Académique Documents
Professionnel Documents
Culture Documents
ii
Contents
1 Syllabus
1.1 Course Packet . . . . . . . . . . . . . . . .
1.2 Course Description and Goal . . . . . . . .
1.3 Use of LaTex . . . . . . . . . . . . . . . .
1.4 Lectures . . . . . . . . . . . . . . . . . . .
1.5 Labs . . . . . . . . . . . . . . . . . . . . .
1.6 Homeworks and In-Class-Exercises (ICEs)
1.7 Reading assignments . . . . . . . . . . . .
1.8 Quizzes . . . . . . . . . . . . . . . . . . .
1.9 Exams . . . . . . . . . . . . . . . . . . . .
1.10 Class Atmosphere . . . . . . . . . . . . . .
1.11 Use of technology in the classroom . . . .
1.12 Communication through email . . . . . . .
1.13 Points allocation . . . . . . . . . . . . . .
1.14 Academic Integrity . . . . . . . . . . . . .
1.15 Students with Disabilities . . . . . . . . .
1.16 FERPA . . . . . . . . . . . . . . . . . . .
1.17 2016 Schedule . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
1
2
2
2
3
3
3
4
4
4
4
5
5
5
5
6
2 Safety
11
3 ABE223 Introduction
15
4 Introduction to hydraulics
4.1 Hydraulic system configuration
4.2 Hydraulic fluid . . . . . . . . .
4.3 Why hydraulics? . . . . . . . .
4.4 A short rant about units . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
23
24
24
25
.
.
.
.
.
33
35
35
35
36
36
iv
CONTENTS
5.3
5.4
5.5
Electric power . . . . . . . . . . . . .
Pump specifications . . . . . . . . . .
Pumps/Motor implementations . . .
5.5.1 Gear Pumps . . . . . . . . . .
5.5.2 Internal Gear Pumps/Motors
5.5.3 Orbiting gerotor . . . . . . . .
5.5.4 Vane Pumps/Motors . . . . .
5.5.5 Piston Pumps/Motors . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6 Cylinders
6.1 Power in equal area cylinders qin = qout . . .
6.2 Power in differential area cylinders qin 6= qout
6.3 Cylinder implementations . . . . . . . . . .
6.3.1 Cylinder cushioning . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7 Valves
7.1 Balanced Piston PRV . . . . . . . . . . . . . . . . . . .
7.1.1 System pressure below preset pressure . . . . .
7.1.2 System pressure equals preset pressure . . . . .
7.1.3 Remote venting principle (control valve closed)
7.1.4 Venting the relief valve (control valve open) . .
7.2 Sequence valve . . . . . . . . . . . . . . . . . . . . . .
7.3 Multipurpose direct acting pressure control valve . . .
7.3.1 Back Pressure Valve . . . . . . . . . . . . . . .
7.3.2 Unloading Valve . . . . . . . . . . . . . . . . .
7.3.3 Sequence Valve . . . . . . . . . . . . . . . . . .
7.3.4 Sequence Valve with integral check valve . . . .
7.3.5 Counter balance valve . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
38
41
43
43
45
45
46
46
.
.
.
.
61
61
63
64
66
.
.
.
.
.
.
.
.
.
.
.
.
81
81
81
81
84
84
84
87
87
87
87
87
92
8 Regenerative circuits
109
8.1 Conventional extension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
8.2 Regenerative extension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
9 Bernoullis law and orifice equation
9.1 Forms of energy in objects . . . . . . . . . . . . . .
9.1.1 Examples: Pumps and motors . . . . . . . .
9.1.2 Examples: Simplified Orifice . . . . . . . . .
9.1.3 Examples: Jet from a tank . . . . . . . . . .
9.1.4 Examples: Heat from friction in a pipe . . .
9.1.5 Examples: Venturi tube . . . . . . . . . . .
9.1.6 Examples: Pitot tube . . . . . . . . . . . . .
9.1.7 Examples: Pressure drop through an orifice
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
119
120
122
122
123
123
124
125
126
CONTENTS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
149
150
156
157
158
12 Ballistics
167
12.1 Case without aerodynamic drag . . . . . . . . . . . . . . . . . . . . . . . . . 172
12.2 Case study: Aerodynamic properties of granular fertilizer particles . . . . . . 174
13 Engine performance
185
13.1 Power and Torque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
13.2 Engine terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
13.3 Engine power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
14 Granular fertilizer application
227
15 Grain harvesting
249
vi
CONTENTS
Chapter 1
Syllabus
Instructor
Credit
Lectures
Labs
Office Hrs
TA
1.1
Course Packet
A course packet is available from the TIS bookstore. This is a hardcopy of a file that can be
downloaded here. Even if you buy the hardcopy, download the file onto a tablet or laptop,
because it contains many links to background materials, videos, etc. All class materials can
be downloaded from the Course Material website
1.2
ABE223 is an introductory course in Agricultural Engineering with special emphasis on OffRoad Equipment Engineering. Off-road machines use Prime Movers (engines) to generate
power; they convert that power using Energy Transfer Systems to do useful work, in Agriculture, Forestry, Construction, and Mining. To address these major topics, this course covers
Hydraulics (4 weeks), Internal Combustion Engines (1 week), and Agricultural Equipment
(3 weeks).
The goal of ABE223 is to familiarize students with basic Fluid Power and Engine power using
a hands-on approach. In addition, the course gives students an overview of some equipment
used in Agriculture such as Fertilizer Applicators and Combine harvesters.
1
CHAPTER 1. SYLLABUS
1.3
Use of LaTex
In this course, we mainly use LaTex for submission of homeworks and labs. LaTex is a free
text editor which has one major advantage over other software. You do not worry about
typesetting anything. It will do it for you, and produce beautiful documents. Another major
advantage is that LaTex is made for equations, either inline or numbered, it will also make
sure that you do not forget closing parentheses etc. In addition, references to figures, and
citations are all automated, and it simply works! LaTex has a learning curve, but once you
used to it you will not want to go back to Word.
LaTex is already installed on the computers in lab 220, but if you want to use your own
laptop you can. To use LaTex on a windows machine, first you need to download and install
MikTEx and then download and install TexStudio. You can also download an excellent user
manual, print pages 200, 201 and hang them above your desk.
To use LaTex on a Mac, download and install MacTex, and then download and install
TexStudio for Mac.
1.4
Lectures
ABE223 is split into four parts: Fluid Power (3 weeks), Air cannon (2 weeks), Engine Power
(1 week), and Implements (2 weeks).
1. The Fluid Power part covers topics such as pressure, flow, pumps, actuators, and circuit
analysis.
2. In the air cannon lectures, you will appreciate the link between adiabatic expansion
theory and ballistic flight theory combined with an actual device. Here you will learn
to write programs to simulate the cannon and flight ballistic and to validate these
models with data from the lab.
3. In the engine lectures, you will learn the basics about internal combustion engines and
you will do an engine dynamometer lab to measure the engine output.
4. In the implement lectures, we will discuss some agricultural equipment such as fertilizer
application and harvesting.
1.5
Labs
1. In the fluid power labs we will build hydraulic circuits that are closely tied in with
the lectures. These labs CAN BE VERY DIRTY. DO NOT WEAR EXPENSIVE
CLOTHING.
2. The air cannon labs consist of a data gathering section where we fire the cannon
outdoors with various projectiles. A follow up lab consists of simulating the motion of
a projectile through air using a ballistic model.
3. The Engine Power and Implements labs consist of an Engine Dynamometer lab where
we collect data from a dynamometer, and analyze these data in the computer lab.
4. Two farm visits are planned, one regarding granular fertilizer application and the other
regarding corn harvesting. These labs take place at the Agricultural Engineering Farm
in October; bring some warm clothes in case it gets chilly.
5. Lab reports are due by the next lab period, submit one report in Compass 2g per group
for the hydraulics labs, all other labs are individual.
1.6
Homeworks will be assigned and submitted through Compass2g. All homeworks and labs are
submitted by downloading a Zip file that contains the LaTex lab document as well as all the
figures it uses. Unzip these files using 7Zip (more reliable than Windows unzipper). Once
you have unzipped all files into one folder, you can double click on the tex file, compile this
file and it should produce your lab handout (if for some reason it complains about finding
files, make sure you unzipped all files in one folder). Then you add to the lab handout, insert
graphs, answer questions, and submit through Compass2g.
In-class-exercises are short ways to emphasize what you learned in that lecture. Typically
they are 15 minutes long, and at the end you will get the answer. These ICEs are handed in,
and you will earn points. If you are absent without excuse you will NOT get these
points.
1.7
Reading assignments
The topics being discussed in class will from time to time be augmented with a reading
assignment. Papers must be evaluated by answering the following questions in your own
words:
1. What did the researchers do?
2. How did they do it?
3. What did they learn?
4. What does it mean to you and society?
1.8
Quizzes
Quizzes will be given to verify the learning of students. QUIZZES ARE NOT ANNOUNCED
IN ADVANCE.
CHAPTER 1. SYLLABUS
1.9
Exams
There will be one final exam during the last class period. The class period before the exam
is a preparation class, no lecture; make sure that you study the material and bring questions.
There is a SAMPLE exam in the course packet.
1.10
Class Atmosphere
The atmosphere in class is very open. Questions are not only encouraged, they are an essential part of the instructor-student interaction. All students are expected to participate in
class.
These are the house rules:
1. No caps
2. No eating
3. Be on time
1.11
Computers and tablets may only be used to enhance your classroom experience; you are
welcome to Google what is being discussed or take notes on an iPAD. However, you may
NOT text, call or check your Facebook status in class, this is distracting you and other
students from following the material.
1.12
When you need to contact me through email, do so in a proper respectful way. Do not send
me one-liner emails that look like a text message; I delete these directly. This is how you
write an email that I will read and respond to:
Good morning Dr. Grift.
I have a question regarding the material you covered in last Mondays class. I would like
to set up an appointment to discuss the material in more depth. Please let me know of a
convenient time for you.
With kind regards,
Student Name.
1.13
Points allocation
1. Final 50%
2. Labs 20%
3. Homeworks & ICEs 20%
4. Quizzes 10%
After the final exam you will be provided with a spreadsheet that you can use to calculate
your final course score. Final letter grades will be amended (B-, A+ etc). The final grade
is not negotiable; do NOT send me emails asking if there is a way to make up for poor
performance, YOU are responsible for keeping up with this class.
1.14
Academic Integrity
1.15
Students with disabilities who have not registered for services with the Division of Rehabilitation Education Services and who wish to request accommodations should contact the
coordinator for Student Services with the Division at 333-4603 (voice or TTY). Reasonable
efforts will be made to accommodate students with disabilities in this course. Inform Dr.
Jin as soon as possible if you require accommodations.
1.16
FERPA
Any student who has suppressed their directory information pursuant to Family Educational
Rights and Privacy Act (FERPA) must self-identify to the instructor to protect the privacy
of their attendance in this course.
1.17
CHAPTER 1. SYLLABUS
2016 Schedule
cannon I
2016 Introduction to safety and air cannon
2016 Lab: Air cannon I (outside AESB, computer lab)
2016 Air cannon theory of adiabatic expansion
2016 Introduction to MatLab
Week 5: Air
September 19,
September 20,
September 21,
September 23,
cannon II
2016 Ballistic model in MatLab
2016 Lab: Air cannon II (computer lab)
2016 Ballistic model in MatLab cont.
2016 Farm of the Future
10,
11,
12,
14,
2016
2016
2016
2016
Combine harvesters
Lab Farm Visit II: Corn harvesting (AGE Farm Race Street)
Exam prep session
Final exam ABE223
The need for a safe environment and the potential dangers that may arise when mistakes are
made are clearly discussed and experienced during the course. There is great emphasis on
following safety protocols in labs.
g. Communicate effectively
Written communication is exercised through reporting of laboratory activities.
h. Broad education to understand engineering in global and societal context
The granular fertilizer application topic is used to emphasize the need for crop yield increases to
feed a global population of 10 billion in 2050.
i. Recognize need and ability in life-long learning
During the course, the students become aware that the technologies in Fluid Power, Engines
(emission standards) and off-road equipment are changing rapidly, and that keeping up with the
latest technologies is imperative.
j. Knowledge of contemporary issues
In the engines lectures, the need for improved efficiency and reduced emissions is explained.
One lecture is spent on the Farm of the Future, an elaborate display regarding the role that
automation and robotics may play in agriculture. Agriculture is the second most unsafe working
environment after mining. ABE223 emphasizes the need to rigorously pursue engineering safety
measures at all times. The topic of safety is dispersed throughout the course materials, as a
constant reminder of its importance.
k. Techniques, skills and engineering tools for engineering practice
The labs are specifically designed to give the students a wide range of skills, such as 1)
constructing, testing and evaluation of Fluid Power circuits (3 weeks). 2) Measuring the
performance of an internal combustion engine (1 week), 3) Evaluation and modeling of a
complete engineered system (Air Cannon) including data collection, programming adiabatic
cannon behavior in Excel, and coupling Excel and MatLab to solve ballistic differential
equations, (2 weeks). 4) Recognition of (fluid) power systems on agricultural equipment,
evaluation of equipment (harvesting, fertilizer application), (2 weeks).
Topics Covered: Fluid Power: Schematics in FluidSim, System components, including,
pumps, valves (Pressure Relief Valve (PRV), Directional Control Valve (DCV), Flow Control
Valve (FCV)), and actuators (cylinders and motors). Principles of Aeration/Cavitation, PRV
efficiency, Pressure Compensated FCV, DCV Center conditions, Metering methods for flow
control, regenerative circuits. Internal Combustion Engines: Basics of SI and CI engines, valve
trains, Power and Torque, traction (Nebraska tractor test data). Air Cannon: Safety
considerations, cannon principle, adiabatic expansion, ballistic projectile model, Excel
programming, MatLab programming to solve differential equations, coupling Excel and
MatLab. Machine Systems: Safety considerations, harvesting equipment performance
evaluation, and granular fertilizer application principles.
A = Considerable achievement
B = Moderate achievement
C = Some achievement
D = Little or no achievement
Chapter 2
Safety
If you are going to read only one chapter of this book, it should be this one. The most
important thing to understand in hydraulics is that it can maim or even kill you! Firstly,
hydraulics has a very high power to weight ratio. Although a cylinder may be small in size,
the force it can generate is very high, more than adequate to cut off limbs or crush bones.
Secondly, small holes in hoses can generate a nearly invisible vapor jet of hydraulic fluid
that can penetrate skin and cause gangrene, an infection caused by the oil contaminating the
wound. Figure 2.1 shows such a jet.
This is very dangerous indeed. To save your limb, they would have to cut out every piece
of tissue that has been in contact with oil. Sometimes the only solution is to amputate a limb
to save your life. At least, you are looking at months of excruciating pain, and rehabilitation.
I will refrain from putting in images of what surgery after fluid injection looks like, but feel
free to Google hydraulic fluid injected into skin and click on Images. It is not pretty.
Therefore, when an hydraulic system is running NEVER feel around hoses or clamps for
leaks. Do NOT assume that because the pump is not running, the system is de-energized.
Cylinders with a load, or accumulators can still have more than enough pressure to hurt you.
A proper system design would include an automatic de-energizing valve, but dont count on
that being the case! If you do accidentally get hydraulic fluid injected under your skin, seek
medical attention IMMEDIATELY. A third potentially lethal hazard is burns: the typical
fluid temperature is higher then 130 degrees F (54 C) which can cause third degree burns.
In summary, it is vitally important that you understand the dangers of hydraulics, visit the
Fluid Power Safety Institute website for more information.
Having said all of this, in our lab we run the benches only at 500 psi where most systems
today run at 5000 psi. You can get as much hydraulic oil on your hands as you want, just
make sure to wash your hands afterward. Also, make sure you are wearing old clothes, there
is a good chance some hydraulic fluid will get on them. It is essential (especially if you want
hydraulics to be part of your career) to be fully aware of the dangers of hydraulics. However,
in ABE223, if you follow our simple house rules, youll be fine.
11
12
CHAPTER 2. SAFETY
Figure 2.1: This fluid jet is visible because the oil pressure is very low, at a realistic pressure
of 5,000 PSI you would not even see it!
&
SAFETY
Hydraulic Systems Safety
by P.D. Ayers
Quick Facts...
Hydraulic systems must store
uid under high pressure.
Three kinds of hazards exist:
burns from the hot, high pressure
spray of uid; bruises, cuts or
abrasions from ailing hydraulic
lines; and injection of uid into
the skin.
Safe hydraulic system
performance requires general
maintenance.
Proper coupling of high and low
pressure hydraulic components
and pressure relief valves are
important safety measures.
Improper Coupling
no. 5.017
Maintenance
Colorado State University, U.S. Department of Agriculture and Colorado counties cooperating.
Cooperative Extension programs are available to all without discrimination. No endorsement of
products mentioned is intended nor is criticism implied of products not mentioned.
Chapter 3
ABE223 Introduction
ABE223 is an introductory course in Agricultural Engineering. Technically, as you know,
our program is called Agricultural & Biological Engineering, but you wont see any biology
in this course. You may look at this course as the Off-Road Equipment Engineering focus
in Agricultural Engineering.
The main topics are Hydraulics, aka Fluid Power, Engines, Air Cannon (Ill explain) and
Agricultural Implements. Do not confuse the term hydraulics with hydrology; that is a whole
different area youll experience in ABE224. In the hydraulics part, it is most important to
be able to distinguish oil pressure from oil flow. Often I hear students say well the pressure
goes from here to there, which is a false statement. I will do my utmost to get students to
understand that oil flows from point a to point b, but that pressure is independent of that
flow. It is the same thing in electric circuits, it takes quite a bit of time before students
understand how to deal with current (flow) and voltage (pressure). We will have three labs
in the hydraulics lab (130) where you will construct various hydraulic arrangements. Note
that these labs are dirty! Hydraulic oil does not come out of clothing easily, so make sure
you wear old clothes. Subsequently, we will study the workings of internal combustion engines (we only scratch the surface, if you are interested in engines, take ABE 466, which is
a semester long course taught by Dr Al Hansen). There will be one lab on engines, where
we test an engine in a test bench (called a dynamometer test).
Then we take a little detour where we use an air cannon as an instructional device. The
reason for doing this is that this device works based on principles you learned in Phys 212,
and also on differential equations. I am sure that many students feel (and rightly so) that
mathematics courses only prepare you for more math; so in ABE223 you get to see that
a combination of physics (adiabatic expansion) and a set of coupled differential equations
can aptly describe a mechanical system. You will also learn how to use MatLab to solve
the (non-linear) differential equations. There are two labs, one where we collect some airtime data, and complete an Excel spreadsheet which contains the equations that describe
the adiabatic expansion process in the cannon, and a second where we crack the differential equations. Next we spend two weeks on agricultural processes, and in the interest of
time, we will study granular fertilizer application, as well as harvesting. Two labs are devoted
to these topics and they will take place at the Agricultural Engineering Farm on Race Street.
Figure 3.1 shows how to get to the Agricultural Engineering farm on South Race Street.
15
16
Attendance is very important in this course. To encourage students to come to class (unless you have a valid excuse), we do a lot of in-class-exercises, where you can score points.
These take about 15 minutes, and afterward you get the answers. You hand in your work
and get points. If you come to class regularly, you will be able to score up to about 30 % of
the total points in ABE223!
Finally, it is my job to teach you skills and knowledge that you can use when you leave
this university. If you have any problems with the material, you need to come see me in office
360J. Preferably, find up to three more student who have difficulties, we will sit in my office
and I will go over whatever you want me to. I guarantee you that spending 30 minutes of time
with me in my office will teach you more than a weeks worth of struggling by yourself. Do
not think that you are the only one struggling. You all made it into the University of Illinois,
which already means that you are smart. Therefore if you have problems with the material, I guarantee you that 50 % of the class does too, they just think theyll read up on it later.
Figure 3.2 shows a distribution of grades in the years 2012-2015. It is clear that you do
not automatically get an A in this course. In fact, about 50 % of students scores and A, 33
% scores a B and the remainder a C or less. Over the years, I have noted that students who
come to class, and visit me in my office are much more likely to score an A than the ones
who dont. So come to class, and come see me if you need help!
17
Figure 3.1: Directions are shown from the Agricultural Engineering Sciences Building to the
Agricultural Engineering Farm.
18
Figure 3.2: Grade distribution in ABE223 from years 2012 through 2015.
8/3/2016
ABE Principles:
Machine
Systems
Main topics
Fluid power
Fundamentals of pressure and flow
Calculation of Pressure, Force/work/Power, flow rate,
piston speed
Implementations
Pumps/Motors/Valves/Actuators
Engines
Working Spark/Combustion Ignition, engine components
Power calculations
Air cannon
Tony Grift, PhD
Dept. of
Agricultural &
Biological
Engineering
Labs
Machinery
Fertilizer application
Combine harvesting
Fluid Power
1.
2.
3.
Cavitation/aeration
Pressure relief valve characterization
Directional control valves
1. Engines
1.
Dynamometer lab
2. Cannon
1.
2.
3. Machines
1.
2.
8/3/2016
dU dq dW dq pdV
Cannon theory:
Isochoric (const. volume)
dU
dW 0
dq
J
cv
mol * K
dU dq ncv dT
dU 0
dW
dq
dq dW pdV
dU
dq 0
dW
dU dW pdV 7
x1 x2
2
2
x2 Kx2 x2 x4
x3 x4
2
2
x4 Kx4 x2 x4 g
40
35
30
25
20
15
10
0
50
60
70
80
90
100
110
A+
12
Chapter 4
Introduction to hydraulics
The sole purpose of hydraulics is to transmit power using a flow of pressurized fluid 1 . Therefore hydraulics is also referred to as a Energy transfer system. The transfer of energy (or
power which is an amount of energy per unit of time) is done by compressing a confined
fluid, and letting it flow under pressure. The basis of all hydraulics is founded on Pascals
principle which says:
Pressure applied to a confined liquid is transmitted undiminished in all directions, and acts
with equal force on equal areas, and at right angles to them.
Figure 4.1 shows a simple hydraulic circuit to demonstrate Pascals law, where we have a
confined fluid shown in red. Lets look at the left hand piston first: a force of 10N is exerted
N
on the piston with an area of 1m2 . This causes a pressure directly underneath of 10
m2
N
where the unit
is also known as the Pascal. Here we have intuitively applied Pascals
m2
law, which says that a pressure is equivalent to a force divided by an area as follows:
N
F [N ]
(4.1)
p
=
m2
A [m2 ]
The principle discovered by Pascal is that in the fluid, the pressure is the same everywhere
(in hydraulics we ignore the fact that there are columns of fluid, we make the assumption
that gravitational effect on the fluid are negligible). This means that the fluid is pushing
against all the walls of its containment with equal pressure. However, some of the walls
are static, they are exerted to a force, but work is only done at walls
that move. In other
N
words work is done only in our pistons. Since the pressure of 10
is present everywhere
m2
(according to Pascals Law), it is also pushing upwards against our piston on the right. Since
its surface area is 10 times larger than the one on the right, the force is ten times larger
too, just apply Pascals law. So for the system to be in equilibrium, we need a 10N force
downward on the left piston and a 100N force downward on the right piston. This force
1
The alternative term Fluid Power captures this better than the term hydraulics.
21
22
Figure 4.1: A hydraulic balance is shown, where on the left side a small piston is causing a
pressure in the fluid, which is counteracted on the right side by a piston with 10 times the
surface area.
increase allows us to move massive loads with low force inputs as is done in hydraulic jacks.
Now lets think about work (see Figure 4.2). Work is equivalent to energy, and the work
we would do if we move the left piston downward say by 1m is 10[N ] 1m = 10[N m]. If
we move the left piston down by 1 meter, the right piston would move up by 0.1m (10 cm)
since the volume we displace on the left is equal to that on the right. We inherently assume
here that the fluid is incompressible, which is not quite true; in reality hydraulic fluid is
compressible by about 3 %, but for our purposes we assume that fluid is incompressible.
Secondly, in this system we assume that there is no resistance to flow, or even if there were,
that the flowrate is so slow that we do not have hydraulic losses. If we did have a resistance
(imagine a very small orifice in the bottom tube) the pressures left and right will equalize
very quickly, but if there is significant flow, we would have a pressure drop across that orifice.
You have to realize that the volume of the fluid cannot change, therefore the flowrate during
the moving of the piston remains the same, and the only parameter that can change is the
pressure.
Finally, we need to discuss power. Power is by definition an amount of work that is done
per unit of time. For instance if the piston on the left side of the circuit shown in Figure
4.2 is pushed downward through a distance of 10 meters in say 2 seconds, then the power
100[N m]
= 50[W att].
generated is
2[s]
Power comes in various forms, and we will derive all of them in later chapters.
23
Figure 4.2: Here the piston on the left is pushed downward at 10N force through a distance
of 1m. This requires an amount of work of W = 10[N ] 10[m] = 100[N m]. If we assume no
losses, that same amount of energy is moving the load on the right upward. Since the force
on the right is 100[N ], the displacement on the right is only 1[m]. This inherently assumes
that the volume of the fluid does not change (we assume incompressibility). If there were a
resistance in the bottom hose, that would result in a reduction in pressure (since the flowrate
cannot change), and this pressure loss would result in a energy loss that would be converted
to heat.
4.1
From an abstract point of view, the hydraulic system converts mechanical energy into mechanical energy, which seems rather pointless. However, what the hydraulic system allows
for is changing the form of the mechanical energy for instance from rotary (from an internal
combustion engine for example) to linear using a cylinder. The hydraulic system also allows
for changing the speed of the actuator (by controlling the flow rate into it) and the force
(by controlling the pressure exerted to it). Therefore, every hydraulic system consists of a
Power supply unit (a pump with some accessories), a Power control section (valves), and an
drive unit (cylinder and motors).
1. Power supply unit. This is where the conversion of mechanical to hydraulic power
takes place. The main components in the power unit are the hydraulic pump (either fixed of variable delivery), the reservoir (tank) which conditions the fluid, and a
Pressure Relief Valve (PRV), which limits the system pressure.
2. Power control section. Here is where the fluid is directed to various actuators,
and the speed and pressure of the fluid are controlled. Valves do this job, such as
Directional Control Valves, which direct the fluid flow into various actuators, Flow
Control Valves (FCVs) which control the speed of the fluid, which in turn controls the
speed of actuators, Pressure Regulators (sometimes called Pressure Reducing Valves,
which causes confusion because its acronym would be the same as that of the Pressure
24
Every hydraulic system has a valve that limits the pressure in the system called the
Pressure Relief Valve (PRV) to a safe level. 2
4.2
Hydraulic fluid
The word hydraulics obviously derives from hydr/hydra/hydro which are the Greek root
words for water. However, water is seldom used in hydraulic applications because it has
a very low viscosity 3 making it hard to seal. The exception to the rule is where water is
used as part of an emulsion in cases of risk of fire, since hydraulic fluid is highly flammable.
Hydraulic fluid (which is what goes into your cars power steering and automatic transmission) is a carefully designed liquid that has good lubrication properties, a relatively constant
viscosity over a large temperature range, and resistance to chemical degradation over time.
The viscosity must be high enough so that it does not leak past seals, and low enough so that
internal friction does not cause too much heat to build up, which reduces the overall power
transmission efficiency. It is also the medium that cools the hydraulic system, it carries away
abrasions to a filter, and at the same time lubricates all components. It also provides a film
of oil on metal surfaces, guarding them against oxydation.
There are five main functions of the fluid being (make sure you know these, this is often
an exam question):
1. Transmission of power
2. Lubrication
3. Cooling
4. Corrosion protection
5. Carry away abrasions to the filter
4.3
Why hydraulics?
One of the main advantages of hydraulic is that it can produce huge amounts of power from
relatively small actuators, in other words, the power to weight ratio of hydraulics is very
2
This does NOT mean that the pressure can never exceed the PRV pressure, we will discuss this in the
cylinder section.
3
Remember that viscosity is the resistance to flow, therefore honey has a high and water a low viscosity
25
high. Secondly producing linear motion is easy with cylinders, which is hard to do, not to
mention expensive, with alternatives such as electric or mechanical power transfer systems.
Thirdly, hydraulic systems can be started under full load, for instance when a lift is used to
elevate a large load such as a truck. Finally, overload protection is easily implemented, in
the form of a pressure relief valve (PRV). As you will soon learn, there is a lot of terminology
and many acronyms in hydraulics. Make sure you make a list of terms and abbreviations
and use them in communication. Finally, hydraulic power can be positioned easily at any
location, since it merely requires properly routed hoses. Compare this to mechanical systems
which require shafts, couplers, gearboxes and transaxles to do so.
Hydraulics also has disadvantages. The efficiency of hydraulic power transmission is very
low, since quite a bit of the input energy is converted into heat. Mechanical and electric
power transfer systems are much better in that respect. Also, hydraulic systems leak slightly,
which is mainly an emotional disturbance. The question to ask is whether you would rather
have some oil flow into the ocean, of some seawater penetrate an hydraulic motor? It is
something we live with.
4.4
It is repugnant that the most powerful country in the world still relies on an archaic, illogical,
and in my opinion, diabolical system of units. No other country in the world clings to a
system that has 16 ounces in a pound, and 7,000 grains in an ounce which nobody (apart
from some gun buffs) uses. Students like yourselves have spent years learning about unit
conversions, losing points on exams because you ran out of time. Just to demonstrate this,
here are the Hydraulics conversion factors provided by the Hydraulic Supermarket company.
The biggest problem with the English unit system is that it has no standards, which makes
it useless from a scientific point of view. I wrote an extended rant about the insanity of the
English units system titled Metric Eye for the AgE guy (and gal), feel free to read it.
Having said all this, we still have to deal with the fact that our hydraulic trainers have
pressure gages with readouts in PSI (they also read in kPa, but that scale is much smaller
and harder to read). They have a label that says Maximum PRV pressure is 500 PSI which
is conveniently placed at the 12 oclock position on the gage. The scale of the cylinder and
their specs are also in English units. The flow meters read in Gallons Per Minute (GPM)
and liters per minute (LPM).
m3
N
, where former
Ideally pressure gages should read in Pascal or 2 , and flow meters in
m
s
number would be very high and the latter very low, so in reality they read in kPa and in LPM.
What this all means is that in lab reports you will still have to do some simple conversions, but at least during exams and homeworks you will get all data in pure SI such that
you do not have to convert any number.
Agenda
Energy Transfer Systems
Pascals Law
Basic Hydraulic Systems
ABE 223
ABE Principles Machine
systems
Pump
Actuators
Pressure Relief Valve (PRV)
Applications
Agriculture
Construction/mining
Forestry
Introduction
to Hydraulics
Tony Grift
Dept. of Agricultural & Biological Engineering
University of Illinois
Introduction
Power steering
Hydraulic brakes
Thrill rides
Basic fluid power
Jack
History
Distance in m
Area in m2
Volume in m3
Time in seconds (s)
Flow rate in m3/s
Hydraulics can lift large loads in linear fashion (bucket of dump truck)
Conversion into useable form and transmission to work location
Compressed fluid is the energy carrier
Disadvantages
Leakages
Sensitive to contamination
Can be dangerous
Stationary systems
Mobile systems
Construction equipment
Vehicles (Off-road and On-road), ships
Aircraft
5
3400 HP
360 Tons capacity
40 MPH Fully loaded
Mechanical drive
vo, Fo
ni, Ti
P1
P4
Q1
Q4
Energy Level
no, To
Basic Principles
10
Liquids form in the shape of the container (like the lines for
transmitting fluid power)
Liquids can be considered incompressible
Fluid properties
11
12
Blaise Pascal
13
100 N
10 N
1 m2
14
10 m2
10 N/m2
15
Conservation of Energy
16
10 m2
m2
100 N
Fi Li Fo Lo
1m
10 m
10 N
10 N/m2
Energy can be neither created nor destroyed
17
18
Linear Actuators
F p1 A1 p2 A2
A1
A2
F
p1
p2
19
Linear Actuators
20
Rotary Actuators
Cylinders
Motors
Double acting
Differential area
Double acting
Dual Rod
21
22
Variable Speed
Reversible
Overload Protection
Actuators can be driven from a high to low speed using a flow control valve.
Hydraulic actuators can be instantly reversed in direction of motion.
Can Be Stalled
23
24
25
26
Mining Equipment
27
28
Backhoe arm
Cylinders
Motors
29
30
31
32
Forestry Equipment
33
34
35
36
37
38
Components of ABE223
Hydraulics (wk 1-3)
Engines (wk 6)
40
ABE 223
ABE Principles Machine
systems
Introduction
to Hydraulics
The End
Dept. of Agricultural & Biological Engineering
University of Illinois
41
Chapter 5
Pumps and motors
Lets begin at the beginning; a pump is a device that converts mechanical rotary power into
hydraulic power. Hydraulic motors closely resemble hydraulic pumps, in fact if you would
hold one in your hand you could probably not tell the difference. The function of a motor
versus a pump is opposite: A pump converts mechanical rotary power into hydraulic power
and a motor does the exact opposite. Therefore in the following discussion about pumps you
may replace the word pump for motor, unless stated otherwise.
Every pump has a suction side and a pressure side and some form of seal between them.
This seal allows the pump to build up pressure. In hydraulic circuits, motors are always of
the positive displacement type, meaning that there is no flow other than leakage possible
from the inlet to the outlet, in contrast to a centrifugal turbine pump. The latter does not
have a seal but relies on momentum transfer from an impeller to the fluid. Pumps typically
are drained internally since they have a suction side that is directly connected to the tank.
Motors are drained externally, since they can have pressures on the inlet and outlet side.
Figure 5.1 illustrates a basic hydraulic pump. The suction side is at the bottom, and
the pressure side at the top. The seal is formed by the teeth of the gears meshing together.
Also, the sides of the gears must be sealed to prevent high pressure oil to leak back to the
low pressure suction side.
At the suction side of any pump, the volume increases (this is what draws the fluid into
the pump) and at the pressure side it decreases which pushes the fluid out of the pump 1 .
Note that in this gear pump, the oil is drawn in at the suction side where the unmeshing
teeth increase the volume. Then the fluid is transported along the perimeter of the pump
housing to the pressure side, where the meshing of the teeth decrease the volume and oil is
pushed out. Gear pumps have only one gear that is driven the other is idling and in fact
driven by the driven gear itself.
1
A common misconception is that a pump is a pressure generator. A pump does nothing more than
creating a flow of fluid which can be constant (constant delivery pump) or variable (variable delivery pump).
Only once the fluid meets resistance, either through pipes and orifices or to move a load will pressure build
up.
33
34
Figure 5.1: Shown is a gear pump, the cheapest, simplest pump in hydraulics; It is a constant
delivery pump that has limited flow output and pressure range and it is very noisy.
At the inlet side of the pump we have a partial vacuum, unless the pump is placed underneath the oil level in the reservoir (sometimes pumps are in fact placed inside the tank,
such as is the case in most cars. This is why oil pumps seldom fail, they are always well
lubricated). Although we can build up very high pressures in hydraulic fluid, on the suction
side we are limited to ambient pressure, which limits the height from which fluid can be
drawn into the pump. Since there is a partial vacuum in the suction line, it is possible that
air enters the oil through a leak (at any other place in the system a leak causes fluid to exit
the system). When air is entrained into the oil, the bubbles pass through the pump, where
they collapse violently at the pressure side. Technically, air entrainment into the system is
called aeration, and it causes cavitation, which over time can erode the pumps parts to the
point of failure. The only way to detect cavitation is to listen to the pump, you will not be
able to tell from the system pressure. The sound of cavitation is highly recognizable though,
you can compare it to the sound of marbles bouncing around in a tin can or a high pitched
scream with intermittent popping sounds. There is another potential cause of cavitation,
which is when an obstruction in the suction line causes a pressure drop caused by the fluid
attaining a very high speed locally (remember that in constant displacement pumps, the flow
rate is constant, and if the line is partially blocked the same flowrate has to go through a
much smaller orifice). The pressure drop (this is according to Bernoullis Law, which we will
discuss later in this course) causes the fluid to boil, and vapor bubbles will emerge. These
bubbles have the same effect as air, they cause cavitation of the pump and can also cause it
to fail. However, the idea that the suction line would be blocked is rather contrived, it would
take a service person to drop a cloth in the tank, which is highly unlikely; nevertheless about
70 % of all hydraulic failures are due to maintenance errors. From the sound you would not
be able to tell whether cavitation is caused by air or vapor bubbles, but your best bet would
be air entrainment.
35
You probably wonder if cavitation happens elsewhere in the system and the answer is all
the time. For instance, when a valve just opens, the fluid speed through this small orifice
is very high and vapor bubbles will present themselves. However, bubbles in the system are
not a problem, because eventually, they will go through the tank, where the fluid is given
enough time to let them float to the surface and disappear. In summary, bubbles either from
air or the fluid itself are only a problem if they go through the pump.
5.1
To drive the pump, mechanical rotary power is generated in either an internal combustion
engine (often used in mobile hydraulics) or an electric motor (most often used in stationary
applications). Then a pump is used to convert the mechanical energy into hydraulic energy.
5.1.1
The power flowing into an engine is often referred to as fuel power. This input power is
equal to the mass flow rate of the fuel in kilogram per second [kgs1 ] times the heating value
of the fuel in Joule per kilogram [Jkg 1 ]. Check that when you multiply these units, you
get the fuel power in Joule per second P owf uel [Js1 ] , which is equivalent to the Watt 2 .
Power generated at the output shaft of an internal combustion engine is essentially mechanical rotary power, symbol P owmech,rot . The output power of the engine (often still
referred as brake power, although it is an archaic term) is calculated from the torque of
the output shaft T in Newton-meter times the rotational velocity of the shaft in radian
per second, therefore
P owmech,rot = T N ms1
(5.1)
The ratio of the output (brake) power and the input (fuel) power is the total efficiency
P owmech,rot
, which is a unitless parameter. The total efficiency of
of the engine or engine =
P owf uel
internal combustion engines ranges from 33 % for Spark Ignition (SI) engines to about 43 %
for Compression Ignition (CI) aka diesel engines. In engines, the total efficiency comprises
the thermal efficiency (limited by the compression ratio) and the mechanical efficiency (due
to friction).
5.1.2
Power generated in an electric motor is equal to the voltage drop across the motor in Volt
[V ] times the current flowing through it in Ampere [A]. It is not directly clear that this
product leads to power in Watt, but you have to realize that the Volt is in fact energy in
Joule per a unit of charge called the Coulomb or V = [JC 1 ], and secondly that the current
2
Note that in this course we use P ow for power, so as to not get confused with the lowercase p which we
use for pressure.
36
is in fact the unit of charge in Coulomb per unit of time in second or A = [Cs1 ]. Therefore
the formula for electrical rotary power is:
P owel = V A Js1
(5.2)
The mechanical input power needs to be converted into hydraulic power. This is done in
pumps.
5.1.3
Hydraulic power
Although we will formally derive hydraulic power later, just by analogy with the electric
power, we can state that the voltage in the electrical domain is equivalent to pressure in a
hydraulic system, and that the current in the electrical domain is equivalent to the flow rate
in hydraulic systems. The unit of pressure p in hydraulic systems is the Newton per square
meter (this is also known as the Pascal [P a]), and the unit of the flow rate q is cubic meter
per second. Multiplying these we get a formula for hydraulic power being:
P owhyd = p q N ms1
(5.3)
and since the Newton-meter is equivalent to the Joule we again get Watt. There is one
caviat here; This formula only applies when, as in the electric motor, the flow rate into
the hydraulic device is equal to the flow rate out of the hydraulic device, which is the case
in pumps, motors and equal area cylinders. It does NOT apply in the case of the (most
common) differential area cylinder, where the flow rate in and out are different.
As in the engine, the total efficiency is the output (hydraulic) power divided by the input
P owhyd
. As in the engine, the total efficiency of
(mechanical) power or: pump,motor =
P owmech,rot
the pump/motor comprises a mechanical part (due to friction) and a volumetric part (due
to leakage). The total efficiency of the pump/motor depends on the rotational velocity, but
overall is quite low, typically around 75 % or so.
As was discussed in chapter 4, the purpose of a hydraulic system is to transfer power.
Input power to the system is always mechanical, rotary power from either an internal combustion engine or an electric motor. Output power comes in a linear form (as produced by
cylinders) and a rotary form (as produced by a hydraulic motor.) To design a hydraulic
system, we need equations that indicated how much power we can generate. Lets start with
the input side, the hydraulic pump.
5.2
From an abstract point of view, a pump does nothing more than converting mechanical power
into hydraulic power, at a certain efficiency. The input is a shaft that turns at a predefined
RPM and it carries a rotational force which we call torque. To find out how much power this
shaft carries, the easiest route is to go through the sequence of Force F [N ], Work (which is
Force through a distance W [N m]) to Power which is work per unit of time P ow in [N ms1 ].
37
Figure 5.2: Shown is the cross sectional area of shaft that carries power in mechanical form.
The force F acts at a distance R. Here we choose to let the force act along a distance R,
this simplifies the derivation of mechanical rotary power.
Figure 5.2 shows a force acting at a distance R[m] from the center. To calculate how
much work that force would do, the easy thing is to move it through that same distance R.
This is because the angle that the force would rotate through is then by definition a radian.
The work that would be done would then be W = F R[N m]. Now we need to find out
how much Power we would produce if the shaft would be turning at a certain speed. It is
easiest to express this speed in [rads1 ]. For Power we need to divide the Work by the time
it would take to move the force through the distance R. If the shaft is turning at [rads1 ],
1
then the time it takes to rotate through ONE radian is [s] (Note that the radian is an
angle and has no unit). Since we have to divide the Work through this time we get:
FR
P owrot = = F R[N ms1 = W att]
1
(5.4)
The factor F R is called the Torque T in N m, so the main equation well use is:
P owmech,rot = T N ms1
(5.5)
This discussion has been about a pump, but the exact derivation works for a hydraulic
motor, just the other way around. Now we need to find an equation for hydraulic power.
The amount of hydraulic power that a pump can generate depends on the flow rate and the
pressure it can achieve. If there are no losses, the mechanical input power must be equal to
the hydraulic power produced, therefore the theoretical power (without any losses) needed
to drive the pump is:
38
P ow = T = pq
(5.6)
Where p is the pressure drop across the pump in P a and q is the flow rate from the
m3
.
pump in
s
In reality, the situation is more complex. Firstly, there is mechanical friction in the pump,
which means that the torque applied in reality is higher than the theoretical torque. The
torque efficiency is a number that quantifies this loss 0 < T < 1. Secondly, the produced
flow in reality is lower than the theoretical flow because of leakage. This loss is expressed in
the volumetric efficiency 0 < V < 1.
The theoretical pump delivery is proportional to its rotational velocity and a constant
displacement per radian as follows: q = D , where D is the displacement in m3 rad1 , and
the pumps rotational velocity in rads1 . In reality the pump delivery is lower because of
leakage as follows:
q = D V
(5.7)
p D
T
(5.8)
p D
T
(5.9)
Now we substitute the rotational velocity from equation 5.7 into equation 5.9 and we
obtain:
P ow = T =
p D
q
p q
=
T
D V
T V
(5.10)
Note that both the volumetric as well as the torque efficiency end up in the denominator
of the equation for the realistic power. Realistic power is theoretical power divided by both
efficiencies.
5.3
Electric power
The electric power generated by a motor has a very similar equation to the hydraulic power
equation for qin = qout as follows:
P owEL = U i[V olt Ampere = W att]
(5.11)
39
The SI unit for power is the Watt, which is equal to a Joule per second, where the Joule
is equal to a Newton meter [N m]). It is not obvious why multiplying a Volt times a current
would give a Watt. Remember that hydraulic power was equal to a pressure differential
p[N m2 ] times a flow rate in q[m3 s1 ] . Multiplying these units obviously gives the Watt.
By intuition it is clear that the pressure differential in hydraulics must be equivalent to
the voltage differential in electric circuits, and that the flow rate must be equivalent to the
current. The current in electric systems is indeed, a unit of electrical charge per unit of
time, where the unit of electrical charge is the Coulomb, hence the current (in Ampere) is a
Coulomb per second. Since the charge of one electron is 1.602176565(35) 1019 Coulomb
(where the (35) is the uncertainty in the last two digits), one Coulomb can also be considered to be the charge of roughly 6.241509 1018 electrons. The earliest determination of the
charge of an electron was through Millikans oil drop experiment.
If the unit of current is Coulomb per second, and the multiplication of the current and
the volt must result in Joule per second (Watt), it is clear that a Volt is equal to a Joule
per Coulomb (see Equation 5.11). This is the way to think about the Volt: it is a measure
of the energy contained in a fixed number of electrons (1 Coulomb).
J
J C
= = W att
(5.12)
C s
s
Figure 5.3 shows a table that has all power forms summarized. Note that all units
become N ms1 or Watt. The Joule J is a unit of work (energy), which is equal to a N m.
As is clear, the unit of Torque is also N m, but that is a coincidence, you cannot conclude
that torque is equal to work or energy.
40
Figure 5.3: Shown are power forms in various domains. Check that the units are always in
Nm
or W att.
s
Here is a table where all power conversion devices are listed, with their input and output
power equations. The units of all P ow symbols is of course W att.
Device
Input Power
Equation
Int. Combustion Engine Fuel power
P ow = m
HV
Electric motor
Electric
P ow = V i
Hydraulic pump
Mech, rot
P ow = T
Hydraulic motor
Hydraulic
P ow = p q
Equal area cylinder
Hydraulic
P ow = p q
Diff. area cylinder
Hydraulic
P ow = pc qc pr qr
where:
m
is fuel mass flow rate in
Output Power
Equation
Mech, rot
P ow = T
Mech, rot
P ow = T
Hydraulic
P ow = p q
Mech, rot
P ow = T
Mech, lin
P ow = F v
Mech, lin
P ow = F v
kg
s
J
kg
rad
s
41
N
or P a
m2
m3
s
m
s
m3
s
5.4
N
or P a
m2
N
or P a
m2
m3
s
Pump specifications
When designing a hydraulic system the selection of a suitable pump is essential, since it
drives the complete system. Firstly the choice has to be made whether a constant or variable delivery pump is suited. In modern equipment, constant delivery pumps are seldom
used since they typically waste a lot of energy when the system is at idle, meaning that all
actuators have stalled. Modern system are load sensing, meaning that they only deliver flow
if it is needed by an actuator. This system gives great energy savings.
The next design criterion is the hydraulic power output needed. This criterion depends
on the pressure-flow characteristic of the pump. Other considerations are efficiency (gear
pumps are much less efficient than piston pumps), projected lifespan, and obviously costs.
The same design criteria are applied to hydraulic motors, but here the other way around.
Here the design is based on the required mechanical rotary output power, which is a function
of torque and rotations shaft velocity (RPM).
Figure 5.4 shows the specs of an Eaton MHT750 vane motor (click here for the complete
vane motor catalog). From the performance chart, it is clear that the volumetric efficiency
increases as a function of the rotational velocity (pumps and motors typically leak less at
higher speeds), and the torque efficiency is nearly constant. Hydraulics in general is inefficient compared to mechanical or electrical energy transfer systems, but it is highly flexible,
it has a huge power to weight ratio, and linear motion is easily accomplished. These are
advantages that no other system can match.
42
5.5
43
Pumps/Motor implementations
Figure 5.5 shows the family tree of pumps (and in most cases, motors). The first division
is between positive displacement pumps and non-positive displacements pumps. The latter
are centrifugal pumps that use an impeller to transfer momentum to a fluid, and are used
often in hydrology, but never in hydraulics (fluid power) because they cannot build on any
significant pressure. The three main categories under the hydraulic positive displacement
pumps are the gear pump, the vane pump and the piston pumps. Gear pumps are always
fixed delivery pumps (since variable delivery would require the shaft to turn at a variable
speed, which is possible, but never done in practice), but vane pumps and piston pumps are
available in either fixed or variable delivery form. Gear pumps also include the internal gear
pump and the orbiting gerotor. The latter is a very complex (and quite elegant) piece of
machinery, but it needs extensive porting to ensure proper flow direction, which makes it
more expensive than the simple two-gear or the internal gear pump.
Overall you have to remember that:
1. Gear pumps are only applicable as auxiliary pumps or in trainers (such as we have in
this class). They are noisy, have low delivery and a limited pressure range.
2. Vane pumps are the middle ground, they are more expensive than gear pumps, more
complicated, have a shorter lifespan, but they can deliver high torque. Most car power
steering pumps are vane pumps.
3. The best (and also most expensive) pumps in fluid power are piston pumps. The
axial version is very simple in design, it has no valves, it seals very well (giving it a
high pressure range) and variable delivery is easily implemented using a variable angle
swash plate. The radial version is more complicated, it needs valves to ensure proper
flow direction, it is more bulky and implementing variable delivery is more complicated
than in the axial case.
5.5.1
Gear Pumps
Figure 5.6 shows a cutaway view of the gear pump. Note that the fluid takes the long way
around the gears, do not think of it as a vane type inertia device. Also, note the simplicity
of design, essentially one gear drives another and the meshing/unmeshing of the teeth comprises the pump action. No valves or springs are needed, and the pump has a distinct suction
side and a pressure side. The drawback of the design is that the pump has a low flowrate
output, limited pressure range (due to limited sealing) and it is very noisy. Therefore if
off-road equipment applications, these pumps are only used as auxiliary devices, never as
the main pump.
44
Figure 5.5: Family tree of pumps (most of them are also in motor form). In terms of
technical sophistication and performance the worst pumps are on the left and the best on
the right. As expected, in terms of costs it is the exact opposite.
Figure 5.6: Cutaway view of a gear pump. Although every hydraulic circuit contains a
Pressure Relief Valve, this pump has a safety valve built in (see if you can discover it in the
picture).
45
Figure 5.7: Internal gear pumps are similar to the two-gear pump we saw earlier in Figure
5.1 in an inside-out kind of way. One gear is driven and it drives the other. The outer
gear typically has one more tooth than the inner gear, so each revolution, the volume of one
tooth-gap combination is displaced. A crescent shaped seal is used to separate the suction
side from the pressure side of the pump.
5.5.2
Figure 5.7 shows an internal gear pump/motor. These devices are built using a matched
inner/outer gear set, where the inner gear has one fewer tooth than the outer. In this
pump/motor both gears have to turn, if the shaft is turning CCW, the left half of the pump
has an increasing volume (suction side) and the right side a decreasing volume (pressure
side). Every pump in hydraulics has a seal between the suction and pressure side, in this
case it is implemented by a metal crescent shaped seal, which is not a positive seal. The main
advantage of the design is simplicity (which implies low maintenance and high durability),
there are no valves needed; the left chamber is simply connected to the suction port and the
right to the pressure port.
5.5.3
Orbiting gerotor
Figure 5.8 shows the principle of the orbiting gerotor motor. Its design is similar to that of
the internal gear pump, with one big difference: the outer gear is stationary and the inner
gear revolves around a shaft that itself orbits. The advantage is that the pump does not
need a seal between the pressure and suction side, but it requires quite complex porting
(similar to a rectifier circuit in electronics) to ensure proper flow direction. An animation of
the orbiting gerotor pump can be found here.
46
Figure 5.8: Compared to the internal gear motor, the orbiting gerotor motor has the advantage that no seal is needed between the suction and pressure sides, since the outer gear is
stationary. This however requires a complex shaft design: since the inner gear orbits inside
the outer gear, the shaft that drives it must orbit as well, requiring a complex coupling. In
addition, the motor does not have a distinct suction and pressure side, so complex porting is
needed to ensure the correct flow direction (this is similar to a rectifier circuit in electronics,
as implemented in a Greatz circuit).
5.5.4
Vane Pumps/Motors
Figure 5.9 shows a vane pump/motor. The image shows how the vanes are placed offset
from the center of the shaft, which creates, as always, a region where the volume increases
(suction side) and a region where the volume decreases (pressure side). The vanes are pushed
against the circumference of the pump chamber, which does not result in a very good seal,
limiting the pressure range of the pump. To improve sealing, the volume underneath the
vanes themselves is typically connected to the pressure side of the pump, so the higher the
pressure, the better they seal. This is an example of a positive seal.
The photo shows an adjustment screw on the left which allows for altering the amount of
offset with respect to the shaft, which alters the displacement and therefore the flow delivery.
However, in this particular pump, it is not meant to be done in real-time therefore this is
not a variable rate pump.
5.5.5
Piston Pumps/Motors
Piston pumps and motors use reciprocating pistons in cylinders to produce flow. The sealing
of pistons inside cylinders can be done cheaply and efficiently, giving piston motors a high
pressure rating. All piston motors can be fixed or variable displacement types. Piston pumps
and motors are the most efficient and provide high pressure and motor speeds. Also the ratio
power to weight is the best.
47
Figure 5.9: Left: Vane pump with displacement adjustment screw, which is not meant as
an instantaneous adjustment; this is not a variable delivery pump. Right: Application of a
vane pump in a vehicle power steering system.
Axial Piston Pumps/Motors
Figure 5.10 shows a photo of an axial piston pump. In an axial piston pump, the pistons
move in the direction of the main shaft. The pistons are mounted in a barrel and connected
to a variable angle swashplate through shoes, which makes this a variable displacement
pump. The complete barrel is rotating inside this pump: in one half of the pump the pistons
are retracting (this is obviously the suction side) and in the other the pistons are extending
which constitutes the pressure side. The pistons can be made to seal very well through the
use of piston rings. The pump does not need valves or ports, it is very simple in design, it
can have a high delivery and high pressure range. Although the design goes back to World
War II, it is the most elegant and effective pump out there. For an animation of an axial
piston motor click here.
48
Figure 5.10: The most elegant pump design in hydraulics is the axial piston pump as shown
here. The pistons are mounted in a barrel that rotates. Since the pistons are connected to
a swash plate which has an angle with respect to the pump axis, they move back and forth
inside the barrel. This means that at one side of the pump they are retracting (suction side)
and at the other they are extending (pressure side). The design does not require valves or
complex porting, and variable rate is easily implemented by changing the swash plate angle,
(which is typically itself actuated by a built-in hydraulic cylinder). This is a marvelous piece
of engineering!
Figure 5.11: The radial piston motor has all the advantages of the piston design (easy sealing
for example), but the piston arrangement makes variable rate far more difficult to implement
compared to the swash plate design, and this type of pump requires valves to ensure proper
flow direction.
49
Figure 5.12: The bent-axis pump/motor is easily recognized by its distinct bend in the
housing. It is a simpler version of the axial piston design, it has all the advantages of the
use of pistons, and it does not require valves. Overall it is a high-power, high flow, reliable
and simple design.
is fixed here. These kinds of motors are easy to recognize, in fact they look rather awkward
with the distinct bend in their housing. For an animation of the bent-axis motor click here.
Agenda
ABE 223
ABE Principles Machine systems
Pascal
Flow
Newton
Watt
Area
Force
Work
Power
N
1 Bar 10 2 105 Pascal
m
5
Torque
N
p 2
m
m3
q
s
Pressure control
Pressure relief valve
Pressure reducing valve (regulator)
Pressure compensation: Only provide the pressure needed to move the
load(s). In idle reduce energy loss by providing a open center condition
Pressure compensated pump. Make the pressure independent of the
flow required to move the load at a preset speed. Deal with multiple
cylinders that need to move simultaneously
A m2
F N
W Nm
Flow control
Nm
Pow
s
TNm
m3
Displacement D
rad
Application
Price
Components
WMECH F * r
Drive motor
Safety valve
Oil reservoir
Pump
1 Radian
Pressure Relief Valve diverts flow back to tank when cylinders are stalled
Reservoir
Cools oil
Cleans oil from suspended particles, water and air which takes time (Capacity)
Filters trap impurities
70% of all malfunctions are due to impurities
7
qin qout )
The power is now equal to this value divided by the time per
radian.
rad
1
If the shaft is turning at
seconds per radian.
it takes
s
Since for Power we have to divide the Work by time, this leads
to:
W F *r
PowMECH
*F
*r
t
1
T
PowHYD p * q
PowMECH * T
Nm rad
s s * Nm
10
PowHYD p * q
PowHYD p * D *
q D *
PowHYD p * D *
D
T p *
PowMECH T *
Constant
q D *
More realistic, with Volumetric Efficiency (why in numerator?)
q D * *V ,0 V 1
T
11
p * D
,0 T 1
12
p * D
q D * *V
*V
And
PowMECH * T
Power Efficiency:
PowMECH
Without loss
p * D q * p
q
*T
*
T
V *T
D *V
Efficiencies
13
14
External Gear
Internal Gear
Vane
Axial Piston
Radial Piston
Poorly sealing
Low flow rates
Low pressures
Fixed displacement
Low cost
15
16
Vane pump
Poorly sealing
Low flow rates
Medium pressures
Fixed displacement
Medium cost
17
Medium sealing
Higher pressures
Inexpensive
Fixed or variable flow
rate
18
Excellent sealing
High pressures
Relatively simple design
Variable flow rate
No need for valves
Expensive
19
20
Excellent sealing
High pressures
Relatively expensive
Variable flow rate
Valves are needed
Complex design
Excellent sealing
High pressures
Relatively expensive
Variable flow rate
21
External Gear
22
Fixed Displacement
23
24
ABE 223
ABE Principles Machine systems
Pumps and Motors
The End
Tony Grift
25
Figure 1 shows the performance chart of an Eaton MHT750 Vane Motor. Calculate the following
quantities using the graph, try to get as accurate as possible, you may have to interpolate.
You need to submit this using LaTex, take this file and extend it. This is an INDIVIDUAL homework.
IDLE
EXTENSION
RETRACTION
Chapter 6
Cylinders
One of the main advantages of hydraulics is that it is quite easy to implement linear power.
This is much harder and more expensive in the mechanical or electrical domain. Cylinders
consist mainly of a piston with a rod attached that moves in a cylinder. Seals are in place
to prevent hydraulic oil to pass from the cap end from escaping the cylinder.
6.1
Now we have an equation for the mechanical rotary power, we need an equation for the
power that the hydraulic fluid contains when compressed. This section explains the power
generated in equal area cylinders. We assume that the flow into these devices is the same
as the flow exiting them. We will evaluate a cylinder with a piston, to which a pressure is
applied on both sides as shown in figure 6.2. The areas on both sides of the piston are equal
being A[m2 ]. The flow rate into the cylinder that moves the piston is q[m3 s1 ], and since
the areas are identical, the flow entering the cylinder is equal to the flow exiting the cylinder
or qin = qout = q. What we have inherently assumed here, is that the fluid is incompressible,
which is not quite true, but its a good assumption for our purposes.
According to the definition of pressure, a force exists when a piston with an area A is
exposed to a pressurized fluid with pressure p: Since the pressure on the left side of the
piston is not equal to the pressure on the right side, the net force is the difference between
the two forces (assuming a static force equilibrium)
F 1 = p1 A
F 2 = p2 A
F = F1 F2 = A (p1 p2 )
(6.1)
(6.2)
(6.3)
where the forces F have the unit [N ], the area A is in [m2 ], and the pressures p are
N
in 2 [P a], also known as the Pascal after French mathematician / physicist Blaise Pascal.
m
Work is done when this force is applied through the distance s through which the piston
moves:
61
62
CHAPTER 6. CYLINDERS
Figure 6.1: Cylinder terminology: The way to learn the terminology of hydraulic components
in general is to remember what each part does.
63
Figure 6.2: Shown is an equal area cylinder. The pressures on the left side p1 and right side
p2 are not necessarily equal. However the flow q into the left side IS equal to the flow q on
the right side. In diagrams like these, unless mentioned otherwise, we assume that the fluid
is incompressible.
W = F s = A (p1 p2 ) s[N m]
(6.4)
A (p1 p2 ) s
Fs
=
t
t
P owHY D = (p1 p2 )
As
t
(6.5)
(6.6)
The first part p = (p1 p2 ) [N m2 ] is the pressure differential across the piston and
the second part is the volume of fluid that is displaced per unit of time, which is the flowrate
q[m3 s1 ]. Therefore we can write:
P owHY D = p q[N ms1 = W att]
(6.7)
It is VERY important to write the in this equation, since it reminds you that you need
the difference in pressure, you need TWO numbers not one. You need only one number for
the flowrate, since we have assumed here that the flow into the device is the same as the
flow exiting it.
6.2
It may seem rather odd to devote a complete section to the situation where the flow rate
entering the actuator is not equal to the flow rate exiting it, but this happens to be the case
in the most common actuator, the differential area cylinder. Figure 6.3 shows this cylinder.
We are going to extend the cylinder, meaning that the piston (and rod) move from left to
right.
64
CHAPTER 6. CYLINDERS
Figure 6.3: Shown is a differential area cylinder. As in the equal area cylinder, the pressures
on the left side pC and right side pR are not necessarily equal, but in this case, the flow
entering the cylinder qC is not equal to the flow exiting qR it either. The left side is called
the cap end and the right side is called the rod end since this is where the rod protrudes
from the cylinder.
Fc = pc Ac
Fr = pr Ar
F = Fc Fr = pc Ac pr Ar
(6.8)
(6.9)
(6.10)
W = F s = (pc Ac pr Ar ) s[N m]
(6.11)
Fs
(pc Ac pr Ar )s
=
t
t
(6.12)
where the forces F have the unit [N ], the area A is in [m2 ], and the pressures p are
N
in 2 [P a], also known as the Pascal after French mathematician / physicist Blaise Pascal.
m
Work is done when this force is applied through the distance s through which the piston
moves:
Ac s
Ar s
pr
= pc qc pr qr
(6.13)
t
t
In other words, the total power we can generate is the power generated at the cap end
pc qc minus the power consumed at the rod end pr qr . For maximum power, of course ideally
the pressure on the rod end side should be zero.
P owHY D = pc
6.3
Cylinder implementations
Cylinders and motors are the main actuation mechanisms of hydraulics. One of the main
advantages of hydraulics is that it enables easy linear motion with a very high power to
65
Figure 6.4: Smaller cylinders are often made with tie rods keeping the cap end and the rod
end together. The complete cylinder consists of a cap end head, a rod end head, a barrel
which is compressed between the heads with o-rings or cork seals, a piston with a seal and
a piston rod with a seal to keep the oil in and a wiper to keep the dirt out.
Figure 6.5: Double rod cylinders have two rods protruding, and are mainly used in off-road
equipment power steering applications.
weight ratio. Figure 6.4 shows a simple, small cylinder that is made by compressing a cap
end head, a barrel and a rod end head together with tie rods. The cap end and rod end
have ports that allow for hydraulic fluid to be pushed in and out of the cylinder. Inside the
cylinder barrel we find a piston that fits tightly inside the barrel, but there is also a dynamic
seal (this is not a positive seal) that separates the pressure side from the non-pressured side
of the cylinder. There are also seals between the cap and rod end heads and the barrel, but
these are static seals. The rod typically has a seal to keep the oil inside the cylinder and a
wiper that keeps dirt out. A cylinder that can be extended and retracted by hydraulic fluid
is called a double acting cylinder.
Figure 6.5 shows a double rod cylinder. These types of cylinders are used in application
where both sides of the cylinder need to do work, the most common application being steering
cylinders in off-road equipment. Double rod cylinders are always double acting.
Figure 6.6 shows a welded cylinder, which is much stronger than the tie rod cylinder
design. The maintain it, the rod end head is taken off and the whole piston-rod assembly
can be pulled out of the cylinder.
Figure 6.7 shows a ram (sometimes called plunger) cylinder. It has a unique design in
66
CHAPTER 6. CYLINDERS
Figure 6.6: Welded cylinders are used where high pressures and forces are encountered.
that the piston and rod are the same. This design enables generating very high forces, and
it is also highly bend resistant. These kinds of cylinders are most often found in high lift
applications such as the beds of dump trucks or in platform lifts. The ram cylinder is a
single acting cylinder, which relies of gravity of a spring force for retraction.
Figure 6.8 shows a telescopic cylinder, which are often used if the length of the retracted
cylinder must be low, but the total extended length must be high. Therefore these types of
cylinders are most often seen in dump trucks and high reach forklifts.
6.3.1
Cylinder cushioning
To avoid high stresses especially under load, often cylinders are fitted with a mechanism
that allow for adjustable cushioning. Figure 6.9 shows how cushioning is implemented. A
small rod protrudes from the piston on the cap end side. This small rod does NOT change
the area of the piston, since it is only the projected area that matters for force calculations.
When the cylinder retracts, the small rod is captured into a cavity that blocks the flow from
exiting through it. This forces the remaining fluid to pass through a small orifice which area
can be adjusted.
67
Figure 6.7: The ram design cylinder is unique in that the diameter of the piston is equal to
the diameter of the piston rod, in essence the piston and the rod are one. This makes the
design extremely bend resistant, justifying its use in applications where very high extension
forces are needed such as shown here in this Genie telescopic boom access platform. For
retraction, these cylinders have to rely on gravity or a spring force.
68
CHAPTER 6. CYLINDERS
Figure 6.8: Telescopic cylinders are used where a long extension distance is required but the
retracted length of the cylinder must remain short. This dump truck shows a good example
of the use of a telescopic cylinder.
69
Figure 6.9: Cylinder cushioning is implemented by having a small rod protruding from the
piston which closes off the regular oil passage at the cap end side, and forces the remaining
oil to pass through an adjustable orifice.
Agenda
Units, Pumps, Pressure Relief Valve
Cylinders
Double acting/ Single acting
Single rod, Double rod
ABE 223
ABE Principles Machine systems
Cylinder construction
Pressure, Flow, Work and Power in cylinders
Cylinder implementations
Cylinders
Tony Grift
Pascal
Flow
Newton
Watt
Area
Force
Work
Power
N
1 Bar 10 2 105 Pascal
m
5
Torque
N
p 2
m
m3
q
s
Pressure control
Pressure relief valve
Pressure reducing valve (regulator)
Pressure compensation: Only provide the pressure needed to move the
load(s). In idle reduce energy loss by providing a open center condition
Pressure compensated pump. Make the pressure independent of the
flow required to move the load at a preset speed. Deal with multiple
cylinders that need to move simultaneously
A m2
F N
W Nm
Flow control
Nm
Pow
s
TNm
m3
Displacement D
rad
3
N
F N p 2 * A m 2
m
ACAP 2* AROD
A
Hydraulic Cylinder
m3
q
s
m
v 2
s A m
m
Nm
F N * v P
Watt
s
s
ACAP 2* AROD
AROD
q
1
vEXT
* vRET * vRET
vEXT
ACAP
ACAP
2
q
vRET
Linear Power: Pow F * v
AROD
7
Cylinder construction
6
8
1.
2.
3.
4.
10
11
12
13
9
5.
6.
7.
8.
9.
10.
11.
12.
13.
Tie rod
Rod end head
Rod end port
Piston seals
pressure)
Cap end head
Cap end port
Rod bearing
Rod wiper
Rod seal
Barrel
Piston rod
Piston
Static seal
10
(
(
(
(
11
12
13
( mounting point)
( fluid entrance/exit point )
( lateral support of the rod )
( keeps dirt out )
(dynamic, seals fluid from environment )
( cylinder )
( mechanical force output )
( pressure to force converter )
( seals fluid from environment )
10
Telescopic cylinder
11
12
Pressure
gage
Actuator
1) Cylinder:
Double acting, differential area
2) Fluid is incompressible
3) Rod and cap end connected
Overload
Protection
Electric Motor
Reservoir Pump
Check in FluidSim
13
qC
qC
qR
s
s
pC
AC
AR
AC
FR
FC
pC
14
AR
pR
pR
AC s
AC v q
q
C
t
R
As
qR R AR v AC AR
t
qC
W F * s pC * AC * s
pR 0
15
pC
AC
qC
pC
pR
AC
FR
FC
AR
pR
W Fs s pC AC pR AR
F pC AC p R AR
FC
qR
FR
AR
qR
s
FC
16
FR
pR 0
17
18
qC
qR
AC
AR
pC
pR
AR
pR
W Fs s pC AC pR AC
A s
A s
W
s
F p C C p R R p C qC p R q R
t
t
t
t
qC
AC
FR
FC
W Fs s pC AC p R AR
Pow
qR
FR
FC
pC
Pow
A s
A s
W
s
F pC C pR C qC pC pR
t
t
t
t
p
qC
qR
19
Power if
qin (as
qoutin motors
q
and pumps)
qC
pC
AC
FR
FC
AR
20
qR
qC
pR
Pow q * p
21
22
Given
pPRV
Required
Torque needed to drive the pump
Power needed to drive the pump
Neglect friction
FROD pROD * AROD pCAP * ACAP pROD * AROD pROD CAP pCAP
AROD
FCAP FROD
23
Check in FluidSim
24
N
p PRV 15 MPa 15 E 6 2
m
cm 1 rev 1 m
m
D 10.54
*
* 6 3 1.6775 E 6 rad
rev 2 rad 10 cm
rad
rad 1 min
rev
* 2
*
1800
188.495
s
rev 60 s
min
3
PRV Setting
pump displacement of
the speed of the pump
Required
Torque needed to drive the pump
Power needed to drive the pump .
Neglect friction
m3
N
T D * p PRV 1.6775 E 6
*15 E 6 2 25.163 Nm
m
rad
Nm
rad
Pow T * 25.163Nm *188.495
4743 s
s
25
26
Answer in SI Units
ABE 223
ABE Principles Machine systems
TMECH D * p PRV
Cylinders
PowMECH T *
The End
Tony Grift
27
28
[ ]
Required:
1.
Compute the pressure needed to
extend the cylinder:
2.
Compute the velocity of the cylinder
and the total time needed for extension:
3.
How much work have we performed
during extension ?
5. How much power are we producing (energy being destroyed during idle ?)
1
For the parallel cylinder as shown in Figure 2 first calculate the pressure needed to extend
either cylinder. Then explain what is going to happen when we switch the DCV to
extend. The cylinder data is the same as in the previous question (the top cylinder has moved to
the right).
80
CHAPTER 6. CYLINDERS
Chapter 7
Valves
7.1
In the following sections, the Balanced Piston concept is explained along with its use in the
Balanced Piston PRV and Balanced Piston Sequence Valve.
7.1.1
The balanced piston has a land in which sensing orifice (a small hole) is drilled which acts
like an orifice (see Fig. 7.1). If you have a hard time imagining where the sensing orifice is
located, a 3D rendering of the valve stem is given in Fig.7.2. The function of the orifice is to
sense the primary system pressure. The poppet valve in the pilot section will remain closed
as long as the system pressure ( times the projected area of the poppet in contact with the
fluid) is smaller than the spring force. Note that in this situation there is no flow through
the orifice which implies that there is no pressure drop. Furthermore, note that the pilot
section is internally drained through the hollow stem of the piston because the dump port
is always connected to tank in the Pressure Relief Valve case.
7.1.2
As soon as the system pressure (sensed through the orifice) times the projected area of the
poppet in contact with the fluid is higher than the spring force, the poppet will open. This
means that the volume above the piston land is directly connected to tank (through the
hollow stem connection). Because of this pressure difference, the piston will rise and a direct
connection is opened between the system pressure port and the dump port. Note that there
is a small flow through the orifice since there is a pressure drop across it. It is tempting to
think that the system pressure drops to zero but it does not, because in that case the poppet
would close and pressure would build up again. Instead the pressure oscillates around the
point where the poppet alternatingly switches between seated and unseated, in other words
the system pressure fluctuates around the preset pressure.
If the pressure above the piston land happens to be equal to the pressure underneath it,
still the dump valve will remain closed, because of a small spring between the piston and
the housing.
81
82
CHAPTER 7. VALVES
83
Figure 7.2: 3D rendering of the hollow stem balanced piston, shown with the sensing orifice
in the land.
84
CHAPTER 7. VALVES
7.1.3
A PRV can be vented remotely by connecting the pilot stage of the valve to a directional
control valve which is connected to tank. The PRV functionality is still present, it is merely
augmented by the remote venting option.
7.1.4
Activation of this valve will force a pressure drop over the piston land and open up the valve
completely. The venting can only take place at pressures lower than the preset pressure
of the pilot stage, because the over pressure protection is still functional. When the valve
is vented the pressure in the system drops (virtually ) to zero, which means that the pump
is unloaded and no energy is wasted. Of course the pressure in the system must be at least
high enough to keep the piston lifted against the small spring force.
7.2
Sequence valve
The sequence valve is quite similar to the PRV. The only difference is the fact that the pilot
stage needs to be externally drained because the pressure in the secondary system is not
necessarily zero as was the case in the PRV where the dump port was always connected
to tank. The function is to allow system pressure from a primary source (A) to enter the
85
86
CHAPTER 7. VALVES
87
secondary system (B) only if the system pressure is higher than a certain preset value. The
primary pressure is again sensed through the orifice in the land and as soon as this pressure
(times the projected area of the poppet in contact with the fluid) exceeds the spring force
the poppet will unseat. This enforces a pressure drop above the piston land and fluid will
flow from the primary to the secondary circuit without obstruction. The system pressure
can exceed the preset pressure of the pilot stage because there can be a pressure buildup
underneath the piston land, whereas in the PRV case this pressure was always zero. In
practical applications the stem of the piston is still hollow, however the fluid will never take
this path since there is a shortcut to tank through the pilot stage.
7.3
The valve is built from three parts which can be configured such that three different types
of valves emerge by switching the orientation of the top and bottom parts. There are two
pistons present to cover a wide range of operating pressures.
7.3.1
The back pressure valve is in fact a direct acting PRV ( Pressure relief valve) because the
secondary circuit is connected to tank. The pressure is sensed upstream, and when the
pressure on the small piston (divided by the area) is higher than the required spring force,
flow will be diverted to the tank preventing further pressure build up. The pressure drop
from the primary to the secondary circuit is constant and determined by the spring setting.
The channel drilled through the center of the spool prevents pressure build up caused by
leakage past the (small) piston.
7.3.2
Unloading Valve
The unloading valve differs from the back pressure (PRV) valve, in that the pressure is not
sensed in the secondary circuit, but externally. It can also be characterized as a remotely
operated direct acting pressure relief valve. In normal mode, there is free flow from the input
to the output, and as soon as the external pressure rises high enough to overcome the spring
force, the valve starts dumping to tank.
7.3.3
Sequence Valve
In a sequence valve, there must be a preset pressure available in the primary circuit, in order
to have a pressure in the secondary circuit. The classical example is a piece of material that
needs to be clamped before a hole can be drilled. The pressure is sensed in the secondary
circuit and the drain is external since neither one of the ports is connected to tank.
7.3.4
When free flow is required, the sequence valve can be fitted with a check valve that is
connected between the primary and the secondary circuit. Note again that the drain is
88
CHAPTER 7. VALVES
Figure 7.7: Multipurpose direct acting pressure control valve, configured as a PRV.
89
90
CHAPTER 7. VALVES
Figure 7.8: Multipurpose direct acting pressure control valve, configured as an unloading
valve.
91
Figure 7.9: Multipurpose direct acting pressure control valve, configured as a sequence valve.
92
CHAPTER 7. VALVES
7.3.5
A counter balance valve is used in case of overrunning loads such as a cylinder load falling
by gravity or a vehicle with a hydrostatic transmission driving downhill. In the normal
situation there must be free flow, therefore a check valve connects the reverse flow circuit. In
the forward flow situation, the pressure is controlled as in sequence valve. Note that there
are only two connections, there is no drain line connected, the valve merely acts as a variable
orifice type pressure regulator.
93
Figure 7.10: Multipurpose direct acting pressure control valve, configured as a sequence
valve with integral check valve from upstream to downstream circuit.
94
CHAPTER 7. VALVES
Figure 7.11: Multipurpose direct acting pressure control valve, configured as a counterbalance valve.
Agenda
Directional Control Valves (DCVs)
ABE 223
ABE Principles Machine systems
Valves
Tony Grift
Throttle
Pressure Compensated Flow Control Valve
Orifice equation derivation
Turbulence prevention
Metering methods
FLOW
CONTROL VALVES
PRESSURE
DIRECTION
Actuation by
PRV
FCV
DCV
lever
foot pedal
mechanically
spring return
electrically (solenoid valves)
Check
Two way
Three way
Four Way
Proportional (not discussed)
: Cylinder/motor connections
: Pressure port
: Tank port
Numbering system
Check Valve
2-1 Valve
10
11
12
Three-way Valves
3-2 Valve
normally open
2-2 Valve,
normally closed
13
14
15
Actuation Methods
16
18
Center conditions
Closed center
19
20
Exercise
21
22
Temperature compensation
23
24
25
26
27
28
Orifice Equation
q Co Ao
2 p
Co Determined experimentally 1
N
p Pressure drop proportional to flow rate 2
m
q Co Ao
2 p
kg
30
31
32
Rule of thumb
*v* D
Re Reynolds number 1
kg
Density of fluid 3
m
m
v Velocity of Fluid
s
D Characteri stic dimension m (here diameter of pipe)
kgm 1
kg m s
Absolute viscosity 3 m 2
s Pa * s
s m2
m s s
33
Am
34
px
Meter Out
Hand knob
Automatic orifice
Aa
Bleed-off
By bleeding the excess fluid off to the reservoir
Inlet port
pi
Outlet port
po
q
35
36
Meter-In Circuit
Meter-In Circuit
37
38
39
Meter-Out Circuit
40
41
42
43
Bleed-off circuit
44
45
46
47
48
Sequence Valves
To sequence hydraulic operations
Unloading Valves
To unload the pump during idle cycles
Counterbalance
49
50
51
52
System Pressure
53
54
To Reservoir
System Pressure
55
56
57
58
Pilot Operated
Relief Valve
Cracking Pressure
System Pressure
59
60
10
Unloading a pump
An unloading valve can directly connect the pump to tank to
save energy
61
62
63
64
Sequence Valves
65
66
11
67
Counterbalance Valves
68
Counterbalance Valves
300 cm2
Optional
Remote Pilot
Direct Pilot
Direct Pilot
150 cm2
90 kN
Unloading valve is set to 10% above the system pressure
required to hold the load.
69
Answer
70
10F
A
90kN
p 10
6MPa
150cm2
p 6.6MPa
71
72
12
Unloading valve
Sensing?
Up/downstream
Sensing?
Up/downstream
73
74
Sequence valve
Sensing?
Up/downstream
75
76
Sequence
Unloading
PRV
Sensing?
Up/downstream
77
78
13
ABE 223
ABE Principles Machine systems
Valves
The End
Tony Grift
79
14
Chapter 8
Regenerative circuits
A regenerative circuit is a ingenious way to take advantage of the differential area cylinder.
Imagine that we have a situation where a cylinder is used to clamp a workpiece. First we
need to extend (advance) the cylinder to the workpiece, and we would like to do that fast,
since there is no load on the cylinder. This can be done in regenerative mode. However, as
soon as we make contact with the workpiece, we want to advance slowly, and use the full
force available. This is what happens in conventional mode, where we try to bring the rod
end pressure to zero, so we have maximum force at a low speed.
Regenerative circuits comprise a somewhat advanced topic in hydraulics, where only
during extension, the cap end and the rod end of a differential area cylinder are connected
to each other. This may seem like a strange thing to do, but there are advantages in certain
applications. Remember that there is only regenerative extension (advance), there
is no such thing as regenerative retraction.
8.1
Conventional extension
To illustrate regenerative circuits, lets first go back to conventional extension. Figure 8.1
shows a conventional setup. It is clear that during extension (when the lever operated,
detented, tandem center DCV is pushed downward), the cap end of the cylinder is connected
to the pump, and the rod end is connected to the tank. Lets find out what the forces and
velocities are during conventional extension.
Figure 8.2 shows the cylinder with forces annotated. Note that the external force (load)
is opposing, which means that the direction of the force is opposite the direction of travel
of the piston. Now we can write a force balance as follows where forces to the right are
considered positive:
FC FR F = 0
When we plug in Pascals law p =
(8.1)
F
we get:
A
pC AC pR AR F = 0
Since in conventional mode, the rod end is connected to the tank we get:
109
(8.2)
110
Figure 8.1: Conventional circuit where during extension, the lever operated, detented DCV
connects the cap end to the pump and the rod end to the tank.
111
pC AC = F = pC =
F
AC
(8.3)
The maximum force we can generate in conventional mode is, limited by the PRV setting
at FM AX = PP RV AC .
We can also calculate the extension (and retraction) velocities in conventional mode as
follows:
v=
qC
qR
=
AC
AR
(8.4)
Here the flow into the cap end is the pump flow (assuming that the PRV is set high
qP U M P
enough and no flow goes through it). Therefore the extension velocity is vEXT =
.
AC
When the cylinder retracts, the same pump flow goes into the rod end, therefore the retracqP U M P
AC
2
tion velocity is vRET =
. For instance, for a
= cylinder, the retraction velocity
AR
AR
1
is twice as high as the extension velocity.
8.2
Regenerative extension
Now lets have a look at regenerative extension, as shown in Figure 8.3. It is clear that now
the cap end is still connected to the pump, but the rod end is connected to the cap end!
This has two effects. 1) The pressure on the cap and rod end are the same pC = pR = p,
and 2), the point where the cap and rod end connect is called a node. The force equilibrium
now becomes:
pAC pAR F = 0
(8.5)
leading to:
pC = pR = p =
F
AC AR
(8.6)
The maximum force we can generate in regenerative mode is again, limited by the PRV
setting at FM AX = PP RV (AC AR ).
To calculate the velocity during extension, we have to realize that the flow into the cap
end is not just the pump flow, but the regenerated flow from the rod end is added to it. In
general in a node, the algebraic sum of the flows (with the proper sign, lets assume flows
into the node are positive) must be zero. In other words, the flow rate of all flows into the
node must be equal to the flow rate of all flows leaving it. In electronics we call this law
Kirchoffs Current Law. It is clear that the pump flow points toward the node, as does the
regenerated rod end flow, and the cap end flow points away from the node. Therefore we
can write:
112
Conventional mode
FM AX = PP RV AC
qP U M P
vEXT =
AC
Regenerative mode
FM AX = PP RV (AC AR )
qP U M P
vEXT =
AC AR
Table 8.1: Comparison between conventional and regenerative modes in terms of maximum
force and extension velocity.
qpump qC + qR = 0
(8.7)
In addition, since this is a differential area cylinder, we can find the relationship between
the flow into the cap end and rod end from Eqn. 8.4
qC
qR
AR
=
= qR =
qC
AC
AR
AC
Substitution into Eqn 8.7 gives:
v=
qpump qC + qR = qpump = qC +
AR
qC
AC
(8.8)
(8.9)
and
qC = qpump
AC
1
= qpump
AR
AC AR
1+
AC
(8.10)
AC
2
=
cylinder we would get: qC = 2qpump . The extension velocAR
1
qP U M P
ity during conventional mode was vEXT =
, but here in regenerative mode it is
AC
qP U M P
vEXT =
. Table 8.1 shows an overview of the difference beween conventional and
AC AR
regenerative mode.
Again, for a
113
qR
pC
Regenerative
Circuits
FR
FC
AC
AR
pR
pC AC p R AR
Force Equilibrium F
FR
FC
Tony Grift
Velocity of piston
Dept. of Agricultural & Biological Engineering
University of Illinois
q
q
v C R
AC AR
2
1
3
qC
qR
AC
Force Equilibrium
AR
F pAC
pAR p AC AR
FR
FC
Supply manifold
qC
qR
qC
qR
KCL in node N
AC
AR
qIN qR qC
AC
AR
qIN qR qC
AR
A AR
qC C
qR qC
qIN qC 1
AC
AC
AR AC
qC qR
AC AR
Summary
Conventional
qR
qC
Force:
Velocity:
AC
AR
Regenerative
F pAC
F p AC AR
q IN
v
AC
q IN
AC AR
A AR
AC
qC qIN
qIN qC C
AC
AC AR v qIN
AC AR
q
v C
AC
Summary cont.
1
3
AC
Rod Area
Supply manifold
AC AR
AR
10
Regenerative
(pressurize cap
and rod end)
Conventional
(rod end to tank)
11
12
13
14
15
16
Figure 1: Here a simple regenerative circuit is shown. This cylinder can only extend, it will
never retract. Why is this?
Chapter 9
Bernoullis law and orifice equation
In this chapter, Bernoullis principle will be derived and some applications in hydraulics
studied. Bermoullis law was derived by Daniel Bernoulli, and it is essentially based on an
energy balance between two points, under the assumption that the total energy must remain
constant between them. This is called energy conservation. To build on what you already
know, lets start with finding out what forms and how much energy are in solid objects. To
determine how much energy is in an object, we are going to find out what it took to get
them in the state they are in:
If we take a solid object with mass m, we can calculate its potential energy since for
the object to reside where it is in a gravitational field, some force must have acted upon it
through a certain distance:
Epot = mgz
(9.1)
where, E is the energy in Joule, m is the mass in kg, g is the gravitational acceleration
in ms2 and z is the height above some reference surface (aka datum). When we drop
this object, the equations of motion are (assume constant gravitational pull, well switch to
symbol a for acceleration):
a(t) = a
Z
v(t) = atdt + v(t = 0)
Z
1
x(t) = vtdt = at2 + v(t = 0)t + x(t = 0)
2
(9.2)
(9.3)
(9.4)
If the initial position x(t = 0) and initial velocity v(t = 0) are zero this reduces to:
a(t) = a
v(t) = at
1
x(t) = at2
2
119
(9.5)
(9.6)
(9.7)
120
Ekin = f x = ma x = ma
1 2
V
2a max
1
2
= mVmax
2
(9.9)
This is the kinetic energy contained in our object. The maximum velocity an object can
attain is when all potential energy is converted into kinetic energy or when:
p
1
2
= Vmax = 2gz
Epot = Ekin = mgz = mVmax
2
9.1
(9.10)
In general, there are three kinds of energy in a solid object being potential Epot , kinetic Ek in
and internal Eint , all of course with unit N m or Joule: Assembling their equations gives:
Epot = mgz
(9.11)
1
Ekin = mv 2
2
(9.12)
Eint = mch T
(9.13)
where ch is the specific heat of the fluid in Jkg 1 K 1 and T is the absolute temperature
in Kelvin. Since energy can neither be created nor destroyed, there is an energy balance:
Epot1 + Ekin1 + Eint1 = Epot2 + Ekin2 + Eint2
(9.14)
In hydraulics, there is a fourth form of energy contained in the fluid: Pressure. Remember
that Power in a hydraulic system was pressure times flow rate or:
3
Nm
N
m
P ow
=p
q
(9.15)
2
s
m
s
Lets convert energy per unit of mass as follows:
Nm
N m kg
P ow
=
s
kg
s
kg
where the massflow rate in
can be written as:
s
(9.16)
3
kg
m
kg
=
=q
2
s
m
s
q
3
3
m
m
s
121
(9.17)
(9.18)
To put all energies on the same mass basis we have to divide the remaining three energy forms by the mass as well, otherwise the energy conservation equation would not be
dimensionally homogeneous.
p Nm
Epot =
(9.19)
kg
Nm
(9.20)
Epot = gz
kg
1 2 Nm
Ekin = v
(9.21)
2
kg
Nm
Eint = ch T
(9.22)
kg
Epress1 + Epot1 + Ekin1 + Eint1 = Epress2 + Epot2 + Ekin2 + Eint2
(9.23)
This leads to the conservation of energy law or Bernoullis law on a unit mass basis:
1
p2
1
p1
+ gz1 + v12 + ch T1 =
+ gz2 + v22 + ch T2
(9.24)
2
We can also convert this equation in a pressure form by multiplying the LHS and RHS
by .
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
(9.26)
2
2
Finally there is a head form which requires
multiplying the LHS and RHS by the
N
specific weight (not mass!) where = g
:
m3
p1
1
ch
W
p2
1
ch
+ z1 + v12 + T1 =
+
+ z2 + v22 + T2
(9.27)
2g
g
g
2g
g
122
In hydraulics we usually ignore the gravity induced pressures because they are much
smaller than the pressure typically present due to loads and resistance. Also, there is usually
not much height to hydraulic systems. That means that we assume that z1 = z2 which
reduces our Bernoulli equation (for example in pressure form) to:
1
1
p1 + v12 + ch T1 = W + p2 + v22 + ch T2
2
2
9.1.1
(9.28)
(9.30)
(9.31)
(9.32)
therefore:
1
Nm
p1 = W + p2 = W = (p1 p2 )
kg
This is the amount of work that is done per kg of fluid (check the units!)
9.1.2
(9.33)
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
(9.34)
2
2
The assumption that the initial velocity v1 = 0 is a rather bold one, this is only the case
if the pipe diameter would be very large. Later well look at a more detailed derivation that
will lead to the orifice equation.
z1
v1
T1
W
= z2
=0
= T2
=0
1
p1 = p2 + v22 = v2 =
2
(9.35)
(9.36)
(9.37)
(9.38)
2 (p1 p2 ) 1
ms
(9.39)
123
Figure 9.1: Shown is a measurement orifice, where the pressure drop measured across the
orifice is a function of the flow rate through it.
9.1.3
Figure 9.1 shows a tank which is assumed much larger than the orifice (in fact so large we
assume that v1 = 0).
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
2
2
p1
v1
T1
W
= p2
=0
= T2
=0
p
p
1
gz1 = gz2 + v22 = v2 = 2g (z1 z2 ) = 2gh ms1
2
9.1.4
(9.40)
(9.41)
(9.42)
(9.43)
(9.44)
(9.45)
The idea here is that the fluid encounters friction in a pipe, and therefore a pressure drop
will occur. However, this lost pressure energy must go somewhere, and it does; it heats up
the fluid. The flow rate however remains the same (therefore v1 = v2 !)
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
2
2
(9.46)
124
Figure 9.2: Shown is the elegant Venturi tube which uses a manometer with its own fluid to
indicated the pressure difference as shown.
z1 = z2
v1 = v2
W =0
p1 + ch T1 = p2 + ch T2 = T = (T1 T2 ) =
9.1.5
(9.47)
(9.48)
(9.49)
p2 p1
[K]
ch
(9.50)
The Venturi tube as shown in 9.2 is a simple device that uses the pressure drop in an orifice
to infer the flow velocity in ms1 (or flow rate in m3 s1 ) of the fluid.
leading to:
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
2
2
(9.51)
z1 = z2
T1 = T2
W =0
(9.52)
(9.53)
(9.54)
125
1
1
1
p1 + v12 = p2 + v22 = p1 p2 = v22 = v12
(9.55)
2
2
2
Apart from these assumptions, we also assume continuity, which means that as the fluid
is incompressible, the amount of fluid at location 1 must be equal to the amount of fluid at
location 2 or A1 v1 = A2 v2 . Substitution gives:
!
2
A1
1 2
p1 p2 = v1
1
(9.56)
2
A2
If we measure the pressure differential with a manometer that contains a fluid with a
density of man we get:
!
2
A1
1 2
1
(9.57)
p1 p2 = hman g = v1
2
A2
Solving for v1 gives:
v
u 2man g
u
u
v1 = u
h
u 2
t A1
1
A2
(9.58)
v1 = K h
(9.59)
Since the factor under the first square root function is constant we can write:
9.1.6
Pitot tubes are widely used to measure the velocity of a fluid (or a gas like in aircraft). The
pressure difference is obtained from two ports, the dynamic (where the fluid has a velocity)
and the static (drilled in the side wall) where the fluid is assumed without velocity.
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
2
2
z1
T1
W
v2
= z2
= T2
=0
=0
1
p1 + v12 + = p2 = v1 =
2
(9.60)
(9.61)
(9.62)
(9.63)
(9.64)
2 (p1 p2 ) 1
ms
(9.65)
126
Figure 9.3: A Pitot tube is an ingenious device that is used widely in aviation and marine
application to measure air or water speed. A pitot tube has a dynamic port (in the front)
as well as a static port (in the side).
v1 =
9.1.7
2
man gh =
2
man g h = K h ms1
(9.66)
Figure 9.4 shows an orifice, where the pressure drop is measured as a difference in height
(aka head) of the fluid.
1
1
p1 + gz1 + v12 + ch T1 = W + p2 + gz2 + v22 + ch T2
2
2
(9.67)
z1 = z2
T1 = T2
W =0
(9.68)
(9.69)
(9.70)
leading to:
1
1
p1 + v12 = p2 + v22
(9.71)
2
2
As we did in the case of the Venturi tube, we assume continuity, or A1 v1 = A2 v2 . Substitution gives:
"
2
2 #
1 2 1
A2
1 2
A2
p = p1 p2 = v2
v2 = v2 1
(9.72)
2
2
A1
2
A1
127
Figure 9.4: Shown is an example of the use of the head form of Bernoullis equation in a
Venturi tube where the pressure heads are different from location 1 to location 2.
A fluid can not follow a 90 degree angle but will form a horn like jet with a reduced
diameter and area. This is called the Vena contracta phenomenon characterized by a constant
CC yielding A2 = CC A0 . Substitution gives:
v
u
"
u
2 #
u
1 2
CC A 0
p = v2 1
= v2 = u
u
2
A1
t
2p
2
CC A0
1
A1
(9.73)
The flow rate is the velocity times the cross sectional area of the conduit q = A2 v2 and
since A2 = CC A0 we get q = CC A0 v2
v
u
u
u
q = A2 v2 = CC A0 u
u
t
2p
s
CC A 0
p
2
2 = s
2
CC A0
CC A0
1
1
A1
A1
(9.74)
This relationship is only valid for high Reynolds numbers, where viscous effects can be
ignored. For small Reynolds numbers is the discharge coefficient CD is introduced as CD =
CC CV . Substitution gives:
CD
q=s
2 A0
CC A0
1
A1
(9.75)
128
(9.76)
The orifice equation relates the flow rate through a resistance (nozzle, orifice) to the pressure drop across that resistance. In electronics we call that relationship Ohms Law U = iR,
where U is the voltage drop, i is the current in Ampere and R is the resistance in Ohm.
However, whereas Ohms law is extremely simple and linear, the orifice equation is non-linear
and involves many physical parameters such as geometries and fluid properties.
The value of K is a function of the Reynolds number which is defined as:
Re =
vd
(9.77)
where is the density of the fluid in kgm3 , v is the velocity of the fluid in ms1 , d is the
characteristic diameter in m, and is the dynamic (absolute) viscosity of the fluid in P as.
vd
(9.78)
4q
Through a circular conduit the flowrate is q = vA2 = v d2 of v = 2 and the Reynolds
4
d
vd
4q
number becomes: Re =
=
. For a given ratio of the diameter of the orifice d and the
d
d
pipe D, being , the K-value can be obtained from a graph that is shown in 9.5. First for a
D
given flowrate q, an orifice diameter d and the kinematic viscosity of the fluid , you deter4q
mine the Reynolds number in the orifice using Re =
which goes on the x-axis. Then you
d
d
ratio, and you read the K-value on the y-axis.
move upward to a curve that represents your
D
When the flow rate q has to be determined from a certain measured pressure head, the
factor K can not be determined because it depends on the flow rate q itself (remember that
4q
Re =
). However, there is a trick for that: Remember that we had the orifice equation
d
from equation 9.76 repeated here:
s
p
q = KA0 2
(9.79)
129
Figure 9.5: Shown is a chart, where the K-value that goes into the orifice equations can be
determined.
130
If we write this in a head form we have to use the conversion p = hg which gives:
q = KA0
Since:
p
p
2gh = K d2 2gh
4
(9.80)
4K d2 2gh
4q
Kd p
4
Re =
==
=
2gh
d
d
(9.81)
Re p
d
= 2gh
K
(9.82)
The slanted lines in figure 9.5 represent this relationship. The value of K can be computed
using the top scale and the slanted lines. Substitution of K in the equation q = KA0 2gh
now gives the flow rate q.
at a
ABE 223
ABE Principles Machine systems
vt adt at v0
xt vdt
Bernoullis Law
1 2
at v0 t x0
2
If the intial position and velocity are zero this reduces to:
Tony Grift
at a
vt at
xt
Dept. of Agricultural & Biological Engineering
University of Illinois
1 2
at
2
1 2
at
2
mgz
2 a
1 2
mv v 2 gz
2
2a
1
1
2
2
E Fx ma * x ma Vmax mVmax
2a
2
3
E pot mgzNm
1 2
mv Nm
2
Eint mchT Nm
Ekin
Nm Nm kg
Pow
*
s kg s
3
kg kg m
s m3 * s * q
p 2 * q p 2
m p Nm
m s
3
kg kg
kg m
3 * q 3
m
m s
There are four forms of energy in a fluid object, here all are given
on a unit mass basis:
Pressure
E press
Potential
p Nm
p1
kg
Nm
E pot gz
kg
Kinetic
Ekin
per kg of fluid
kg
1 2 Nm
v
2 kg
p1
Nm
Eint chT
kg
Internal
1 2
p
1 2
gz1 v1 chT1 2 gz 2 v2 chT2
2
1 2
p
1 2
gz1 v1 chT1 W 2 gz 2 v2 chT2
2
p1 gz1
1 2
1
N
v1 chT1 W p2 gz2 v2 2 chT2 2
2
2
m
Multiplying left and right gives various forms of the equation but
the energy conservation remains key
N which is the
There is the head form, divide by g
m 3
specific weight (not mass!)
p1
z1
Nm
kg
Nm
p1 1 2
p2 1 2
v chT1 W
v chT2
2 1
2 2
kg
1 2 ch
1 2 ch
W p
v1 T1 2 z 2
v2 T2 m
2g
2g
g
g
g
Pressure form 2
m
p1
1 2
1
N
v1 chT1 W p2 v2 2 chT2 2
2
2
m
Head form m
W p
1 2 ch
1 2 ch
v1 T1 2
v2 T2 m
2g
g
g 2g
g
p1
p1 gz1
1 2
1
N
v1 chT1 W p2 gz2 v2 2 chT2 2
2
2
m
10
p1 gz1
1 2
1
N
v1 chT1 W p2 gz2 v2 2 chT2 2
2
2
m
Assumptions
Assumptions
z1 z 2
z1 z 2
v1 v2
v1 0
T1 T2
T1 T2
W 0
p1 W p2 W
p1 p2
p1 p2
11
1
2 p1 p2
v2 2 v2
2
12
p1 gz1
1 2
1
N
v1 chT1 W p2 gz2 v2 2 chT2 2
2
2
m
Assumptions
Assumptions
v1 0
p1 p atm
p1 p2
p1 patm
p1 p2
T1 T2
v1 0
W 0
T1 T2
p2 p atm
v2
W 0
p2 patm
v2
v1 0
gz1 gz 2 v2 2 v2 2 g z1 z 2
1
2
v1 0
13
p1 gz1
14
A1v1 A2 v2
Assumptions
1 2
1
N
v1 chT1 W p2 gz 2 v2 2 chT2 2
2
2
m
z1 z 2
T1 T2
Assumptions
W 0
A1
z1 z 2
A1
A2
v1 v2
W 0
p1 chT1 p2 chT2 T T1 T2
v1
v2
v1
p1
p2
p1
p2 p1
ch
15
A1
2
p1 p2 1 v12 A1 1
2
A
2
16
A
2
p1 p2 h man g 1 v12 A1
2
A1
A1
2 man g
A1
A2
A2
v1
v2
v1
p1
p2
p1
v1
v2
v1
p1
p2
p1
v1
A1
1
A2
A1v1 A2 v2
h
h
z1 z 2
v1 K h
T1 T2
W 0
17
18
Pitot tubes are widely used to measure the velocity of a fluid (or a
gas like in aircraft). The pressure difference is obtained from two
ports, the dynamic (where the fluid has a velocity) and the static
(drilled in the side wall) where the fluid is assumed without
velocity.
v1
p1
Assumptions
z1 z 2
p2
T1 T2
1
2
p
p2
2
1
W 0 p1 2 v1 p2 v1
v2 0
When a manometer is used with fluid density
v1
man gh
T1 T2
W 0
1 2
1
v1 p2 v2 2
2
2
v1 A1 v2 A2
p1
Continuity equation
h1
man
h2
man g h K h
A1
A0
19
1 2
1
v1 p2 v2 2
2
2
v1 A1 v2 A2
p1
A fluid can not follow a 90 degree angle but will form a horn like
jet with a reduced diameter and area. This is called the Vena
contracta phenomenon characterized by a constant CC
A2
20
p1
1 2
1
v1 p2 v2 2
2
2
A2
v2
A1
v1
2
A 2
1
1 A2
1
2
2
p p1 p2 v2
v2 v2 1 2
2
2 A1
2
A
1
v1
A2 CC A0
A2 CC A0
h1
h2
C
A
1
p v2 1 C 0 v2
2
2
A1
1 CC A0
A1
A1
A0
A2
21
The flow rate is the velocity times the cross sectional area of the
conduit:
CC A0
C A
1 C 0
A1
q A2 v2
C A
1
v2 2 1 C 0
2
A1
v2
2
C
A
1 C 0
A1
q A2 v2 CC A0
2
1 CC A0
A1
22
CV CC C D
q
A2 CC A0
CV CC A0
C A
1 C 0
A1
C D A0
C A
1 C 0
A1
CC A0
C A
1 C 0
A1
23
CD
C A
1 C 0
A1
q KA0 2
Orifice Equation
24
q KA0 2
kg
3
m
m
v velocity of the fluid in
s
d ' characteristic' diameter in m
absolute viscosity in Pa s
vd
Re
vd
q vA2 v
Re
Re
Re
vd
vd
kinematic viscositym 2 s
d2
4q
kinematic viscositym 2 s
d Conduit diameter
D Orifice diameter
25
Re
d
2 g h
K
26
When the flow rate q has to be determined from a certain measured head or
pressure drop, the factor K can not be determined because it depends on the
flow rate q itself.
Re
vd
4q
q KA0 2 gh K
4 K d 2 2 gh
4q
4
Kd
Re
d
d
d 2 2 gh
2 gh
Re
d
2 gh
K
Re
4q
27
p1
p1
1 2
1
v1 p2 v2 2
2
2
p2
v1
2 p p
1
2
q A2 v2 A2
2
A2
1
A
1
28
1 2
1 2
v1 h2
v2
h1
p1 1 2 p2 1 2
2g
2g
v1
v2
2g
2 g
h1
, h2
p2
h h1 h2
v2
v1
v2
A2
A1
p1
A2
29
A1
2 g h h
1
2
q A2 v2 A2
2
A2
A1
30
ABE 223
ABE Principles Machine systems
Bernoullis Law
The End
Tony Grift
31
32
Chapter 10
Pressure drop in pipes and orifices
10.1
If a pipe conducts fluid, the flow through the pipe is constant, since the fluid cannot leak out.
If any energy is lost due to friction, that means that only the fluid pressure can change. To
calculate how much pressure drop a pipe will cause, we need to go back to elementary fluid
dynamics. There are two flow regimes in pipes, being laminar and turbulent. An example
of laminar and turbulent flow can be seen from a smoking cigarette. Close to the cigarette,
the smoke travels upward in straight lines, but at some point the flow will become turbulent.
The indicator for the flow regime is the dimensionless Reynolds number (Re) which is
defined as:
vd
Re =
(10.1)
where [kgm3 ] = the density of the fluid, v[ms1 ] is the velocity of the fluid, d[m] is the
diameter of the pipe and [P as] is the dynamic (absolute) viscosity of the fluid. Another
notation that is sometimes used is as follows:
Re =
vd
(10.2)
cross sectional area, the Reynolds number determines whether laminar flow (Re < 2000),
transitional flow 2000 < Re < 4000 or turbulent flow (4000 < Re) is present. In the case of
laminar flow, the flow profile in the pipe will be quadratic with no fluid velocity component
at the pipe wall (this is called the no-slip condition). This type of flow is called HagenPoisseuille flow (type this in a search engine to learn the proper pronunciation; it is not
poi-sell!). In the laminar flow case, the equation to calculate the pressure drop in a pipe is
64
easy, since the friction factor f =
. Now the pressure drop can be calculated using the
Re
DArcy-Weissbach equation which reads:
where [m2 s] is the kinematic viscosity of the fluid where =
p = f
L1 2
v
d2
137
(10.3)
138
Figure 10.1: The Moody diagram allows for calculation of the Friction Factor in the calculation of pressure drop in pipes. In the case of laminar flow (Re < 2000), the Friction
64
. In turbulent flow (Re > 4000) the Friction Factor depends on
Factor is simply f =
Re
the Relative Pipe Roughness . In the transitional region, the Friction Factor is unreliable,
d
because external forces can turn a laminar regime into a turbulent regime without warning.
In this region assume that the flow is turbulent (worst case scenario)
.
64
is the Friction Factor, L[m] is the length of the pipe and d[m] is its
Re
diameter, [kgm3 ] is the density of the fluid and v[ms1 ] is the velocity of the fluid, which
q
can be calculated from the flow rate through the pipe assuming incompressibility as v = ,
A
where q[m3 s1 ], is the flow rate and A[m2 ] is the cross sectional area of the pipe.
where f =
If the flow is turbulent, meaning that Re > 4000, the situation is very different. We can
no longer assume that there is no velocity component at the wall, and in fact the wall has a
distinct effect on the pressure drop in the pipe. Figure 10.2 shows the velocity profiles of a
fluid in a pipe in the laminar (a) and turbulent (b) regime.
Although we will cover the pressure drop in the turbulent regime, it is very important
139
Figure 10.2: (a): Flow profile in case of laminar flow. Notice that the flow velocity at the
pipe wall is zero. (b): flow profiles in the case of turbulent flow. Notice that the velocity
at the wall is non-zero, in fact the wall roughness has a distinct effect on the pressure drop
along the pipe.
that a designer of hydraulic systems prevents turbulent flow in pipes at all cost, since it
increases the pressure (energy) loss, and it introduces vapor bubbles that cause cavitation.
Nevertheless, in the turbulent regime we can read the friction factor from the Moody diagram
as shown in Figure 10.1. One needs to know the Relative Pipe Roughness of the pipe which
is expressed in the ratio of the average of the wall protruberances and the diameter of the
pipe as in . When this number is known, the associated curve is followed until it intersects
d
with the Reynolds number. Then the Friction Factor can be read on the left scale.
10.2
To calculate the pressure drop across an orifice, we need to use the orifice equation as shown
in Equation 9.76 repeated here:
s
p
q = KA0 2
(10.4)
Lets rewrite this a little so it looks more like the DArcy-Weissbach equation by assuming
q
incompressibility as v =
, where q[m3 s1 ], is the flow rate and A0 [m2 ] is the cross sectional
A0
area of the orifice.
s
p
vA0 = KA0 2
(10.5)
140
p =
1 1 2
v
K2 2
(10.6)
1
is
K2
also known (lets just call that thing K too). That factor is now the friction factor and we
can write the equation as:
Since the factor K is determined experimentally, we can assume that the factor
1
p = K v 2
2
Where K is a experimentally determined friction factor.
10.3
(10.7)
Summary
For pressure drop calculations in pipes, you get the friction factor from the Moody diagram
(Figure 10.1) but you need to first calculate Reynolds number as:
Re =
vd
(10.8)
where [kgm3 ] = the density of the fluid, v[ms1 ] is the velocity of the fluid, d[m] is the
diameter of the pipe and [P as] is the absolute viscosity of the fluid. Now for the pressure
drop calculation, you use the DArcy Weissbach equation with the friction factor from the
Moody diagram:
L1 2
v
(10.9)
d2
where f is the Friction Factor, L[m] is the length of the pipe and d[m] is its diameter,
[kgm3 ] is the density of the fluid and v[ms1 ] is the velocity of the fluid, which can be
q
calculated from the flow rate through the pipe assuming incompressibility as v = , where
A
q[m3 s1 ], is the flow rate and A[m2 ] is the cross sectional area of the pipe.
p = f
For pressure drop calculations in orifices you use the (slightly modified) Orifice Equation
and you get the friction factor K from a data sheet.
1
p = K v 2
2
(10.10)
Remember that a good design avoids turbulent flow in pipes at all times. Turbulence
cannot be avoided in valves, when they just open, they will cause very high velocities (the
pump is trying to squeeze all pump flow through a very small opening) and therefore the
pressure will drop (from Bernoullis equation), and cavitation will occur. This is not a big
deal, as you know, bubbles (either from air entrainment or from the fluid boiling due to
low pressure) will be taken out by the tank. It is however unacceptable to let bubbles pass
a pump, because over time, cavitation will destroy a pump.
= 800[ kgm3 ]
1000N
m2
= 32*106
10
= 0.004
Equations
Reynolds
Re =
vd
Diameter of piping 10 [ m]
2
11
8
Orifice equation
1
p = K v 2
2
DArcy-Weisbach eq.
12
13
L 1
p = f v 2
D 2
Pipe lengths:
0-1
2-3
4-5
6-7
8-9
10-11
12-13
0.1 m
0.1 m
0.2 m
0.2 m
0.4 m
0.6 m
0.1 m
5
4
K-values of components:
1-2
5-6
7-8
11-12
10
15
20
25
(filter)
(PRV)
P-A DCV
B-T DCV
2
1
0
10
0
11
12
13
Patm
3
0
Questions:
1.
vd
L 1
p = f v 2 , Re =
D 2
Section L
0-1
2-3
4-5
6-7
8-9
10-11
12-13
0.10
0.10
0.20
0.20
0.40
0.60
0.10
Re
12.73
12.73
12.73
12.73
12.73
6.37
6.37
Calculate the pressure losses in all Components using the Orifice equation
p = K
Component path
1-2
5-6
7-8
11-12
10.00
15.00
20.00
25.00
1 2
v
2
E=
PLOAD
PLOAD + PLOSS
(1.1)
5. Moody diagram
2p
(1)
m3
where q
is the flow rate through the orifice, A0 [m2 ] is the opening area of the orifice,
s
N
kg
p
is the pressure drop across the orifice,
is the density of the fluid, and C0
2
m
m3
is an experimentally determined calibration constant, derive an equation for:
px = f (pi , po , Am , Aa )
1
(2)
Figure 1: Here the schematic of the pressure compensated flow control valve is shown. The
valve contains two orifices, a manual one that is used to set the desired flow rate, and a
second one that automatically adjusts itself to maintain that desired flow rate, even if the
pressure drop across the valve varies.
m3
s
d) Assume that the pressure at the outlet port rises to po = 8 106 [P a], with no change
3in
3 m
Am , to what value must Aa automatically change to maintain the flow rate of 6 10
s
? First, find out if px remains constant or not.
148
Chapter 11
Air cannon
When you read that two weeks would be spent on an air cannon 1 you probably wondered
why this is part of an introduction to Agricultural Engineering course. Heres why: Air
cannons are used for a variety of purposes. In agriculture, there is the concept of a puncher
planter where seeds are shot into the soil using compressed air. In addition, air cannons
are very useful for high speed safety testing for protective glazing in forestry machines, for
testing wind shields in trains and airplanes and for testing buildings from flying debris in
case of hurricanes and tornadoes. The air cannon is essentially a machine that can release
a rather large amount of energy in a very short time frame, and compared to alternative
explosive devices they are safe, inexpensive and reusable.
However, the main motivation for studying the air cannon is that students like you have
to take classes such as linear differential equations, physics, and computer science. How
many times have you seen a math professor who actually talks about the application of the
topic at hand? I would bet very few, and this is because mathematicians are proud to be
removed from practice, they always have this slight arrogance about applications being for
the blue collar crowd. The result is that most math courses only prepare you for ... more
math! Your physics courses may be somewhat more applied although often the examples
given are contrived, and the same goes for computer science classes.
To illustrate that taking these theoretical courses actually help you become an engineer,
in ABE223, we work the other way around. We have a device from which we know exactly
how it functions, we can write equations based on gas laws your learned in physics, and
we can analyze the performance using differential equations. We will collect data from an
experiment, we will then analyze the cannon and use equations that we capture in an Excel
spread sheet, and in the second week, we will solve the non-linear differential equations that
govern the process of projectiles flying through air (aka ballistics). For this you will need
MatLab skills that you (hopefully) acquired in CS101. Once we have a valid model of the
machine, we can use that model for design (for instance for a bigger cannon) and for perfor1
Cannon is an interesting word, its plural can be cannon or cannons, and it can be used as a verb. Also
think of the expression loose cannon. The worst thing that could happen to a crew in the days of sail
battleships, was a cannon that was not tied down securely to the ship. When fired, it would fly around the
cannon deck with guaranteed catastrophic results.
149
150
mance prediction.
The cannon was originally designed to be used as a testing device. In forestry, large
machines are used to harvest trees, called feller-bunchers. Some of these machines are built
on a bulldozer frame, and have a large (5 ft diameter) saw disc. These machines can fell
large trees, by driving up to them and, without even slowing down, cut the tree. Then the
machine pulls the tree in to be bunched with other trees, hence the name. Figure 11.1
shows a picture of a feller buncher.
When one of the sawteeth on the disc accidentally dislodges, it will travel through the
air with a very high speed. A sawtooth weighs about 1 kg, and it is not hard to imagine
what would happen if the operator of a machine would be struck by such a projectile. It is
obvious that operators need to be protected from this hazard. To design the safety glazing
around the operator, tests are needed where the impact process is simulated. This was the
reason to develop the cannon; it should be powerful enough to propel a 1 kg object to a very
high speed after which the object slams into a piece of safety glazing (typically laminated
polycarbonate or lexan).
It is clear that to accelerate a 1 kg object to a high speed in a very small amount of time,
it is imperative to store a lot of energy and to release it instantly. This is what the cannon
does. Lets have a look at how it works:
Figure 11.2 shows the design of the air cannon; It consists of a barrel through which
the projectile is emanated, a reservoir that holds the compressed air, and a clever piston
mechanism that is used to fire the cannon. There are only two stages in the cannon design,
charging and firing. Charging the reservoir takes place due to air leaking past the piston
which has a loose fit. Figure 11.3 shows a detail of the piston mechanism sealing off the
barrel.
Leakage of air from the reservoir into the barrel was prevented by applying a rubber
gasket on the piston. Since the pressure in the barrel is atmospheric, the piston and gasket
are pressed against the barrel with a force proportional to the applied pressure, creating
a positive seal (see Figure 11.4). Once the pressure in the reservoir has stabilized and
equalized to the applied pressure (typically 100 PSI or 700 kPa) the charging valve is closed.
The cannon is fired by rapidly dropping the pressure on the left side of the piston using
a large ball valve. This has the effect of the piston being pushed back instantly onto its seat
(see Figure 11.5). This opens the reservoir to the barrel pushing the projectile through the
barrel at very high speeds. Figure 11.6 shows the cannon being fired, notice the huge stream
of air (water vapor) emanating from the cannon at the speed of sound.
11.1
Before we dive into the analysis of our cannon, we need to review some basic gas laws. The
granddaddy of them all is the ideal gas law, which can be derived from kinetic theory (this
is a great derivation, it is quite amazing how scientist derived this one from very few simple
assumptions).
151
Figure 11.1: A feller buncher based on a dozer frame has a 5 foot diameter saw disc that
allows it to cut trees without slowing down. Imagine what happens if a saw tooth dislodges,
and flies through the forest at 150 mph. It is obvious that operators need to be protected
from this hazard using safety glazing, and to test said safety glazing we use the air cannon.
152
Figure 11.2: 3D drawing of the air cannon produced in SolidWorks. The cannon was built
from steel tubes with welded flanges and bolted together. The flanges have static seals
whereas the piston (the only part in the cannon that moves) has a dynamic seal, which
wears out over time.
Figure 11.3: Here the piston assembly is shown, the yellow part is the barrel, and the salmon
colored part is the piston, which has a rubber seal that is pushed against the barrel. Notice
that the piston fits inside a cylinder, but it does not have a very tight fit, allowing air to be
pushed into the reservoir (only indicated by solid lines) past the piston itself.
153
Figure 11.4: The cannon is charged by pushing air past the piston (which has a loose fit)
into the reservoir. Since the pressure in the barrel is atmospheric (therefore colored blue),
on the left side of the piston there is a very low pressure, but on the right side there is high
pressure. This has the effect of the piston being pushed against the barrel and closing it
off. In fact the higher the pressure differential across the piston, the better it seals: This is
called a positive seal.
Figure 11.5: The cannon is fired by dropping the pressure behind the piston (on the right
side) quickly. Since on the left side of the piston we had a high pressure in the reservoir, and
now we have a low pressure on the right, the piston is slammed to the right with a very high
speed. This opens up a direct pathway between the reservoir and the barrel. Essentially,
the air is allowed to expand instantaneously, and theoretically will reach the speed of sound
(343 m/s or 767 mph)
154
Figure 11.6: Air cannon being fired during a demonstration. The air is being pushed out
of the cannon at the speed of sound. Since this air is expanding very fast, the moisture in
the air becomes a vapor, which creates the white color. Also notice a small amount of air
coming from the area behind the piston (under the boys right hand).
155
pV = nRT
(11.1)
where p is the pressure of the gas in N m2 , V is the volume of the gas in m2 , n = number
of gas molecules in mol, R is the ideal gas constant in Jmol1 K 1 and T is the temperature
of the gas in K. It is important to realize that in a closed system (no gas escapes or enters)
the factor nR is a constant.
Furthermore,Mayers relation is a relationship where the specific heat of a gas for a
constant volume and a constant pressure are related to the ideal gas constant R.
R = cp cv
(11.2)
We already have a cannon that works fine, but imagine that you need to design a new
cannon, that is more powerful, to accelerate a larger projectile. We could simply build a
bigger one, but there are more than one parameter that we can change, including the barrel
length, and the reservoir diameter and length. To design a new cannon in a methodical
way, we need to have a model based on theory that we can use to find out what design
modification would give us the most bang for our buck. What is most important, is that this
model is valid, that means we can use measurements to verify our predictions with the model.
Intuitively, what is the most cost effective way to build a more powerful cannon?
To analyze the cannon behavior, we need to look into gas laws. Figure 11.7 shows a
simplified cylinder with a piston. There are three types of energy present here, we can have
heat exchange (either cooling or heating) with the environment dq, we have energy contained
in the gas itself (termed internal energy, which causes it to have a temperature), and we
can do work by moving the piston (this can be work on the gas (by compressing it) or by
the gas (by letting it expand). Since energy cannot be generated nor destroyed, the sum of
these energy forms must remain constant or:
dU = dq + dW
(11.4)
The reason for writing this relationship in differential form is that we can only state this
for infinitesimal changes in our variables. The work dW can be written as a force through
a distance change, we can also write it as a pressure (force/area) times a volume change
(distance * area). We can at the same time get rid of that pesky negative sign by assuming
that work done onto the piston is positive. That all leads to:
dU = dq + dW = dq pdV
(11.5)
156
Figure 11.7: Elementary gas process parameters including internal energy dU , heat exchange
with the environment dq, and work dW
1. dW = 0 = dV = 0: The volume stays constant (piston does not move, no work is
done). This is called an isochoric process.
2. dU = 0: The internal energy (temperature of the gas) stays constant, that means we
do have heat exchange with the environment, and we do work (the piston moves). This
is called an isothermal process.
3. dq = 0: There is no heat exchange with the environment, that means either that the
system is perfectly insulated, OR that the process is so fast that there simply is no
time for heat exchange with the environment. This is called an adiabatic process.
Which one of these processes do you think is most representative of the cannon?
11.2
Isochoric process
In an isochoric process there is no volume change; therefore no work is being done. We are
only left with heat exchange with the environment, and internal energy (which is related to
the temperature of the gas). Figure 11.8 shows this situation.
dV = 0 = dU = dq
(11.6)
T =T1
where cv is the heat capacity of our gas for a constant volume. It is clear that the
isochoric situation is as simple as heating up a cup of tea (or any other medium with a
constant volume). The amount of energy needed to heat it up is proportional to the number
of molecules n, mol, the temperature differential T2 T1 and the specific heat (again for
constant volume if you have a gas, in fluids we typically assume they are incompressible
anyway).
157
Figure 11.8: Isochoric process where the volume of the system is constant, and no work is
done.
Figure 11.9: Isothermal process, where the internal energy (temperature) of the gas remains
constant. This means that the heat exchange with the environment must equal the work
being done by the piston.
11.3
Isothermal process
In the isothermal process, the internal energy of the gas (that is its temperature) must
remain constant, and therefore the heat exchange with the environment must be equal to
the amount of work being done. Figure 11.9 shows this situation.
dU = 0 = dq pdV = 0 = dq = pdV
q=
(11.8)
V =V2
pdV
(11.9)
V =V1
Using the ideal gas law we can write the pressure p as a function of the volume V as
follows:
158
Figure 11.10: Adiabatic process, where no heat exchange with the environment is assumed,
either the system is perfectly insulated, or the process is so fast that there is no time for
heat exchange.
pV = nRT
(11.10)
Now we can integrate and found out the amount of energy that is needed to change the
volume from V1 to V2 : Notice that the factor nRT is a constant: since this is an isothermal
situation = T = constant.
q=
V =V2
V =V1
11.4
1
nRT dV = nRT |lnV |VV21 = nRT [ln(V2 ) ln(V1 )] = nRT ln
V
V2
V1
(11.11)
Adiabatic process
In the adiabatic process, the assumption is made that there is no heat exchange with the
enviroment. This assumption is valid in cases where the cylinder containing the gas is
perfectly and completely insulated from the environment, or where the process is so fast
that there simply is no time for heat exchange. Figure 11.10 shows this situation.
dq = 0 = dU = pdV
(11.12)
We now use the same tricks we used in the isochoric and isothermal process as follows:
dU = ncv dT
(11.13)
1
p = nR T
V
(11.14)
and
Substitution gives:
159
1
dV
(11.15)
V
Moving the temperature to the LHS and volume and cv to the RHS, while cancelling n
gives:
1
1
(11.16)
cv dT = R dV
T
V
Integrating the left side with respect to temperature and the right side with respect to
volume gives:
Z T =T2
Z
1
R V =V2 1
dT =
dV
(11.17)
cv V =V1 V
T =T1 T
ncv dT = nRT
The factor
k as follows:
leading to:
R
can also be written as using Mayers relation and the heat capacity ratio
cv
R
cp cv
cp
=
=
1 = (k 1)
cv
cv
cv
Z
T =T2
T =T1
1
dT = (k 1)
T
V =V2
V =V1
1
dV
V
(11.18)
(11.19)
After integration we get (notice that we get rid of the negative sign by flipping the ratio
V2
upside down):
V1
T2
V1
ln = (k 1)ln
(11.20)
T1
V2
(k1)
T2
V1
(11.21)
ln = ln
T1
V2
Taking the e-power of the LHS and RHS gives:
T2
=
T1
V1
V2
(k1)
(11.22)
By substituting the relationship following relationship, that results from the ideal gas
law:
p2 V2
p1 V1
=
(11.23)
T1
T2
We obtain:
T2
=
T1
V1
V2
(k1)
(k 1)
p2
k
p1
(11.24)
By multiplying the two RHS terms in this equations we obtain the characteristic adiabatic
equation:
160
p1 V1k = p2 V2k = C
(11.25)
where k is the heat capacity ratio we saw in Eqn. 11.3, and C is a constant.
With the equations as shown here we now can, given the initial volume, pressure, and
temperature, as well as the final volume (as the tennis ball leaves the cannon) calculate the
final pressure and temperature of the air. We can also calculate how much energy was in
the cannon before firing, and, assuming a 100 % transfer of the compressed air energy into
the tennis ball, predict how fast it leaves the cannon.
The amount of work that is being done while the air volume changes from V1 to V2 is the
integral of the Force through a Distance (that is the definition of work). However, we dont
directly have the force nor the distance, but since the cross sectional area of the barrel is
constant, we can also write Force integrated over distance as pressure (Force divided by an
area) integrated over a volume (which is distance times an area). Therefore the formula for
the work from V1 to V2 is:
W =
VZ=V2
p(V )dV
(11.26)
V =V1
VZ=V2
C
dV
Vk
(11.27)
V =V1
VZ=V2
C
dV = C
Vk
V =V1
V21k V11k
1k
(11.28)
p2 V2 p1 V1
1
= mp vp2
(11.30)
1k
2
where mp in kg is the mass of the tennis ball, and vp is its instantaneous speed in m/s.
Solving for the speed vp gives:
s
2 p2 V2 p1 V1
vp =
(11.31)
mp
1k
W =
161
Figure 11.11: Spreadsheet that contains the equations needed to calculate the energy output
of the cannon.)
The computer lab associated with the cannon comprises building a spreadsheet that
contains all the pertinent cannon parameters, the intial volume, temperature and pressure,
and the final volume, along with the parameters of the projectile (see Figure 11.11). Then
all equations are inserted that allow us to calculate the exit velocity of the projectile. Then,
we will pass these values to MatLab which will use them as inputs for a ballistic model that
allows us to calculate the trajectory of the projectile. Notice that, although some input
parameters may be given in English units, all parameters will be calculated in SI units, so
that in the equation we do not have conversion factors.
Laser receiver
Laser
transmitter
Cannon barrel
Air cannon as a
pneumatic safety
testing device
Upper bearing
Cover plate
Tony Grift
Swing arm
Cutting blade
Lower bearing
Agenda
Laser
receivers
Cannon
barrel
pV k Constant
Laser
transmitters
Cutting
blade
Damper
Swing arm
Miscanthus
mount
Drawings in SolidWorks
3
Charging the reservoir takes place while the piston seals the
barrel
Thermodynamics 101
pV nRT
N
J
p 2 *V m3 n mol R
*T K
m
mol * K
First law of thermodynamics: Internal energy is composed of
heat and work performed on or by the gas
Enthalpy
dU dq dW dq pdV
10
dU dq dW dq pdV
dU
dW 0
dq
J
cv
mol * K
dU dq ncv dT
dU 0
dW
dq
dq dW pdV
dU
dW
dq 0
dU dW pdV
11
pV nRT
R c p cv
dU dq dW dq pdV
cp
cv
dW pdV
dW 0
dV 0
dU
dW 0
dq
dU dq ncv dT
U ncv
9
T T2
dT nc T
v
T T1
T1
J
cv
mol * K
12
pV nRT
dU dq dW dq pdV
R c p cv
dT 0 dU cv dT 0
cp
cv
dU 0
dW pdV
dq
V V2
pdV
V V2
V V1
V V1
C
dV
Vk
dq pdV
dq nRT
q nRT
pV nRT
W C
1
dV
V
V V2
V V1
V V2
V 1 k V11 k
1
dV C 2
k
V
V V1
1 k
V2
1
dV nRT
ln V2 ln V1 nRT
ln
V
C
C
V1
p2V2 p1V1
1 k
13
dU
dU dq dW dq pdV
dq 0 dU dW pdV
dW pdV
R c p cv
cp
k
1
1
dT nR dV
cv
T
V
1
R 1
T
R V
R c c
dT
dV ln 2 ln 1 where p v k 1
T
cv V
T1 cv V2
cv
cv
V
T2
k 1 ln 1
V2
T1
ln
V
T2
ln 1
T1
V2
k 1
T2 V1
T1 V2
k 1
p
2
p1
p2V2 p1V1 1
2
m pv p
1 k
2
2 p2V2 p1V1
vp
1 k
m p
k 1
k
17
14
pV nRT
Assuming that all energy from the gas is converted into kinetic
energy of the projectile (this is a major assumption) we get:
pV nRT
dq 0
dU pdV ncv
ln
16
pV
nR constant
T
p1V1 p2V2
T1
T2
T2 V1
T1 V2
k 1
p
2
p1
k 1
k
p1V1 p2V2
T
p V V
2 2 2 1
T1
T2
T1
p1V1 V2
p1V1k p2V2 k
Adiabatic equation
k 1
p1V1k p2V2 k
15
18
Your assignment
Think of everything that could go wrong
Make a safety procedure sheet for the test cannon before and after we fire it
19
20
21
166
Chapter 12
Ballistics
In the previous chapter, we determined how much energy was contained in the compressed
air in the cannon. We did this based on an assumption (a very reasonable one at that),
being that the sudden expansion of air must comprise an adiabatic expansion process. We
determine the volume of air before firing and after, and that gave us a number for the total
energy contained in the air. Then we assumed that all the energy that was contained in the
air was somehow transferred into the projectile (this is quite an assumption by the way, it
needs to be validated experimentally). Here are the equations repeated for the amount of
work that is being done, while the projectile accelerates through the barrel, and the velocity
the projectile would attain if there were 100 % conversion of the energy from the compressed
air into the kinetic energy of the projectile.
1
p2 V2 p1 V1
= mp vp2
(12.1)
1k
2
where mp in kg is the mass of the tennis ball, and vp is its instantaneous speed in m/s.
Solving for the speed vp gives:
s
2 p2 V2 p1 V1
vp =
(12.2)
mp
1k
W =
In the chapter of ballistics, we will find out what the trajectory of the projectile (in our
case a tennis ball) would be. What we know is the initial velocity (again based on the assumption of 100 % energy transfer from air to projectile), we have the mass and the diameter
of the tennis ball. We also need to know the initial location, and the angles of the cannon.
A cannon has two angles being the elevation (angle in a vertical plane) and the azimuth,
the angle in a horizontal plane. The latter is not relevant for us, we simply assume that the
projectile will stay within the vertical plane without any lateral movement.
Before we can determine the trajectory, we need to know how the forces on the tennis
ball while traveling in air affect it. For this we need to draw a diagram as shown in Figure
12.1.
If we assume that the tennis ball is traveling upward along the vector v as shown, there
is a aerodynamic drag force FD in N acting on it in the exact opposite direction of travel.
167
168
(12.3)
(12.4)
(12.5)
(12.6)
According to Newtons law, the net sum of the forces must result in the inertia force which
is equal to the mass of the tennis ball times the acceleration. Also since these equations are
starting to look like differential equations, we typically define in the x-direction that x =
distance, x = velocity and x is the acceleration. Putting all that in gives:
m
x=
or
CD A p 2
x x + y 2
2
p
x = K x x 2 + y 2
(12.7)
(12.8)
169
Figure 12.2: This diagram shows how the drag coefficient CD of a sphere is related to
Reynolds number.
CD A
. The drag coefficient CD (and therefore K) is a function of Reynolds
2m
number (see Figure 12.2), but for simplicity, we assume that the drag coefficient is equal to
0.4 (meaning we assume a turbulent flow regime that starts at a Reynolds number of 2,000).
There are better approximations, as shown in the discussion under section 12.2.
where K =
The exact derivation as above for the x-direction can be made for the y-direction with
one exception: In the y-direction we also have a gravity force pulling the projectile down to
earth. Therefore the equation for the y-direction becomes:
y = K y
p
x 2 + y 2 g
(12.9)
Putting them both together we obtain a set of coupled non-linear ordinary differential
equations as follows:
p
x = K x x 2 + y 2
p
y = K y x 2 + y 2 g
(12.10)
(12.11)
These equations are coupled since we have an x in the y equation and vice versa. The
square root function makes them both non-linear. These kinds of equations typically cannot
be solved analytically, therefore we need to resort to numerical methods. These methods are
170
Figure 12.3: To simulate this set of ballistics equations, MatLab requires that you present
them in a standardized form, as a set of first order equations.
based on Eulers method but typically Runge-Kutta methods are implemented in software
such as MatLab.
For MatLab to solve these differential equations, they need to be presented in a set of
first order equations in the form X = f (X, t) where X is a vector of state variables and f is
a (possibly nonlinear) function. We can do this by defining the following:
x1
x2
x3
x4
=x
= x 1 = x
=y
= x 3 = y
(12.12)
(12.13)
(12.14)
(12.15)
Now we have to present these equations as a set of first order equations as follows:
x 1 = x2
x 2 = x1 = x = K x2
x 3 = x4
(12.16)
x22 + x24
q
x 4 = x3 = y = K x4 x22 + x24 g
(12.17)
(12.18)
(12.19)
Figure 12.3 shows the MatLab file that contains this code, compare this with the equations
given here.
In the computer lab you will write a program that allows you to calculate the trajectory
of the projectile, the output is shown in Figure 12.4.
The MatLab simulation uses Runge Kutta method to solve the non-linear differential
equations. You have to watch this one hour lecture where MIT professor emeritus Arthur
Mattuck explains how this method works (Figure 12.5).
171
Figure 12.4: The MatLab program that you will develop in the lab shows the simulation
output.
172
Figure 12.5: To understand how computers are used to solve even non-linear differential
equations, you need to watch this video by MIT professor emeritus Arthur Mattuck.
12.1
Now we have the generic equations for ballistic flight with aerodynamic drag, lets see what
happens in the case of no aerodynamic friction, which simply means that the drag coefficient
CD is assumed equal to zero. Now the equations become very simple:
x = 0
y = g
(12.20)
(12.21)
If we integrate these equations to obtain equations for the speed in the x, y direction we
get:
x = x(0)
y = gt + y(0)
(12.22)
(12.23)
Where x(0)
and y(0)
+ x(0)
1
+ y(0)
y = gt2 + y(0)t
2
(12.24)
(12.25)
173
Lets assume that the cannon represents the origin, so x(0) = 0; y(0) = 0 which simplifies
to:
x = x(0)t
y = gt2 + y(0)t
2
(12.26)
(12.27)
Lets find out what elevation angle of the cannon we need to shoot our projectile the
farthest. When the projectile lands its y coordinate has to be equal to zero, lets find out
what time is associated with that event.
1 2
1
gt + y(0)t
= 0 t gt + y(0)
=0
(12.28)
2
2
It is clear that t = 0 is a solution, this is when the projectile leaves the cannon. Another
solution is:
t=
2y(0)
(12.29)
Substituting this back into the equation for the displacement in the x-direction (Eqn.
12.22) gives:
x = x(0)t
= x(0)
2y(0)
(12.30)
Note that after substituting a constant value for the time t, x is no longer a function of
time. As mentioned, the initial velocities in the x, y directions are related to the elevation
angle as follows:
x(0)
= V cos
y(0)
= V sin
(12.31)
(12.32)
Where V is the constant velocity in ms1 at which the projectile leaves the cannon. Now
we can obtain the relationship between the displacement in the x-direction and the elevation
angle by substitution:
x() = x(0)
2V sin
2V 2
2y(0)
= V cos
=
cos sin
g
g
g
(12.33)
(12.34)
174
(12.35)
yielding:
(12.36)
4
So it is clear that the farthest distance can be achieved when the elevation angle is 45
degrees and the total distance is .
=
2V 2
V2
cos sin =
x( ) =
4
g
4
4
g
(12.37)
However, this is without aerodynamic friction, which is not realistic. This is why we
need to solve the set of differential equations to find out what this value is for the real life
situation where friction is present.
12.2
To illustrate the importance of a proper model for the behavior of a particle moving in air,
here we address the problem of granular fertilizer application. The method of ground based
fertilizer application mimics a human arm swing, by dropping granular fertilizer particles on
a disc with vanes. The design of these kinds of spreaders has been largely empirical, since
modeling the flow of particles through a gate, their (largely interactive) acceleration along a
vane, and their travel through air to a landing spot are quite difficult. However, to optimize
the process, and to improve upon existing designs, there is a need for models of all these
subprocesses.
One of the most complicating factors in the process is the material itself. In classical mechanics, the particle is always assumed spherical, which is fine because it allows researchers
from all over the world to repeat experiments. This is where agricultural engineering is
different from classical engineering. We always have to deal with non-ideal situations that
are very hard to model, and we need to be creative to translate a highly variable (stochastic) form into a manageable deterministic one. A good example is the behavior of fertilizer
particles through air, after being emanated from a disc.
Fertilizer particles are not spherical. Some forms are close to spherical (Ammonium
Nitrate and Urea for example) but others look as if they are mined from a rock (such
as Potassium Chloride, which it is). To determine how fertilizer particles behave while
traveling through air compared to spherical particles a fall test was devised. Here particles
were dropped from a height of 15.83 m, and the fall times recorded. Figure 12.6 shows the
top and bottom parts of the fall tower.
A model was used to calculate the theoretical fall times for spheres, and the arrangement
was tested by dropping spheres with density close to that of common fertilizers. When a
Figure 12.6: These photos show the top and bottom of the fall tower. A carousel was used
to insert up to 100 particles at once, and the machine slowly rotated until all particles had
been dropped. The timer was started by an optical sensor just below the carousel disc, and
it was stopped when the particle hit a metal plate that was fitted with a microphone. The
total height of the fall tower was 15.83m, and it was lined with heavy plastic to prevent
external airflow influencing the measurements.
176
particle falls in still air, there are three forces at work. Firstly a gravity force Fg is pulling
the particle down, while a friction force with the air (drag force FD ) resists that motion.
The sum of these forces results in an acceleration according to Newtons law with an inertia
force FI . In formula form:
Fg FD FI = 0
(12.38)
1
If we call the downward direction y, since Fg = mg, FD = CD A air y 2 , and FI = m
y.
2
Substitution gives:
1
y=0
(12.39)
mg CD A air y 2 m
2
where m in kg is the mass of the particle, CD is the drag coefficient of the sphere,
A is the projected area of the sphere, air is the density of air in kgm3 , and y and y
are the
acceleration and velocity of the particle respectively. By defining
instantaneous
1 CD Aair
K=
we obtain:
2
m
y + K y 2 = g
(12.40)
This equation obviously contains the acceleration of a particle, but there is a special name
for the limit situation where the particles acceleration is zero and the velocity
r constant: This
g
.
is called terminal velocity, and from the equation it is equal to yterm =
K
From your coursework in differential equations, I hope you recognize that equation 12.40
is a non-linear ordinary differential equation due to the squared function of y.
There is
an analytical solution for this kind of equation (technically, it is called a Riccati equation)
but this is an advanced math topic. Another problem is that the drag coefficient CD is not
constant, but it depends on Reynolds number which contains the speed of the particle. This
is another reason why this problem needs to be solved numerically.
An analytical solution to equation 12.40 was found using MatLabs Symbolic Solver
Toolbox (which was called the Maple kernel (from Canada) at the time). Nowadays, I would
probably have used Mathematica for this purpose. The solution was:
p
io
1 n h
y(t) =
ln cosh
gK(t C1 )
+ C2
(12.41)
K
Putting in the boundary conditions of y(0) = y0 and y(0)
( "
ln cosh arctanh
K
vo
g
!!#)
(12.43)
t=
1
arccosh eKy
gK
(12.45)
Now I could split up the total fall distance into a large number (say N = 100, 000) of
small distances, calculate the K value for each distance, and use equation 12.45 to calculate
ymax y(0)
the time it would take for a particle to fall through this small distance yi =
.
N
The total fall time tend would then simply be the sum of all these:
tend =
N
X
i=1
1
arccosh eKi yi
gKi
(12.46)
As you can see from the table in 12.7 the simulated fall times were very close to the
measured fall times for spheres. It was essential that we calibrated the complete system, to
evaluate the level of error before dropping actual fertilizer particles.
Figure 12.8 shows the fall data for Calcium Ammonia Nitrate (CAN) fertilizer, which is
rather spherical. Note that all fertilizer particles take longer to fall down the 15.83 m tube
than spherical particles (simulated in the blue line). There are also many outliers, which
are particles that did not drop down straight, but bounced off the plastic walls of the fall tube.
Now that we had established a relationship between fertilizer particles and spheres, it
was possible to produce correction factors to be used in ballistic models. The idea would be
that for a given fertilizer in a given diameter range, the fertilizer particle would be treated
as a larger sphere to compensate for the drag coefficient being higher. For example, a CAN
fertilizer particle with a diameter of 4mm would have the same fall time as a spherical particle of 3.5mm, therefore the correction factor would be 0.875.
In summary, this research is typical in agricultural engineering: We try to come up with
a translation of materials and behaviors in our world to well defined shapes that can be
modeled using theory. For the complete story see [1] or download the paper here.
178
Figure 12.7: This table shows how good the simulation results were by comparing the predictions for spheres with data from real spheres made from the plastics PTFE, POM and
PP.
Figure 12.8: Here the total fall times are shown for CAN fertilizer (which is rather spherical).
It is clear that fertilizer particles have a longer fall time (thus a higher drag coefficient) than
spheres (blue continuous line). For comparison, if the particles fell in a vacuum, the time it
would take is 1.8 seconds.
Agenda
Derivation of dynamics in equation form:
What happens to a flying projectile in air?
Solving non-linear equations in MatLab
Assuming that all energy from the gas is converted into kinetic
energy of the projectile (this is a major assumption) we get:
p2V2 p1V1 1
2
m pv p
1 k
2
2 p2V2 p1V1
vp
1 k
m p
The drag coefficient (Cd) value for a sphere is about 0.4 for
turbulent flow regimes (Re>2000)
vy
FDx
vx
Fg
F
FD
FDy
F
1
FD Cd * A * v 2
2
0 .4
vy
vy
FDx
vx
FDx
vx
Fg
Fg
FD
v
Fy FD *sin mg Cd * A * v 2 * y mg
2
FDy
1
1 F
Fx FD * cos Cd * A * v 2 * x Cd * A * vx v
2
2
Fx mx Cd * A * vx v
2
Fy my Cd * A * v y v mg
2
Cd * A *
2
Fy y
vyv g
m
vx x
Cd * A *
2
Fx x
vx v
m
2
2
v vx v y x 2 y 2
FD
FDy
F
v y y
2
2
v vx v y x 2 y 2
y Ky x 2 y 2 g
x Kx x 2 y 2
7
x Kx x 2 y 2
y Ky x 2 y 2 g
y Ky x 2 y 2 g
y Ky 2 g 0
g
K
1
Cd A
K2
m
y
10
x Kx x 2 y 2
function xp = aerodyn(t_sim,x_state)
y Ky x 2 y 2 g
global g K
x1 x
x2 x1 x
Define
xp = [x_state(2);
-K*x_state(2)*sqrt(x_state(2)^2+x_state(4)^2);...
x_state(4);
-K*x_state(4)*sqrt(x_state(2)^2+x_state(4)^2)-g
];
x3 y
x4 y1 y
Result
x1 x2
x1 x2
2
2
x2 Kx2 x2 x4
2
2
x2 Kx2 x2 x4
x3 x4
x3 x4
x4 Kx4 x2 x4 g
2
11
2
2
x4 Kx4 x2 x4 g
12
Simulation can yield a graph that relates the exit velocity to the air
time
14
12
10
0
0
50
13
150
200
Exit velocity in m/s
250
300
350
14
Lab
Outdoor part
180
160
Indoor part
140
Height achieved in m
100
120
100
80
60
40
20
0
50
100
150
200
Exit velocity in m/s
250
300
350
15
17
16
18
http://www.youtube.com/watch?v=LbKKzMag5Rc
19
20
184
Chapter 13
Engine performance
From an abstract point of view, an internal combustion engine is a device that converts
chemical energy (from fuel) into mechanical energy. If we divide both input and output side
by time, we can also claim that an engine converts fuel power into mechanical (rotary) power
(Figure 13.1). We can also then define the overall engine efficiency as the ratio between output power 1 and fuel power.
The first lab regarding engines in ABE223 comprises a test where we measure the output
power generated by the engine, as well as the input power flowing into the engine. We
already have the relationship for mechanical rotary power which was:
rad
Nm
=
T [N m]
(13.1)
Pmech,rot
s
s
The radian is an unitless angle, therefore
this equation is dimensionally homogeneous.
Nm
Also, remember that another name for the
is Watt and another name for [N m] is
s
J
J
Nm
=
= Watt. The fuel power into the engine
is equivJoule, therefore a
s
s
s
J
alent to the energy density of the fuel (aka the heating value) in
times the mass flow
g
hg i
rate of the fuel into the engine in
. As a reference, #2 (summer) diesel has a heating
s
MJ
value of 42
.
kg
13.1
As you know by now, power is the product of torque in [N m] and the rotational velocity
rad
of the engine shaft in
. Given an output power requirement, engine designers have
s
1
The name that still persists for output power is brake power, which reverts back to the days when the
engine output power was measured with a genuine brake mechanism.
185
186
Figure 13.1: An abstract model of an engine defines it as a device that converts fuel power into
mechanical (brake) power. The ratio between input (fuel) power and output (brake) power
is the overall engine efficiency. The lost power is composed of a mechanical loss component
(friction) which is expelled through the engines cooling system, as well as a thermal loss
component, which is expelled through its exhaust system. Even after a century of engine
development, the highest overall efficiency achievable is about 43 % for a compression ignition
(diesel) engine with a turbocharger. Spark ignition (gasoline) engines typically have overall
efficiencies up to 33 %.
187
Figure 13.2: Ferrari 12 cylinder, normally aspirated Formula 1 engine rated at 700 HP at
18,000 RPM with a maximum torque of 271 Nm.
the freedom to produce an engine with high torque or high rotational velocity (lets call it
RPM, since that is the most common term used). These engines will look very different
though. Low RPM engines such as ships engines are high torque engines since they drive
large low-speed propellers. High-torque engines last longer and are more efficient than high
RPM engines, but they are much heavier because of the high torque, which requires heavier
shafts, bearings, pistons etc. Youll see these engines where weight is not a problem (such
as in stationary application), and where durability and efficiency are paramount such as a
mining trucks, off-road equipment and ships. You see high RPM engines where low weight is
premium, such as in small aircraft, passenger cars and in the extreme in racing applications.
For instance, have a look at the Ferrari Formula 1 engine in Figure 13.2.
To achieve the 700 HP from a lightweight engine that would easily fit on a coffee table,
they had to get the engine speed very high, but with a piston speed limited by potential
lubrication breakdown, the torque would be relatively low. Therefore these engines have
pistons that have large bores (diameters), but short strokes. If the bore is equal to the
stroke, this is called a square piston arrangement, if the bore is larger than the stroke it
is called over square, but in the case of a formula 1 engine the bore is much larger than the
stroke, therefore these are called supersquare piston arrangements (see Fig. 13.3).
Another problem with these very high revving engines is that the valves at some point can
no longer follow the camshaft lobes, they will start floating. This is a highly undesirable
phenomenon, because when the valves no longer completely open and close, the engine
efficiency decreases, there is a chance of the pistons actually hitting the partially closed
valves, in addition to exhaust valves becoming so hot that they disintegrate and fail. For
this reason, Formula 1 engines do not use springs as support for the valves, but they use
pneumatic (air) springs (see Fig 13.4).
188
Figure 13.3: The piston in a Formula 1 car is what is called supersquare, since the bore (98
mm) is much larger than the stroke (40 mm). Notice that the piston skirt is very short,
it just supports the compression ring and an oil distributor ring. The overall aim is to keep
the piston and connecting rod as light weight as possible, but they have to be made from
steel and aluminum, exotic materials such as magnesium and beryllium are not allowed.
Figure 13.4: At extreme rotational velocity (RPMs) of an engine, the classical metal spring
return mechanism no longer works. This is why Formula 1 engines that can do up to 20,000
RPM, the valve is returned using compressed air. At 20,000 RPM, this valve opens and
20, 000
= 166 times per second, a major engineering feat.
closes
2 60
189
Figure 13.5: Volvo Penta D16 engine rated at approx. 700 HP at 1,850 RPM, with a
maximum torque of 3,265 Nm. This is a large 16 liter displacement diesel engine weighing
in at 2,110 kg. These engines are mainly used in marine applications.
13.2
Engine terminology
Engines have always had quite a few components, and modern engines only add to that
because of strict emission control hardware. Figure 13.6 shows a Spark Ignition engine with
a single camshaft that is mounted in the cylinder head. This configuration is called Single
Over Head Camshaft. Well see what the camshaft does in a minute.
First lets go through this engine in alphabetical order.
(a) Intake Valve, Rocker Arm & Spring. Since we have only a single camshaft, and
the valves are mounted at an angle with respect to the engine centerline, rocker arms
are used that are pushed up by the camshaft, and then in return push down the valves.
The left valve itself is the intake valve, which are typically smaller than the exhaust
valves. Between the intake port and the valve there is usually a hardened seat, a
ring that is pressed into the cylinder head, which is typically made from aluminum.
Older engines relied on lead in the fuel to seal the valve to the port, but since lead
has been outlawed in fuel, newer engines have hardened seats. The spring returns the
valve after the lobe on the camshaft has passed, typically there is a inner spring and
an outer spring. These springs very slowly rotate to distribute wear of the valve and
springs.
Ideally there should not be any play between the lobes of the camshaft and the valve
stem, but in reality when the engine warms up to operating temperature, the valves
(especially the exhaust valve which gets really hot) expand. Therefore a small clearance
is built in between the rocker arm and the valve, which can be adjusted such that at
ideal operating temperatures, there is no slack or play. Modern engines no longer have
this adjustment, instead they use hydraulic lifters, these are hollow cylinders which are
pressurized by engine oil, effectively eliminating valve play at all times.
190
Figure 13.6: Terminology of a Single Over Head Camshaft (SOHC) Spark Ignition engine.
191
(b) Valve cover. The valve covers role is to protect the valve train from external debris
and at the same time, keep oil inside the engine. It has a gasket that often leaks
in older engines, it can be easily replaced even with a home-made gasket using high
temperature silicone.
(c) Intake port. The intake ports role is to allow the fuel-air mixture into the cylinder
(remember this is an SI engine, so it draws in the correct ratio of air and fuel). The
fuel can be supplied by a carburetter ,but modern cars have computer controlled fuel
injection.
(d) Cylinder head. The cylinder head is a complex part, since it contains the valves,
that run in valve guides and rest on seats, it has bearings for the camshaft, contains
the sparkplugs, but also has channels for coolant, in addition to oil passages. Modern
cylinder heads are cast from high-strength aluminum to save weight, but highly stressed
parts are typically inserted such as valve guides, valve seats, camshaft bearings and
threaded inserts (helicoils) for the sparkplugs.
(e) Coolant. To keep the engine cool, typically a jacket is surrounding the cylinders and
passages are drilled into the cylinder head, since it contains the combustion chambers
that become extremely hot. The coolant is typically a mixture of glycol alcohol and
distilled water. This mixture gives the best protection to freezing up to -35F. Never
use pure water (even distilled water) in an engine, since during freezing
conditions the water will expand and crack the cylinder mantle and/or the
head.
(f) Engine block. The engine block in modern engines is made from high-strength aluminum to save weight. However, the cylinders need walls that let oil adhere well,
therefore these often have sleeves made from cast-iron. Also, during production, these
sleeves are honed to a cross-hatch pattern which lets oil adhere even better.
(g) Oil pan. The oil pan contains the lubricating oil for the engine, typically the oil pump
is also mounted here.
(h) Oil sump. Shown here is a wet-sump system where the oil simply drains back to the
lowest point in the engine. However in high g-load applications such as airplanes and
race cars/motorcycles, a separate tank is used, a dry sump system. This is to avoid the
oil slushing around in the sump and starving the engine of oil under hard cornering.
Dry sump system use a scavenger pump to push oil from the sump into the tank, and
a separate oil pump to push oil into the engine.
(i) Cam shaft. The sole purpose of the camshaft is to open and close the intake and
exhaust valves at the right moment during the cycle of the engine. In a four stroke
(cycle) engine, it takes two revolutions of the crankshaft to complete the four strokes.
During these two crankshaft rotations, the intake and exhaust valve open only once.
Therefore the camshaft must turn at exactly half the rotational speed of the crankshaft.
This is most often done using a timing belt, although some cars have timing chains or
gears. Most modern engines are interference engines, meaning that if the connection
192
(j) Exhaust valve, Rocker arm and Spring. The purpose of the exhaust valve is
to expel the combustion gases from the cylinder. After the fuel-air mixture has been
ignited, its volume increases dramatically (this is what pushes the piston down), this
is why the exhaust valve (or valves, since most modern engines have four valves per
cylinder) is much larger than the intake valve(s). It also passes very hot exhaust gases,
so it becomes much hotter than the intake valve which only passes a cool fuel air
mixture.
(k) Spark plug. The spark plus, as the name suggests, is the device that delivers the
spark that initiates the combustion process of the fuel-air mixture. In Spark Ignition
(SI) engines, in contrast to Compression Ignition (CI, aka diesel) engines (remember
these acronyms) the compression does not cause a high enough pressure for the fuel to
combust spontaneously. In fact, self combustion in SI engines is highly undesired (it
is called pinging), in fact over time this can destroy and engine. The fuels octane
rating number (such as 95 or regular) up to 98 in super) tells you how resistant
the fuel is to self combustion. Most engines run fine on regular fuel, but if your engine
has a high compression ratio (racing engines), or if it has a supercharger compressor
or turbocharger, you need to use a higher octane number to prevent pinging. Putting
a high octane fuel in a typical normally aspirated engine is a waste of money.
(l) Exhaust port. The exhaust port allows the spent gases to leave the cylinder.
(m) Piston. The piston moves up and down (reciprocates) in the cylinder rapidly (for
the fun of it, calculate how many times per second if an engine runs at 6000 RPM).
A typical piston has compression rings to seal the combustion chamber, and at the
bottom of the skirt, an oil scraper ring that maintains a constant film of oil on the
cylinder walls. Modern pistons are made from high-strength and high temperature
resistant aluminum. Inside the piston is a bearing that allows the wrist action of the
piston, and inside the bearing is the wrist pin. In larger engines this typically (not
always) is a well lubricated sliding bearing, in smaller engines youll find needle sleeve
bearings.
(n) Connecting rod. The connecting rod (or con-rod in nascar parlance) connects the
piston to the crankshaft. At the piston side (small end) it contains the wrist pin
bearing and at the crankshaft side (big end) it contains a much larger sliding bearing.
When engines fail, often the connecting rod breaks and pushes right through the engine
block, this is what is meant when you hear a nascar commentator say the engine just
threw a rod.
(o) Connecting rod bearing. The connecting rod is attached to the crankshaft through
a large sliding bearing. This bearing is made from a soft metal that has a large area,
so that the pressure is relatively low. An oil film makes sure that there never is any
193
13.3
Engine power
The power flowing into an internal combustion engine is called fuel power, and it can be
calculated by multiplying the fuel flow rate into the engine (in kgs1 ) and the fuel heating
J
kg J
194
pressure (in N m2 ) or Pascal is given versus the volume (in m3 ) of the cylinder (which of
course varies since the piston moves up and down). When the piston is at the bottom position this is called either Bottom-Dead-Center (BDC) and sometimes Crank-Dead-Center
(CDC) since it is the closest the cylinder will get to the crank shaft. The highest point in
the cylinder reached by the piston is called the Top-Dead-Center or TDC. Lets have a look
at the pressure in the cylinder during the four strokes.
During the intake stroke, the intake valve opens and there is a slight vacuum in the cylinder which draws the fuel-air mixture into the cylinder (this is assuming the engine does not
have a turbo-charger or a super-charger, in other words it is naturally aspirated). This is
shown at the bottom of the pV diagram where the piston moves from the TDC to the BDC.
During compression, the fuel-air mixture is compressed where the pressure and temperature
increase adiabatically, as shown in the exponential line where the piston moves from the
BDC to the TDC. At this point the fuel-air mixture is fully contained in the combustion
chamber.
This is the point where ignition takes place. The spark ignites the fuel-air mixture, which
gives an instantaneous increase in the temperature and pressure of the burning gases. This
process is so fast that we assume that the piston does not even move, which is shown by the
vertical line at the TDC of the cylinder (left).
After ignition the gases expand violently pushing the piston downward with very high
force. This is the power (work) stroke. Remember that this is the only stroke where engine
power is produced.
Finally after the piston has reached the BDC, the exhaust valve opens, and the burnt
gases are expelled while the piston moves back up to the TDC. Now the whole process begins
again.
Firstly, you notice that the intake and exhaust are necessary, but essentially parasitic
strokes. However, they take very little energy, so we typically ignore them in power calculations. The only thing we have to do is remember that the cylinder only has a net energy
output during compression (negative energy) and power (positive energy), but it takes two
revolutions of the crankshaft for a full four cycles, so in the end we need to divide by two.
Secondly, during compression, the piston has to do the work to compress the fuel-air
mixture into the combustion chamber. Since we count energy produced BY the engine as
positive, this is in fact negative energy. In the power stroke, we do get positive energy out of
the cylinder, but to calculate the net output energy, we need to subtract the energy we spent
in compressing the fuel-air mixture. The pressure that is present during the combustion
cycle is shown in a pV diagram (Fig. 13.8).
To calculate how much power an engine can produce, we need to find out how much
power a single cylinder can produce, and simply multiply by the number of cylinders that
the engine has (which can range from 1 in motorcycles to about 12 in cars, and even higher
in ships and vintage airplanes). To figure out how much power a cylinder produces, we
195
196
average forces, we simply multiply by the projected area of the piston. Then, to translate
these into average work, we multiply the forces by the stroke of the piston.
FC = pC A
C = pC A s
W
(13.2)
(13.3)
Now we have average work, we need to calculate the power produced in the cylinder. We
do that by dividing the average work by the time it takes the cylinder to do not one but
two revolutions. Why was this again? Because the power that we produce in a four stroke
engine requires two revolutions, since we have the parasitic intake and exhaust strokes. If
1
2
rev
then it takes
second to do a revolution and
seconds
an engine is running at say
s
for two revolutions. If we divide the average work by this factor we get:
pime A s
(13.4)
2
To get the power of the engine, we have to multiply by the number of cylinders N , and
by the mechanical efficiency (M ech ). This gives the complete power equation for the engine
as:
P owCyl =
pime
AsN
M ech
P owEngine =
2
(13.5)
Agenda
Power measurement
Dynamometers, waterbrake, eddy current
Torque and Power curve
Power computation
Pressure-Volume curves
Indicated Mean Effective Pressure
Power computation in cylinder, engine
Power
Measurement
Tony Grift
Power Measurement
PTO dynamometer
Dynos
Mechanical (friction)
Electrical (generator)
Hydraulic (water,
eddy current)
P T
Engine
Torque
Engine
Speed
Exercise
Compute the Torque of the following vehicles
136
HP @ 6000 RPM
500
100,000
800
HP @ 7000 RPM
HP @ 100 RPM
HP @ 18,000 RPM
rad
Pow
* T Nm
s
Nm
1HP 745Watt 745
s
1
min
1 rev 60 s
s
Power Computation
F = p* A
Work
Power Measurement
F = p* A
W= F*L
Dynamometer
Power
in fuel
ip
Indicated
thermal
efficiency
bp
Mechanical
efficiency
Rejected
Heat
Friction
Power
Compression stroke
FC p C * A
WC pC * A * s
p
TDC
BDC
Power stroke
W WP WC
WP p P * A * s
TDC
TDC
BDC
W WP WC p P * A * s pC * A * s p P pC * A * s pime * A * s
pime
BDC
RPM rev
60
s
Nm pime * A * s * * N
* MECH
PENGINE
2
s
pime Indicated Mean Effective Pressure during Compression and Power strokes
p * A * s *
W
Nm W
PENGINE
W * ime
* MECH
2
s t 1
A
s
Homework
The indicated mean effective pressure measured in the cylinder is
10 Bar (that is or ). The engine is a V8 four-stroke. The engine is
running at 3600 and the mechanical efficiency is 90%.
100 mm
120 mm
Required:
1) Compute the average Force acting on the piston during the
compression/work strokes.
3) Compute the work done in 1 second for the whole engine, which is the
Power
100 mm
120 mm
Required:
1) Compute the average Force acting on the piston during the compression/work strokes.
2) Compute the Work being done in one cylinder during compression/work strokes.
3) Compute the work done in 1 second for the whole engine, which is the Power
55
Units (SI)
Pressure
Pascal
Flow
Newton
Joule
Watt
Engine
Performance
Area
Force
Work
Power
Tony Grift
Torque
Dept. of Agricultural & Biological Engineering
University of Illinois
N
p 2
m
m3
q
s
A m2
F N
W Nm
Nm
Pow
s
TNm
Force is in Newton
Work is Force through a distance = Newton meter (Joule)
Power is Work per unit of time = Newton meter / second (Watt)
F
W F *D
Pow
N
Nm
W F * D Nm
s
t
t
lb f ft
s
550
lb f ft 0.3048 m 4.448 N
*
*
1 lb f
1 ft
s
Nm
or W
745.7
s
Force
Work (Energy)
F m*a
W FD Nm
The unit for work is also called a Joule.
1 Nm 1 Joule
N kg ms
Linear power
Torque
In case of a motor shaft, the Work can be found by multiplying a
force through a distance. Suppose we assume a force at a
distance r. The total work per radian of the shaft is now equal to
PLINEAR
W Nm
t s
WMECH F * r
PLINEAR
FD Nm
t s
1 Radian
Nm
The factor t
Rotary power
The power is now equal to this value divided by the time per
radian.
rad
1
If the shaft is turning at
seconds per radian.
it takes
s
Since for Power we have to divide the Work by time, this leads
to:
W F *r
PowMECH
*F
*r
t
1
T
PowMECH * T
Nm rad
s s * Nm
Linear (Drawbar)
m
PLINEAR F N v Watt
s
Rotary (PTO)
rad
PTOP
T Nm Watt
s
Hydraulic
3
N m
Hyd p 2 q Watt
m s
Electrical (alternator)
10
Rule of thumb..
The rule to remember is that in worst case, 2WD, soft field the
total usable power at the drawbar is approx. 50 % of the rated
power at the PTO.
11
12
No lugs in front
(just for steering)
Large wheels,
tires with lugs
13
Articulation
14
16
Engine Implementations
17
18
Four stroke engines are heavy, relatively low power, but with
proper appendages can have decent emissions
20
Ignition of the fuel-air mixture takes place just before Top Dead
Center, which pushes the piston downward
22
During the exhaust stroke exhaust gases are pushed out of the
cylinder by opening the exhaust valve
24
25
During the intake stroke air (not fuel) is drawn into the
cylinder by opening the intake valve
26
28
30
Two stroke engines are light weight and relatively high power but
have poor economy, poor durability and very poor emission
quality. They also run rough at idle due to gas mixing with exhaust
32
After the combustion (work, power) stroke is completed, fresh gases move from
the sump to the combustion chamber: The engine is flushed with fresh fuel that
mixes with combusted fuel (exhaust), which is a really bad idea.
35
34
Diesel two stroke engines flush the cylinder with air, not fuel/air. This is a great
principle, with twice as many powerstrokes as a four stroke engine
36
37
38
39
40
41
42
43
44
Honda GL 1500
Flat 6 SOHC
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
Perkins (Caterpillar)
62
61
Perkins 404C-22T
(44.7 kW @ 2800 rpm)
63
Perkins 403C-11
(19.5 kW @3000 rpm)
64
Naval applications
65
66
11
67
68
Lets listen
Ships engines
PistonRemovalMAK.mov (QuickTime)
70
69
Elly Maersk and her sister ships are the largest container
ship in the world. Image the engine in this vessel
71
72
12
Crankshaft Bearings
Crankshaft
74
Cylinder sleeves
75
76
Wartsilla Engine
Horse Power TV 26:11 30:00
77
MAN
MAN
MAN
MAN
MAN
2
2
2
2
2
stroke
stroke
stroke
stroke
stroke
1
2
3
4
5
(6:18)
(4:28)
(4:20)
(3:00)
(6:45)
78
13
3400 HP
360 Tons capacity
40 MPH Fully loaded
Mechanical drive
Mining Trucks
http://hutnyak.com/HaulTrucks.htm
79
80
81
82
83
84
14
85
86
87
88
89
90
15
Military
91
92
93
94
95
96
http://www.geocities.com/Pentagon/Quarters/4980/engine9.html
16
Race engines
97
98
99
100
F1 Engine running
Senna Qualifying Suzuka
101
102
17
103
104
105
106
107
108
18
109
110
111
112
113
114
19
115
116
117
118
119
120
20
121
122
123
124
Exotics
125
126
21
127
Wankel motorcycles
Advantages
Hercules W2000
Staubsauger
Slower
Since the rotors spin at one-third the speed of the output shaft, the main moving
parts of the engine move slower than the parts in a piston engine.
128
Disadvantages
129
NSU RO80
130
Mazda
RX7 / RX8
http://www.mazdarx7engines.com/
131
132
22
133
134
Lets listen..
135
136
Transmissions
137
138
23
139
140
141
142
143
144
24
145
146
147
148
25
Chapter 14
Granular fertilizer application
One of the most influential factors in the grain yield increase accomplished over the past 150
years was the application of artificial ammonia-based fertilizer, through a process invented
by Fritz Haber, commonly known as the Haber-Bosch process. The historical yields of corn
in the US remained stable from the civil war until 1936 at around 26 bu/ac. After the
Dust Bowl of the 1930s, farmers adopted hybrid seed which led to an annual increase of
0.8 bu/ac/yr in the period from 1937 1955. Following this period a constant increase of
1.9 bu/ac/yr was established, in part due to mechanization, and improved agronomics and
genetics. In 2010, a record yield was set at over 160 bu/ac, passing the magical metric mark
of 10,000 kg/ha or 1 kg/m2 . However, the Great Drought of 2012 showed that natural causes
can have a devastating effect on maize yield, pushing the average US yield down to a mere
123 bu/ac.
Even though Haber was awarded the Nobel Prize in 1918, his true intentions were far from
noble: Not only was his process of producing artificial ammonia from air targeted at the
development of explosives, he was in fact the main scientist behind the development of chlorine gas and other poison gases used in the gruesome trench warfare of World War I. His
wife was so disgusted by his war contributions that she committed suicide with his service
gun. The father of chemical warfare soldiered on, though apparently quite unfazed. Later,
attempts were made to strip Haber of his Nobel Prize for his involvement in the war effort;
but, after he smartly pointed out that Alfred Nobel himself had made his fortune from the
pursuit of war, these voices went silent. Although Habers name may never be detached from
chemical warfare, currently, the food production of half the world relies on the large scale
production of artificial fertilizer he enabled.
The efficiency of applying nitrogen fertilizer is very low. The total amount of fertilizer
applied worldwide is 80 million tons annually. The amount that ends up in food amounts to
17 million tons annually, hence the efficiency is 21.25%. It is therefore paramount that we
have equipment that attempts to apply the material as uniformly as possible, to minimize
the undesired side effects on the environment.
Fertilizer can be applied in liquid form, but the granular form has major advantages
in terms of handling, storage and safety. Most granular fertilizer applicators use two discs
to propel the material outward and create a spread pattern behind the machine, see figure
227
228
Figure 14.1: This picture shows a granular fertilizer applicator (Vicon Rotaflow) with dual
discs. This kind of spreader is very popular mainly in Europe.
(14.1). Figure 14.2 shows the adjustment that can be made on a Vicon dual disc spreader.
A smart spreader control system could adjust these two automatically to realize an optimal
spread pattern.
The ultimate goal of granular fertilizer application is to reach the target application rate
(in kg/ha, (lbs/ac)) at the highest uniformity possible. This goal can be accomplished by
having a spread pattern that can be overlapped to yield a high uniformity. The overlapping
process amounts to several parallel passes of the spreader, in either race track (circuitous)
mode or back & forth mode. The distance between consecutive overlaps is called the swath
width. It is obvious that the chosen swath width is inversely proportional to the application rate, but it also affects the uniformity of the overlapped pattern. Ideally, a pattern is
robust, meaning that the uniformity remains acceptable, even if the swath width varies
(either by choice or owing to errors in the driving path).
The uniformity of a spread pattern is expressed in the coefficient of variation (CV ), which
is a statistical measure equal to the standard deviation of the overlapped pattern divided
by the mean of the overlapped pattern. A rule of thumb for row crops is that a CV lower
than 15 % will not cause damage to the crop, although the requirements are more stringent
for vegetables. It is clear that an ideal pattern would have no variability at all, and therefore,
the CV would be zero.
CV =
(14.1)
After overlapping it at a distance termed the swath width, this pattern should be highly
uniform; overapplication can burn the crop (and let nitrogen leach into the groundwater) and
underapplication can starve the crop of nutrients. Conventional spreaders such as shown here
are designed for a constant application rate, and they require annual calibration to maintain proper uniformity. Testing these machines can be time consuming and expensive since
229
Figure 14.2: The Vicon rotaflow spreader has an adjustment for the application rate on the
right (that controls a gate) and one for uniformity on the left (that rotates the whole unit
and therefore the spread pattern).
large halls are needed to collect the material in collection trays, the traditional method of
calibration. Figure 14.3 shows a photo of such a large calibration hall. It is obvious that
these facilities are quite costly because of the free roof span needed. The building as shown
calibrates spreaders by having transverse collection trays, all of which are moved to simulate
the machine being driven across a field. During the test, the machine is turned on, fertilizer
is emanated from the machine, and the trays move all the way from the left to the right of
the building. After the material is collected, the tray contents are weighed automatically,
after which the overlapped pattern can be simulated using software. Figure 14.4 shows such
a program that takes collection test data from an Excel spreadsheet and calculates the overlapped pattern parameters (application rate and CV ) as a function of the swath width. The
program can simulate both race track and back & forth patterns [2] (pdf), [3] (pdf).
The US standard for calibration of ground based fertilizer spreaders is ASAE S341.3 [4].
Figure 14.5 shows the calibration procedure conducted under field conditions.
It is clear from the previous discussion that proper calibration of granular fertilizer spreaders is important. An added level of complexity is that modern spreaders are often Variable
Rate capable, to allow farmers to apply fertilizer in management zones, where, depending
on topography, soil type and crop status, the application rate is varied in real time. However,
it is already difficult to achieve a uniform pattern with a constant application rate machine,
let alone a varying one! It is clear that we need to develop spreaders that can control their
application rate and uniformity automatically. This would circumvent calibration and ensure
that the rate is achieved at the maximum uniformity possible. One of the key elements of a
control system is a feedback sensor that can determine the spread pattern in real time. This
concept is shown in figure 14.6.
230
Figure 14.3: Granular fertilizer spreaders are typically calibrated indoors in countries with
frequent adverse climatic conditions. Shown here is a fertilizer spreader (left) that is being
prepared for testing. During the test, the collection trays (shown in the center) are moved
from left to right, material is deposited in the trays which contents are then conveyed out
to be weighed automatically.
The idea is that a sensor can measure the emanation velocity as well as the diameter
of particles, and then use a ballistic model to predict where these particles would land on
the ground, which was developed earlier [5]. Before the sensor was developed, a list of
requirements was made as follows:
1. The sensor must be able to detect particles ranging from 1 mm to 5 mm in size
2. The sensor must be able to detect particles traveling at a very high speed
3. The sensor cannot interfere with the flow of particles (be non-intrusive)
4. If the sensor fails, this must be detectable instantaneously, rather than remain obscured
5. The sensor must be rugged and robust so it can be used for outdoor application
6. Frequent re-calibration must be avoided
Optical sensors use electronic components that can detect events that are minute in time,
satisfying requirement 2. However, optical sensors when used in an outdoor environment often suffer from contamination issues, which need to be dealt with. Furthermore, the choice
needs to be made to create an analog or a digital sensor or a hybrid. Analog sensing arrangements have the nasty tendency to become less accurate over time, without notifying the end
user. A digital sensor in contrast fails completely, which can be easily detected. From the
list of requirements, it became clear that the sensor had to be based on radiation or light
interruption to satisfy requirement 2. The most restrictive requirements were items 1, 5 and
231
Figure 14.4: Spread Pattern Analysis Tool (SPAT) is a program that allows for overlapping
patterns and to calculate the CV and application rate as a function of the swath width.
Figure 14.5: Granular fertilizer applicators can be calibrated under field conditions, given
that the environmental parameters are within those stated in the standard.
232
Figure 14.6: An alternative to using large halls for granular fertilizer spreader calibration is
to use a sensor that measures the diameter and the velocity of individual fertilizer particles,
and to predict the landing locations of each particle, forming a spread pattern (top right).
Here is a movie that shows how the sensor rotates around a single disc spreader.
6. Virtually all sensors require calibration, and developing one that doesnt is a tall order,
but not impossible. For instance, a true balance to weigh objects does not need calibration,
since it is based on the principle of a balance of force.
The high speed of fertilizer particles was not a big concern, since opto-electronics is quite
fast. However, a true challenge was to detect small fertilizer particles, since the smallest
optical sensors available at the time were about 5 mm wide. The idea was to make two
photosensitive arrays by placing up to 30 optical sensors side by side, and then connect
all outputs in a logical OR function. However, if the sensors are spaced 5mm apart, and
the fertilizer particle is only 1mm wide, it would most likely go undetected. Here is where
innovative thinking comes in. What was needed is optical magnification, such as found in
telescopes, microscopes, but in projection form. That is where it struck me: How about if I
project the images of particles using a slide projector principle? I wrote a piece of software in
MatLab that allowed for dimensioning of the sensor, by placing the lenses in those locations
that minimized the overall size of the machine as shown in 14.7.
For the sensor arrays themselves I considered using line cameras (although these were
very expensive and not that fast at the time), but, for the sake of simplicity, I ended up using
Honeywells SDP8601 OptoSchmitts. These are technically infrared sensors (that required
me to make sure I did not put in an infrared filter such as slide projectors have to prevent
burning the slides), with a very fast fall time of 15 nanoseconds (ns). The other nice feature
of this sensor is that it already has a voltage regulator, and a Schmitt-Trigger circuit built
in. Figure 14.8 shows the SDP8601 electronic circuit and its package.
The sensor was tested on a single disc granular fertilizer spreader, and the results published in [6] (pdf). This paper received an ASAE Superior Paper Award in 2003. Eventually,
together with my PhD student Giyoung Kweon, we worked out a complete system where
233
Figure 14.7: To design a projection mechanism with two lenses (akin to a slide projector) a
MatLab program was written that uses elementary thin lens equations.
Figure 14.8: The Honeywell SDP8601 is an optical detector with a photodiode (left), an
amplifier (LA), Schmitt Trigger and an output stage (transistor and pull up resistor on the
right). It also has a built-in voltage regulator. The cost of an SDP8601 is about $1.75. A
spec sheet of the device can be found here.
234
Figure 14.9: The spread pattern sensor is essentially a slide projector, in which the shadows
of particles are projected onto dual sensor arrays. This Time of Flight principle allows to
calculate the velocity and the diameter of each particle that passes the sensor.
the drop location of fertilizer on the disc was varied in order to optimize the pattern uniformity [7] (pdf). Later that same year, we published another paper in which we showed that
the friction coefficient of particles sliding along a straight radial vane could also be measured
in real time using the sensor [8] (pdf). This work was done together with CEMAGREF,
Frances premier agricultural engineering research institute. CEMAGREF was very interested in using my sensor for experiments, so I donated one to them. One of the problems
that we had was to prevent fertilizer particles from hitting the sidewalls and pass the sensor
at reduced speeds. To alleviate this problem, we fitted the sensor with a small window (that
did not work), then we put deflectors on it (did not work either), but finally we decided
to build a wide frame around the sensor that allowed particles to pass uninterrupted. The
latter worked very well. Here is a video where these stages are shown.
This design as shown in figure 14.9 was eventually adopted and published in [9] (pdf).
Figure 14.10 shows the sensor where the dual photo sensor arrays can be seen in the back
and the diverging lens in the front.
In summary, an optical sensor was developed that meets all requirements, is reliable,
inexpensive and well understood. So one wonders why this sensor was never commercialized,
it is after all published, which means anyone can use it (when an article is published, it is
no longer possible to obtain intellectual property (IP) rights). The answer is simply that
in academics, we produce various prototypes, we publish papers, get promoted etc., but we
merely hope that someone else will bring our ideas to actual practice. This, however, seldom
happens. Therefore it is time that departments start their own companies, and have a real
impact on society beyond the perceived intellectual impact!
235
Figure 14.10: The optical sensor was built with a PVC body that allowed for easy mounting
of the dual sensor arrays (in the back). In the front, the second (diverging) lens is visible as
well as part of a diaphragm.
Disadvantages
more expensive than liquid chemicals
poor metering characteristics, hard to
distribute uniformly
limited to soil applications as moisture is
required for activation
Storage is more complicated
bulky
dry conditions required
Granular Fertilizer
Application
Tony Grift
The objective of granular application is to apply the material with high field
capacity in optimal amounts and with high uniformity.
Selection of fertilizer
Metering
?
Transport ?
in spreader
Rotary disk
Pneumatic
Central hopper
Aircraft
1100-2400 lb payloads
80-120 mph application rate
swaths 40-50 ft
hopper
metering
discs
Disc spreaders
Application
rate
adjustment
agitator
Emanation
angle
adjustment
metering
unit
disc
adjustable
vane
10
11
12
Dual disc spreaders can accomplish high field capacity, and good
uniformity, if properly calibrated. Note the narrow spreading field
Disc spreaders
13
14
Application rate
setting
Border
spreading
15
16
18
19
20
Gate and vanes in an aerial spreader form the throat. Vanes are
usually asymmetric to compensate for the propeller air flow
21
22
23
Calibration methods
Dry materials
Run off test (material flow rate)
Collection tray test (distribution)
Spreader Height
Weather Station
Temperature
Humidity
Wind Velocity
Collection Bins
Aircraft velocity
Wind
25
27
28
30
31
32
33
34
36
37
38
1
N
x
i 1
std
CV
mean
1
N
x
i 1
Spread Pattern
Analysis Tool
(SPAT)
1 N
2
xi x
N 1 i 1
1 N
2
xi x
N 1 i 1
x
39
Given is the data collected using a collection test from an aerial fertilizer spreader.
There are 11 pans, the distance between the pans is 1m, the area of the pan is and
the collected amount is in grams. The swath width (SW) is 5m.
1
x
N
i 1
1 N
2
xi x
N 1 i 1
CV
41
13 18 24 24 18 13 12 13 13 12
16 g / pan
10
1
N
x
i 1
16 g
1 kg 10,000 m 2
kg
*
* 1 ha 160 ha
m 2 1000 g
42
Coefficient of variation
(13 16) 2 (18 16) 2 (24 16) 2 (24 16) 2 (18 16) 2 (13 16) 2 (12 16) 2 (13 16) 2 (13 16) 2 (12 16) 2
4.76 g
9
1 N
2
xi x
N 1 i 1
CV
CV
4.76 g
29.76%
16 g
43
44
Triangle-Wide
45
Triangle-Narrow
46
Trapezoid (3/12)
47
48
Trapezoid (4/12)
Trapezoid (5/12)
49
Trapezoid (6/12)
50
Trapezoid-Skewed
51
Crown
52
M-shape
53
54
M-shaped Skewed
55
The pattern with the lowest average CV has the highest robustness.
The Gaussian is best followed by triangles, trapezoids. Crown and
M-shapes are caused by poorly adjusted spreaders
M-Shaped
M-ShapedSkewed
Crown
TrapezoidSkewed
TrapezoidWide
TrapezoidNarrow
TriangleNarrow
TriangleWide
Gaussian
10
15
56
20
57
58
Automatic calibration of
granular fertilizer spreaders
59
60
10
61
62
120
60
4
150
velocity
Absolute
Angle
Encoder
180
[ms ]
diameter
30
2
Optical
Sensor
-1
210
[m]
330
240
300
270
h[ m]
L[m]
63
Mennel, R.M. and Reece, A.R. 1962. The Theory of the Centrifugal Distributor,
Part III: Particle Trajectories. Journal of Agricultural Engineering Research 3(1): 78-84
Converging
lens
Diverging
lens
Sensor
Array 2
Sensor
Array 1
Layer 1
Layer 2
11
SmartSpreader.mov
69
12
NAME:_________________
Given is the data collected using a collection test from an aerial fertilizer spreader. There are 11 pans, the distance between the pans is 1m, the
area of the pan is 1m 2 and the collected amount is in grams. The swath width (SW) is 5m.
1) Compute the overlapped pattern (total amount after overlapping, use the bottom row), in Back & Forth (BF) mode, which is the most
popular in aerial application.
2) Compute the application rate in kg/ha (1 ha is 10,000 m 2 )
5 10 14 13 11 9
Chapter 15
Grain harvesting
249
Feeder
House
Header
Gather
Cut
Convey
Separator
Threshing
cylinder
Thresh
Separate
Clean
straw
chaff
Cleaning
Shoe
chaff
Unload
Store
Unloading
Auger
Grain Tank
Convey
Clean Grain
Auger
Rotor Type
Case Axial Flow - 174-280 HP
New Holland Twin Rotor - 240-280 HP
Functional Processes
Header
Gather
Cut
Convey
Threshing
Cylinder
Separator
Thresh
Separate
Clean
straw
chaff
Cleaning
Shoe
chaff
Unload
Store
Unloading
Auger
Grain Tank
Convey
Clean Grain
Auger
10
11
12
Functional Processes
Feeder
House
Header
Gather
Cut
Convey
Separator
Threshing
cylinder
Thresh
Separate
Clean
straw
chaff
Cleaning
Shoe
chaff
Unload
Store
Unloading
Auger
Grain Tank
Convey
Clean Grain
Auger
13
14
Combine Layouts
Conventional
Threshing
Feeder
House
Header
Gather
Cut
Convey
Thresh
Separator
Threshing
cylinder
Separation
Separate
Clean
straw
Unload
Store
Unloading
Auger
Grain Tank
Cleaning
Dyna Flow (JD)
Quadra Flow (JD)
Cleaning
Shoe
chaff
Straw walker
Specials
Convey
Clean Grain
Auger
15
16
Cylinder- concave
combination
Rasp bar
Threshing between
rasp bar and
concave
Grain fall through
concave on grain
pan (Gp)
Spiked tooth
Threshing between
teeth on cylinder
and concave
18
JD Conventional layout
Unloading Auger
Concave
Straw walkers
Blower
Feeder house /
Beater
conveyor
Transverse cylinder
20
Straw Walker
Straw Walkers
Straw
Grain
Beater speed tied to
cylinder speed
21
22
24
Separation
Rotary separation
Based on centrifugal effect on heavier grain particles
JD Combs material over grates
25
26
27
JD 9750 STS
28
29
30
Combination of
threshing and
separation in one
cylinder
31
32
Cleaning
Fan blows fine chaf away from grain kernels
Adjustments
Fan speed
Chaffer opening
Sieve openings
Overcleaning:
Fan speed too high
grain blows out back
Clean grain in tailings
Undercleaning
Fan speed too low
Chaffer and sieve openings too far open
Heavy tailings
Straw
Grain
33
34
35
36
Functional Processes
Feeder
House
Header
Gather
Cut
Convey
Separator
Threshing
cylinder
Thresh
Separate
Clean
straw
chaff
Cleaning
Shoe
chaff
Unload
Store
Unloading
Auger
Grain Tank
Convey
Clean Grain
Auger
37
Unloading rate
38
Harvest losses
Preharvest losses
Wind shatter losses
Lodged grain
Field too wet for harvest
39
40
Cylinder
loss
Separating
& cleaning
loss
Total loss
Oats
1.5
0.8
2.0
4.3
Wheat
3.4
0.1
0.6
4.1
Soybeans
5.8
0.6
0.7
7.1
41
42
Symptom
Lodged and wind shattered grain out of reach of cutter bar
Remedy
Reel losses
Symptom
Kernels counted after header passes before straw output
Remedy
44
Concave
clearance
Cylinder
speed
Results or indicators
wide
wide
slow
wide
fast
tight
tight
slow
tight
fast
correct
correct
45
46
Residue Management
Overthreshed situation
Grain walks out of the back of the machine because walkers are
plugged with ground up straw. Concave clearance is too tight.
Reduce erosion
Straw Chopper
Chaff Spreader
47
48
49
50
51
52
LEXION 480
LEXION 480
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
The cutterbar
The cutterbar
Reel
Feed rake
Crop lifter
Intake auger
Divider
EDUCATION SERVICE
EDUCATION SERVICE
+ 200 mm - 100 mm
+ 200 mm - 100 mm
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
Laser Pilot
Laser Pilot
The Laser Pilot transmits pulsing
beams of light. These are
reflected by the standing crop
and the stubble. Due to the
greater distance, the reflection
from the stubble takes longer
than from the standing crop.
The Laser Pilot uses this time
difference to calculate the edge
of the crop and guides the
combine harvester precisely
along it.
EDUCATION SERVICE
EDUCATION SERVICE
Main concave
Pre-separation concave
EDUCATION SERVICE
CLAAS Multicrop
EDUCATION SERVICE
CLAAS Multicrop
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
Threshing
drum drive
Variable speed drive
Accelerator
drive
Diesel engine
Impeller
drive
Output
EDUCATION SERVICE
EDUCATION SERVICE
Return pan
Above each straw walker rack there are two driven tines which shake out the straw
from above and speed up the straw flow. This improves grain separation.
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
Cleaning
Cleaning
Grain auger
Returns auger
EDUCATION SERVICE
EDUCATION SERVICE
EDUCATION SERVICE
CEBIS
EDUCATION SERVICE
CEBIS
EDUCATION SERVICE
EDUCATION SERVICE
Yield measurement
Photo cell
Humidity meter
Quantimeter
module
The CEBIS / IMO
or ACT display shows:
Front axle
Throughput in t/h
Yield in t/ha
Quantity harvested in t
CEBIS
Grain humidity in %
ACT
Lateral and
longitudinal position
module
Dry humidity in t
Average humidity
per Job in %
EDUCATION SERVICE
Harvest Practices
Harvest Timing
Harvesting grain at too high of moisture can result in
severe kernel damage during threshing and drying.
Conversely, allowing corn to field dry too long can lead
to reduced yield and quality if stalk or ear rot diseases
or insect feeding damage are increasing.
Monitor stalk quality. If stalks lodge, ears can come in
contact with the ground which can lead to ear rot
development.
Hybrid Selection
Selecting hybrids rated highly for grain quality traits is
the first step in producing highest quality grain.
Pioneer brand hybrids are rated for resistance to
prevalent ear rot diseases and test weight.
High test weight is a grain trait that helps kernels resist
breakage during threshing and handling.
Combine Settings
Proper combine settings and operation are critical to
preserve grain quality.
Optimum ground speed depends on the condition of
the crop, but should generally be as fast as possible
without plugging the head or threshing mechanism.
Snapping rolls should be set relative to ground speed.
A setting that is too fast will cause kernels to be shelled
and lost and increase breakage of ear butt kernels.
Use the lowest possible cylinder/rotor speed that will
shell the grain within acceptable loss levels (1% in
good-standing fields).
Begin with manufacturer suggested sieve and fan
settings and check and adjust frequently. Crop
conditions can change rapidly during autumn days.
Pioneer Agronomy Sciences , TM, SM Trademarks and service marks of Pioneer Hi-Bred.
2009, PHII
Pioneer Agronomy Sciences , TM, SM Trademarks and service marks of Pioneer Hi-Bred.
2009, PHII
CONVERSIONS
From
To Pure
SI
N
m2
Multiply by
9.81 N
1 kg 1550 in 2
*
*
= 6897.17
1 kg 2.2046 lb
1 m 2
Pressure*
PSI =
Flow rate
GPM =
Gal
min
m3
s
1 m3 1 min
*
= 6.25 * 10 5
266.67 gal 60 s
LPM =
l
min
m3
s
rad
s
Nm
s
1 m3 1 min
*
= 1.666 * 10 5
1000 l 60 s
Rotational
velocity
Power
Displacement
lb
in 2
rev
min
lb * ft
HP = 550
s
RPM =
m3
rad
N
1Bar = 105 2
m
CIR =
N
* 1[MPa ] = 106 2 ,
m
in3
rev
2
1
rad 1 min
rev * 60 s = 0.1047
1 kg 1 m
9.81 N
550 *
*
*
= 745.9
1 kg 2.2046 lb 3.28 ft
m3 1
1
*
61023.74 in3 2
rev
6
rad = 2.608 * 10
Unit
Meter
Second
Radian
Kilogram
Kelvin
Ampere
Candela
Mole
SI Symbol
m
s
rad
kg
K
A
cd
mol
Prefix
giga
mega
kilo
hecto
deca
deci
centi
milli
micro
nano
pico
1.
2.
3.
4.
5.
Chicago
Bloomington
Alton
Effingham
Freeport
The actual flow Q A is computed using the flow meter reading QR . Both values should be in
l
min
TRAINER SCHEMATIC
The reservoir (1) supplies hydraulic oil to a gear pump (7), which has a fixed displacement of
3
cm 3
6 m
or
1
.
028
*
10
6.46
rad
rev
by the electric motor (8). Since there may be times when oil cannot be transmitted though the
pressure manifold (27) (for example, when no circuit is connected), an adjustable pressure relief
valve (PRV, 5) is provided to allow an escape path for the oil from the fixed-displacement pump.
The relief valve is ordinarily set to open at zero pressure at startup; then, the pressure can be
increased gradually to more safely check the correctness of a newly connected circuit. A vent
line provides another path from the pump to the reservoir; the vent line is a convenience on the
trainer and is not present on most practical hydraulic circuits. The vent line can be blocked by the
vent valve (4), but this valve should always be open when starting or stopping the pump. This
prevents damage to the pump due to transient hydraulic shocks. Except for the special cases
discussed in Lab 2, the suction-line valve (3) should always be open and the aeration valve (30)
should be closed when the pump is running.
1.3B PROCEDURE
1. Set the PRV (5) for minimum pressure by turning the knob fully counterclockwise (ccw),
and the vent valve (4) to open, i.e., with the handle horizontal.
2. Use the start switch on the controller (10) to start the pump.
3. Close the vent valve, then turn the relief pressure setting knob clockwise until the
pressure gage (6) reads 500 psi ( 3.45[MPa ] ).
4. Cycle the vent valve between open and closed several times while observing the pressure
gage and listening to the noise level from the trainer. Record the pressure readings when
the vent valve is open and when it is closed.
1.4B REPORT
1. Using the pressure information from step 4, describe the purpose of the vent valve on the
fluid power trainer.
2. Noise level can be an important clue to the operation of a fluid power system. Does the
noise level decrease, remain the same, or increase when the vent valve is closed? What
clue does this provide as to the status of the circuit?
m3
3. Assume the pump delivery is 0.166 * 10 3 .
s
a. Note the general Hydraulic power equation here (use Equation Editor) in Pure SI
Units (no conversion factors).
b. Calculate the hydraulic power generated when the vent valve is open
c. Calculate the hydraulic power generated when the vent valve is closed
4. What path does the oil take from the pump back to the reservoir, and what is the power
being used for
a. when the vent valve is open
4. Turn the suction line valve towards the closed position until you hear a steady, highpitched squeal and gravel-like rattle from the pump. Quickly record the pump pressure
and the suction line vacuum, and then open the suction line valve. These are the
readings for cavitation.
5. Partially close the suction line valve until the vacuum gage reads 5 inches of mercury
(16.7 kPa, or 0.167 atmospheres). Then open the aeration valve by turning it counter
clockwise. The vacuum gage reading should drop to zero. Listen for an erratic popping,
banging, high-pitched noise as the air enters the pump. Quickly record the pump pressure
and the suction line vacuum, then open the suction line valve and close the aeration
valve. These are the readings for aeration.
6. Reduce the relief pressure setting to zero, open the vent valve and stop the pump.
1.4B REPORT
1. Suppose you suspected aeration or cavitation or was occurring in a hydraulic pump, but
there was no gage connected to measure vacuum in the suction line. Could you
distinguish between cavitation and aeration using only your ears? On the basis of the
exercise you have just conducted, briefly compare the noise from cavitation with that
produced by aeration.
2. If a vacuum gage was connected in the suction line, briefly explain how you could use
the reading to distinguish between cavitation and aeration.
3. Make a list of some likely causes of cavitation in a hydraulic pump. For each item, list
one or more ways to correct the cause.
4. Aeration can cause damage to hydraulic oil because it can oxidize in the presence of air
and high temperatures. What are some likely causes of aeration? Again, list one or more
ways to correct the cause.
10
Notes:
11
Objectives
To understand the purpose and characteristics of the Pressure Relief Valve (PRV), as well
as to become familiar with the working of two types of Directional Control Valve.
A pressure relief valve (PRV) is a valve that virtually every hydraulic system contains. Its
purpose is to limit the pressure on the pressure side of the pump to make sure that no
damage is done when no flow is allowed into the system. Hence, the ideal characteristic of a
PRV is simple: It should do nothing (meaning stay closed) until the PRV setting (500 PSI
in our labs) is reached. At that point, it should open completely and dump all pump flow
(12 l/min in our lab) back to the reservoir. This will cause the pressure to drop, so the PRV
should close again. This process should repeat itself very quickly so that exactly the PRV
setting (500 PSI) is maintained.
The fact that we call this ideal behavior implies that in reality this is not realizable. If we
use a simple PRV where a ball is backed up by a spring, the characteristic is far from ideal.
There is a better implementation, which is called the Pilot Operated PRV. This is a dual
stage valve which separates the sensing of the pressure from the dumping of hydraulic fluid
to tank. Its principle is shown in Figure 1. Here is how it works:
First let us assume that the pump pressure is lower than the PRV setting: The pressure
on the pressure side of the pump is sensed through a small hole called an orifice, which is
drilled in the land of the hollow stem piston (see dashed lines in Figure 1). It is important to
note that at this point there is no flow through the orifice, since the poppet valve is closed.
Therefore, the pressure above and below the orifice are the same, no matter how small the
hole (students often find this confusing, they think that wherever there is a flow resistance
(obstruction), there must be a pressure drop, but this is only the case if there is flow). Since
the pressure above the land (and orifice) is equal to the pressure below it (pump pressure),
the piston is balanced, and a spring keeps the connection to tank closed.
Figure 2: The pump on the HTU-01 trainer is a fixed displacement gear pump.
A gear pump is the simplest and cheapest version of a pump. It has limited pressure (due
to the limited sealing between the teeth of the gears), low flow rates, and it is noisy. The
displacement of a pump is equal to the theoretical amount of oil that would be moved from
the inlet port to the outlet port of the pump during one radian of revolution of the input
shaft (there are 2 radians per revolution). In formula form:
q = D
(1)
m3
m3
, D is the pump displacement in
, and is the angular
where q is the flowrate in
s
rad
rad
velocity of the pump in
. In the US, you will see the abbreviation CCR (Cubic centimeter
s
per revolution), and the CIR (cubic inch per revolution). The gear pump in the hydraulic
l
.
benches have a constant delivery of about 3 GPM, or 11.25
min
2.1
Procedures
Connect the inlet port of the flow meter to the pressure manifold, the outlet port of the flow
meter to the pressure gage and to port P of the open-center DCV (12). Connect the T port
of the DCV to the return manifold. Using the longer hoses, connect cylinder port A of the
DCV to the lower port of hydraulic cylinder (25) and the upper port to port B of the DCV.
1. Set the PRV for minimum pressure (knob fully ccw), the vent valve open, the suctionline valve open, and the aeration valve closed (fully cw).
2. Start the pump, close the vent valve and set the PRV to 500 psi (3.45 MPa).
3. Open the vent valve and stop the pump, leaving the PRV setting where it was.
4. Connect the inlet port of the flow meter (16) to the pressure manifold (27), the outlet
port of the flow meter to one port of the needle valve (20) and the other port of the
needle valve to the return manifold (the lower set of connections on 27).
5. Open the needle valve by turning its knob fully ccw.
3
6. Start the pump, close the vent valve, and record the system pressure and the flow
meter reading (in l/min). Note that the inner ring of the gage reads in hundreds of
kPa (in Bar); a reading of 30 on the inner ring, for example, indicates 3000 kPa or 3
MPa.
7. Take readings of system pressure and flow as the needle valve is closed down in small
increments, starting with 400 psi , then 410,420, 430, 440, 450, 460, 470, 475, 480, 485,
490, 495 and 500 PSI (fully closed). Insert the data in Excel for easy manipulation.
Calibrated Flow and the data in Pure SI are computed using the calibration equations
and conversion factors.
Target
pressure
(PSI)
400
410
420
430
440
450
460
470
475
480
485
490
495
500
Measured
l
flow
min
Calibrated
l
flow
min
Measured
pressure
N m2
Calibrated
m3
flow
s
Hydraulic
Nm
power
s
Make sure to always be closing down on the needle valve during this part of the test.
If you overshoot the reading, back up to a lower pressure and come up on it again.
Think about where the flow is going as you close off the needle valve.
8. Continue to take pressure and flow data as the needle valve is opened back up in similar
small increments. Again, make sure to be always opening the needle valve during this
part of the test.
Target
pressure
(PSI)
500
495
490
485
480
475
470
460
450
440
430
420
410
400
Measured
l
flow
min
Calibrated
l
flow
min
Measured
pressure
N m2
Calibrated
m3
flow
s
Hydraulic
Nm
power
s
9. Open the vent valve, turn the PRV setting fully ccw, and stop the pump. Then
disconnect your circuit and put the components back into storage on the trainer.
2.2
Questions
Q1:Calculate
3 the theoretical flow rate inh Purei SI, given that the pump displacement is
cm
rev
6.46
, and the pump speed is 1725
. Conversion factors can be found here.
rev
min
A1:
Q2: Considering your observations from Steps 7 and 8 of the procedure, briefly explain how
the flow path of the oil from the pump back to the reservoir changed as the needle valve was
adjusted.
Directional Control Valves control the direction of fluid in a hydraulic circuit. A typical
application is to change the direction of movement of linear actuators (cylinders) or rotary
actuators (motors). The DCVs used in this exercise are 4-3 DCVs, which stands for 4-way
(for the 4 ports) 3-position (for the 3 possible positions of the control spool). You will be
using an open-center 4-3 DCV initially (see Figure 4 and later a closed-center 4-3 DCV as
shown in Figure. A schematic of this arrangement is given in Figure 5.
3.1
Procedures
Connect the inlet port of the flow meter to the pressure manifold, the outlet port of the flow
meter to the pressure gage and to port P of the open-center DCV (12). Connect the T port
of the DCV to the return manifold. Using the longer hoses, connect cylinder port A of the
DCV to the lower port of hydraulic cylinder (25) and the upper port to port B of the DCV.
1. Set the PRV for minimum pressure (knob fully ccw), the vent valve open, the suctionline valve open, and the aeration valve closed (fully cw).
2. Start the pump, close the vent valve and set the relief pressure to 500 psi (3.45 MPa).
3. Open the vent valve and stop the pump, leaving the PRV setting where it was.
4. Connect the circuit as shown in Figure 6. Note that the symbol for the flow meter is
different (it looks like a tennisball on the trainer).
5. Start the pump, close the vent valve, and record the pressure and the flow meter reading
l
). To aid your understanding of the circuit, consider how the oil returns from
(in
min
the flow meter to the reservoir, i.e., what is its path? Note that the cylinder is not
moving in Step 5; the DCV is in the neutral. Why is the pressure at gage 17 not zero?
6. Use the lever on the DCV to alternately extend and retract the hydraulic cylinder.
Observe and record both the flow and the pressure while the cylinder is moving. Also
record the flow and pressure when you continue to hold the lever away from center
after the cylinder has reached the end of its stroke. Consider the path of the oil while
the cylinder is moving and also while you continue to hold the DCV spool off of center
after the cylinder has reached the end of its stroke. These considerations should enable
you to understand the reasons for the changes in flow and pressure in Step 6.
Measured
pressure
(PSI)
Measured
l
flow
min
Cylinder
Extending
Cylinder
Stalled
Cylinder
Retracting
7
Calibrated
l
flow
min
Measured
pressure
N m2
Calibrated
m3
flow
s
Figure 5: Here the electrically (solenoid) operated closed-center Directional Control Valve
is shown. The fact that the pump and tank port are not connected means that in idle, the
pump must pressurize to the PRV setting, which wastes a lot of energy. The fact that the
a,b ports are not connected means that this kind of valve can keep a cylinder locked in place
which is an advantage. However, if this kind of valve would control a motor, the abrupt
cutting of the flow to the motor will cause seals to fail, or, in case of a hydrostatic drive, the
passengers to be launched through the windshield.
Figure 6:
7. Now shift all connections from the open center DCV (12) to the corresponding ports
on the closed center DCV (14) shown in Figure 5.
8. Repeat Step 5 above, with the closed-center DCV in the circuit. Remember to open
the vent valve before starting or stopping the pump.
9. Repeat Step 6 above, with the closed-center DCV in the circuit. Use the solenoid
selector switch (9) to activate the solenoids to extend or retract the cylinder. Moving
the switch to A directs oil from the pump to port A of the DCV, while moving the
switch to B directs the oil to port B.
10
3.2
Questions
Q1: Explain the reason for the differences in pressure and flow in neutral for the two different
DCVs (open center and closed center). Include in your explanation the flow path of the oil
in each case.
A1:
Q2: Explain why the pressure and flow changed when the cylinder reached the end of its
stroke (as compared to while the cylinder was moving) in steps 6 and 9.
A2:
Q3: What are the similarities and differences observed in the circuit when the connections
to the hydraulic cylinder are reversed as in step 10?
A3:
Q4: Assuming the power efficiency of the pump is equal to 0.85, calculate the power delivered
to the pump during each of Steps 5, 6, 8, and 9 of the procedure. Make a table summarizing
the flow, pressure and power for each step of the procedure. This will include data for each
DCV in neutral, while extending and retracting, and while being held at end-of-stroke. Remember that the flow is never zero from a fixed displacement pump; it has to go somewhere!
A4:
Q5: What happens to the power that you calculated for Steps 5 and 8 of the Procedure
(with the DCV in neutral), i.e., how is the power used? Taking your answer into account, is
the open center or closed center valve better suited for use with a fixed-displacement pump?
Why?
A5:
11
Objectives
To understand the working of a Sequence Valve and, en passe, the working of a Pressure
Regulator Valve.
Introduction
A sequence valve is used to make events in a hydraulic circuit happen in sequence. The
classical example is where a workpiece needs to be clamped by a cylinder, before a drill is
allowed to come down and drill a hole. Therefore, we need a valve that measures a pressure
in a primary (clamping) circuit, and when this pressure exceeds a preset pressure, then a
path is opened to allow flow to a secondary circuit. This sounds a lot like a PRV, with the
only difference that in a PRV the secondary circuit is always the tank. Figure 1 shows the
Sequence Valve that is present on the hydraulic trainers in the lab.
The ANSI symbols for a PRV and a Sequence valve are very similar, and as a consequence,
so is their functioning. Both valves are normally closed and both sense the pressure upstream. Both open when the pressure upstream reaches a preset limit. The difference is
that when the upstream pressure is reached, the PRV dumps fluid back to tank as a safety
measure, whereas the Sequence Valve passes on the fluid to a secondary circuit. This also
explains why the PRV has no separate drain line; Since it is always connected to tank, it
can be drained internally. The sequence valve however does not have a tank connection, so
it needs to be drained externally to tank. Figure 2 illustrates the difference between a PRV
and a sequence valve symbol. Figure 3 shows the symbol used in the schematic, which is
borrowed from FluidSim. FluidSim does not have a sequence valve, because essentially, it is
a shutoff-counteracting valve where the remote port X is connected to the pressure port P.
Procedures
The circuit demonstrated in the exercise represents a Clamp-Drill circuit that utilizes a sequence valve. Before a sequence valve can be used in a circuit, its switching pressure must
1
Figure 1: Here the sequence valve is shown that is present on the hydraulic trainers in the
lab.
Figure 2: The PRV on the left can be distinguished from the sequence valve on the right,
merely by recognizing that the PRV has no external drain line. It does not need it because
a PRV is ALWAYS connected to the tank, so it can be drained internally. The Sequence
valve has a primary circuit, and a secondary circuit, neither one of which are necessarily
connected to the tank, therefore it does need an external drain line.
Figure 3: The sequence valve here is shown in FluidSim form, that means it is a shutoff
counteracting valve with the remote control port X connected to the pressure port P. Note
that the symbol in the American ANSI system is different.
be adjusted such that it opens are a preset value. In reality it does not have an binary
open/closed relationship, but it opens gradually over a certain pressure range. This kind
of behavior is typically modeled as an S-curve (mathematical sigmoid function). Figure 4
shows this relationship.
1. To set the sequence valve, we first need to measure the total pump flow by directly
routing the manifold flow through the flow meter and return to tank (should be close
l
), write down this number.
to 12
min
2. Build the circuit as shown in Figure 6. This circuit allows you to set the flow rate to
half of the pump flow rate, and then adjust the sequence valve such that this happens
at a preset pressure (400 PSI). To begin make sure that the set screw on the sequence
valve is turned all the way out (CCW). Since the pressure to the right of the sequence
valve is zero (it is directly connected to the tank), the pressure at the pressure gage is
essentially the pressure across the sequence valve.
3. To set the midpoint, you need to set the pressure regulator valve such that it produces
the desired pressure at which the sequence valve conducts half the pump flow rate
l
@ 400 PSI). This valve is shown in 5. Make sure you connect the drain,
(say 6
min
otherwise it will not work.
4. Now you turn the set screw on the sequence valve until the flow rate reaches half the
l
pump flow rate (6
).
min
To study the behavior of the sequence valve, build the circuit as shown in Figure 7.
Figure 5: The pressure regulator (aka pressure reducing valve) maintains a constant pressure
even if the flow rate through the valve varies. We can use this to assure a constant pressure
across the sequence valve when we are setting the flow rate to half the pump flow rate.
Turn up your PRV to 600 PSI (since we have a normally closed valve you can do this without
removing any hoses). Check the working of your circuit. If all is correct, the clamp cylinder
should extend at a low pressure (since the cylinder has no load) all the way until it stalls
out. Then the drill cylinder will extend at the Sequence valve setting (400 PSI), until it is
fully extended, at which point the pressure increases to the PRV setting (600 PSI). There is
no preset order for retraction: Since neither cylinder has a load, one can retract before the
other or they can retract simultaneously.
Figure 6: The sequence valve needs to be set to a value at which it opens. In reality it does
not have a binary open/closed relationship, but it opens gradually over a certain pressure
range. This circuit allows you to set the pressure drop across the sequence valve using a
pressure regulator (400 PSI), and then set the flow rate through the sequence valve to half
l
of the pump flow rate (6
).
min
Figure 7: The complete circuit for a clamp-drill arrangement is given here, it requires quite
a few T-connections. Note that the sequence valve on the trainer does not have an X-port
(it is internal) but it does have a drain port which is not shown in this FluidSim schematic.
Also note that the sequence valve can only conduct flow from the P to the T port, therefore
you need to bypass it with a check valve to allow the drill cylinder to retract.
Questions
Q1: Imagine that you do not have a pressure regulator valve laying around, how would you
be able to set the sequence valve to conduct half the pump flow at a preset pressure?
A1:
Q2: If you had the task of characterizing the behavior of the sequence valve (meaning how
the pressure is related to the flow as shown in 4), how would you do that? Imagine that you
have an hydraulics trainer as we do in the lab.
A2:
Q3: The flow meter as used on the hydraulics bench is hard to read and not very accurate.
What element on the hydraulics bench could be used as a proxy, and how would you go
about calibrating it?
A3:
Objectives
This lab is designed to 1) acquire real life data from an air cannon experiment, 2) to model
the cannon in a spreadsheet using standard gas laws.
Introduction
Air cannon have great use in testing protective materials because they can launch projectiles at high speed that impact for instance safety glazing for off-road equipment, trains,
airplanes etc. This particular cannon design was developed to test safety glazing for the cabins of feller bunchers, which are large dozer frame based machines that can cut and collect
trees. These feller bunchers have a saw disc that spins at 5000 RPM, and it is not hard to
imagine what happens if one of the 1kg saw teeth accidentally dislodges. It is obvious that
operators must be protected from this danger with safety glazing (which is typically made
from multi-layered polycarbonate aka Lexan). To test this material, we developed a large
cannon that can launch a sawtooth into the Lexan at very high speed.
This is however not the main reason why we use an air cannon in ABE223: As an engineering student you will be exposed to many physics and particularly mathematics courses
where youll see little to no practical application. Therefore in this class, we work the other
way around, starting with an application and working backward to see what theory can
be applied to understand its working. We will model the cannon using gas laws that you
learned about in PHYS212. There are three major gas laws for a closed system, and one
of them describes the cannon to a tee. Once we have the proper model, we will use Excel
to predict the amount of energy contained in the compressed air, which is a function of
boundary conditions (starting pressure, starting temperature and starting volume) as well
Procedures
This lab consists of two parts being an outdoor experiment, and a subsequent computer lab.
3.1
Outdoor experiment
Firstly, we will experiment with an air cannon that was built for safety demonstrations. It
is a very powerful machine but it is built very strong, there is no chance of it disintegrating
under pressure. We will launch tennis balls into the air and measure their air time while
varying the pressure in the cannons reservoir. The second part consists of completing an
Excel spreadsheet that allows us to calculate the exit velocity of the tennis ball.
You need to understand how the cannon is built and how it works, otherwise youll have
a hard time understanding the computer lab. The cannon consists of three main parts, a
reservoir (tank, main chamber), a barrel, and a piston which is slightly larger than the barrel,
see Figure 1. The principle of the cannon is as follows: During charging, air flows past the
piston into the main chamber, in which the pressure rises. Since the pressure in the barrel is
atmospheric, the piston is pushed against the barrel, sealing it off (this is another example
of a positive seal). The main chamber is further pressurized to 100 PSI. To fire the cannon,
the pressure behind the piston is dropped quickly using a ball valve: Since the pressure on
barrel side of the piston is 100PSI and now we dropped the pressure on the ball valve side
to atmospheric, the piston will slam back at a very high speed, which opens up a connection
between the main chamber and the barrel, releasing all pressure in the process.
During the test we will measure the air time. The clock starts when you hear the cannon
fire, and stops when you see the tennis ball hit the ground. In the second part of this lab,
we will correlate these air times with predicted air times from a ballistic model.
Target pressure (PSI)
60
60
80
80
100
100
Figure 1: Here the back of the cannon is shown with the main chamber outline, barrel
(yellow) and piston (salmon).
3.2
Computer lab
After we have collected some data outdoors, we need to build a spreadsheet that contains all
inputs and equations needed to calculate the energy that is released during the acceleration
of the projectile in the barrel. A template is provided which contains the basic structure of
the spreadsheet where the left side contains required inputs in gray, as well as conversion
equations in yellow. The right hand side in blue contains all equations we need to calculate
the Volume before firing (V1, the volume, pressure and temperature at the point where the
projectile leaves the barrel (V2, p2 and T2 respectively) as well as the work being done
during the acceleration of the projectile in the barrel W . Note that the volume before firing
(V1) is the reservoir volume minus the barrel volume, and the volume after firing V 2 is the
volume of the reservoir (we ignore the wall thickness of the barrel).
Every equation you need can be found in the Air Cannon chapter in the ABE223 course
packet. In the next lab, we will grab all the parameters and Work from this spreadsheet in
MatLab and calculate the trajectory of the projectile. MatLab will then return its calculated
parameters to this spreadsheet in the bottom right block (light gray).
Figure 2: Here a template of the Excel Spreadsheet is shown. All yellow fields must contain
formulae. All inputs are in gray. Note that none of the equations contain any conversion
factors, we do this by a priori converting all inputs to pure SI units.
Figure 3: Here the complete spreadsheet is shown, make sure that your numbers are close
to the ones shown here.
Question
Based on the spreadsheet, how would you go about designing a more powerful cannon?
Elaborate here.
Assignment
You are an engineer in charge of conducting high speed launch tests as we did today with
the air cannon. Write a complete safety procedure (1/2 page single spaced) that contains
the following:
1. Detailed procedures for firing the cannon in a safe manner.
2. What kind of safety measures must be in place?
3. A list of what to do in case of eventualities. This means you need to think through
everything that can go wrong and write a response for all what if scenarios you can
think of.
Notes:
Lab5: Ballistics
STUDENT NAME
October 14, 2015
Objectives
Introduction
In the first part of the cannon lab, we determined the energy and exit velocity of the projectile
after being accelerated in the cannon through the adiabatic expansion of air. The ballistics
part of the cannon lab deals with the trajectory of the projectile through air. We will use
the Excel spreadsheet we developed before (see Figure 1) to pass on the initial conditions
for the simulation to MatLab. MatLab is then used to solve twin differential equations that
represent the trajectory of the projectile through air. After the simulation completes, we
pass the results back to the Excel spreadsheet.
Procedure
The ballistic model for a spherical projectile in air is shown in Figure 2. This led to the
following model:
p
x = K x x 2 + y 2
p
y = K y x 2 + y 2 g
(1)
(2)
CD A
kg
is the density of air,
where K =
, CD is the drag coefficient of the projectile,
2m
m3
A[m2 ] is the frontal projected area of the projectile, and m[kg] its mass.
h m i x, y are the horizontal and vertical position, and g is the gravitational acceleration in 2 . These equations
s
were translated into a set of first order equations so MatLab can read them in a file called
Aerodyn.m. MatLab and Excel can be linked via a Dynamic Data Exchange protocol after you 1) define the location where MatLab can find the Excel file and 2) open the Excel file.
Figure 1: The complete Excel Spreadsheet is shown here, make sure that your numbers are
close to the ones shown here.
Figure 2: This simple force balance model yields two coupled non-linear differential equations
that can only be solved using a numerical method.
Make sure your MatLab programs AND your Excel spreadsheet are in the same
folder so MatLab can find everything.
x 1 = x2
(3)
q
x 2 = x1 = x = Kx2 x22 + x24
x 3 = x4
x 4 = x3 = y = Kx4
(4)
(5)
x22 + x24 g
(6)
The Aerodyn.m file is given here, copy and paste this text (except for the line numbers) in
the MatLab editor and save as Aerodyn.m:
1
2
3
4
5
xp
You can download the main program template file here. Note that this file is in ascii
format, you need to copy and paste it into the MatLab editor and save it as
YourNameBallistic.m. To initiate the communication between MatLab and Excel
the following code is used (change the DateFileName and DataFilePath variables.):
1
2
3
close all
clc
clear Channel
4
5
6
% =======================
global g K
Globals
====================================
7
8
9
10
% ======================= Constants
====================================
DataFileName
= 'CannonEnglish2.xls';
DataFilePath
= 'C:\Cannon\';
11
12
13
ChannelActive
g
= 0;
= 9.81;
14
15
16
17
18
19
20
21
22
23
24
25
Figure 3: The cells that contain the parameters that MatLab needs to calculate projectile
trajectories are given here.
MatLab takes some inputs from the Excel spreadsheet to initiate the simulation using the
ddereq command (Complete the list):
1
2
3
4
5
6
7
8
9
10
The simulation needs boundary (in this case starting) conditions. Since we are simulating
the position and velocity in both x and y direction, we need to specify the initial position in
the x direction , the initial velocity in the x direction, the initial position in the y direction ,
and the initial velocity in the y direction. You need to calculate these from the exit velocity
of the projectile V you got from the Excel spreadsheet. In MatLab, sin and cos take an
argument in radians, but if you use sind and cosd, you can provide the argument in degrees.
Further, the simulation needs the start and final time (take 0 and 100 seconds). Since in the
equations there is a K, and a g, these need to be defined as well, and they must be declared
global in both the calling and the called file (check both your main program and Aerodyn.m
program if this is the case).
2
3
4
5
7
8
10
11
12
...
13
14
15
16
...
17
18
19
20
%
x0
[t,x]
=
=
x0 Vx0 y0
Vy0
[0 Vx height Vy];
%
ode23(@Aerodyn,[t0 tfinal],x0);
You now have the time vector and a 4 column data vector x. The 4 columns represent
x, x,
y, y that is position in the x-direction, velocity in the x-direction, position in the ydirection, velocity in the y-direction. Once the particle hits the ground, the position in the
y-direction changes from positive to negative. Now find out at what time this happens using
min of find where the y-position < 0.
1
index
min(find(x(:,3)<0,1);
x dist
x vel
y dist
y vel
time
=
=
=
=
=
x(1:index,1);
x(1:index,2);
x(1:index,3);
x(1:index,4);
t(1:index);
2
3
4
5
6
7
8
9
%============
=================
Here you need to determine the exact maximum distance and the total air time. You do
this by linear interpolation of the distance plot and time plots as follows: From the figure it
is clear that:
x
xi xi1
=
yi1
yi1 yi
(7)
Figure 4: To calculate the landing position in the x direction, we use linear interpolation.
xi xi1
yi1 yi
(8)
You need to program this relationship in MatLab to obtain the maximum distance:
1
MaxDist
AirTime
MaxHeight=
2
3
4
When MatLab has simulated the complete trajectory, it outputs its data back to Excel with
the ddepoke command (Complete the list):
1
2
3
4
5
6
7
8
Figure 5: The variables that MatLab calculates are being pushed back into the Excel Spread
Sheet using these cell locations.
Output is generated through two plots, one where the x,y coordinates are shown versus
distance (this is called a trajectory).
1
2
3
4
5
6
7
8
9
10
11
12
Your program is complete when it generates a dual output plot (see Figure 6), AND when
you see that it transfers values back to the Excel Spread Sheet.
Figure 6: If your MatLab program runs properly, you will see this output graph.
Questions
Q1: If you were to launch a very light projectile, the exit speed would be high / low (circle).
In that case the object would decelerate fast/slowly in air. If you were to launch a heavy
projectile, the exit speed would be high / low (circle). In that case the object would decelerate fast/slowly in air. There is obviously a tradeoff here. Use your working MatLab
program to determine what the optimal mass is of an object to achieve the maximum air time and height. Assume a diameter of 65 mm (tennis ball) and an elevation
angle of 1 degree.
A1:
Q2: The optimal elevation angle to shoot a projectile at if there is no Aerodynamic friction
is 45 degrees. However, if there is Aerodynamic friction, this is not the case. Use your
working MatLab program to determine what the optimal elevation angle is in
the presence of Aerodynamic friction.
A2:
Q3: If you would take the cannon to the moon, and shoot a tennis ball from the cannon at
a 60 degree elevation angle, how far would the projectile reach along the moons surface and
how high above the surface would it reach? First use your brain to set the stage, and
then the MatLab function to determine this value.
A3:
Notes:
Objectives
This lab is designed to 1) acquire real life performance data from a diesel engine in a dynamometer test, 2) to understand the concepts of fuel power (engine input), brake power
(engine output), torque and efficiency.
Introduction
In this lab, we will test a 4.5 liter John Deere 4045H PowerTech diesel engine as shown
in Figure 1. The specifications of this engine are shown in Figure 2, you can download the
complete specs of this engine here.
Procedures
The engine lab consists of two parts, 1) a dynamometer test where we measure the input
and output power of the engine, and 2) a computer lab where the data will be analyzed.
3.1
Dynamometer test
This test can go awry in a hurry unless we take precautions, and run it very systematically.
When we input Torque values, these should be in the range [0, 500]. While inputting RPM
values these should be in the range [1400, 2600]. Imagine what would happen if we key
in 2400 (assuming we are in RPM mode) while we were in fact in Torque mode. The
dynamometer would try to put a load of 2400 Nm on the engine, which would directly stall
it under overload conditions, which could destroy its bearings or worse. To avoid this kind
of potential disaster, we need a procedure and follow it to a tee.
1. Enable the fuel valve
2. Turn on the ceiling fan
3. Turn on the console (on-off switch)
4. Turn on the dynamometer controller (switch on power strip)
1
Figure 1: Here the John Deere 4045 turbocharged diesel engine is shown. It is a 4 cylinder
in-line turbocharged compression ignition (diesel) engine with a displacement of 4.5 liter,
rated at 129kW peak power and 645 Nm of peak torque.
Figure 2: The specs of the John Deere 4045 engine show a maximum (intermittent) power
of 129 kW at 2200 RPM, and a maximum torgue of 645 Nm at 1400 RPM.
5. Turn on the engine: Make sure you have water flow into the dynamometer
6. Engage the dynamometer
7. Hit Control, OnOff (throttle field should become yellow)
8. Gradually increase engine speed to 79% (20% increments) to 2600 RPM
9. Hit Control, OnOff (puts it in RPM mode)
10. TORQUE MODE
(a) Select (on-screen) Torque field
(b) Dial in 100 Nm
(c) Work up to 500 Nm and back down to 100 Nm
(d) Dial in 0 Nm to take the load off the engine
11. RPM MODE
(a) Select (on-screen) RPM field
(b) Dial in 2400 RPM
(c) Work down to 1400 and back up to 2400
12. Select (on-screen) Torque field
13. Dial in 0 Nm to take the load off the engine
14. Hit Control until the throttle field is selected
15. Turn down engine using the throttle back to 20%.
16. Disengage the dynamometer
17. Turn off the engine
18. Turn off the exhaust fan, fuel valve, lights
During the test you need to fill out the data form as shown in Figure 3:
3.2
Data analysis
After we have collected data in the lab, we need to build a spreadsheet that calculates the
Fuel Power in Watt (number of Joules per second flowing into the engine), the Brake Power
in Watt, the resulting overall engine efficiency, as well as the engine efficiency in gram of fuel
per kWh produced. The spreadsheet is shown in Figure 4, you can download it here. The
data is given in the yellow colums A-D, also the Energy density of the diesel fuel is given at
42000 Joule/gram. You need to put equations in the blue columns to calculate the following:
1. Fuel consumption in gram/second: You have to calculate this from the data since
you have the time it takes to consume 100 gram of fuel.
2. Rotational
You need to get this from the RPM, convert
speed
in radian/second:
radian
revolution
to
.
minute
second
3. Brake Power: This is the output power of the engine: You
it
directly
can calculate
rad
Nm
=
T [N m].
from the definition of mechanical rotary power which is P ow
s
s
4. Fuel Power: This is the power flowing into the engine: You need to calculate this from
hg i
J
the fuel mass flow in
, and the heating value (energy density) of the fuel in
.
s
g
Note that the cell in which the energy density is given (D23) needs to remain fixed
(this is called an absolute reference in Excel), so if you refer to it in your equations,
you need to put a dollar sign in front of the 23 to keep it constant while you copy the
equation to other cells. (technically you do not need a dollar sign in front of the D
column, but you can put it there if you prefer.)
5. Efficiency: The overall efficiency of an engine is simply output power (Brake Power)
divided by input power (Fuel Power). If you want you can change the format of all
efficiency cells to a percentage (it just looks nicer).
h gram i
. This is a tricky one, but it is given in the spec sheet
6. Fuel consumption in
kW h
hg i
, and you have output
so we need to calculate it. You have the fuel mass flow rate in
s
h gram i
h gram i
Joule
power in
. Combine these to arrive at
and convert to
. Note
s
Joule
kW h
that a kW h is NOT a measure of power, but rather a measure of energy. Before you
do this last part convert kWh to Joule on paper.
Figure 4: Here a template of the Excel Spreadsheet is shown. If you are in a group that does
the data analysis before the dynamometer test, type in the data from this figure, otherwise
use the data you just collected during the dynamometer test in Figure 3.
Now you need to produce five graphs. Every time you see the word versus (vs) that always
means Y versus X, so in our graphs the RPM is always on the x-axis. Use the scatter plot
graph, with or without a curve drawn through.
1. Torque in Nm vs RPM
2. Brake Power (in kW) vs RPM
3. Fuel Power in (in kW) vs RPM
4. Engine efficiciency in percentage vs RPM
5. Fuel consumption in gram/kWh vs RPM (read kilowatt hour)
Submission
To submit the lab, you need to take a screen shot with the Snipping Tool, showing the data
and the 5 graphs wrapped around it. Make sure it looks good, for instance make sure that
the RPM axis runs from say 1200 to 2600 RPM, not 0 to 2600 RPM. Add the screen shot
to this document and submit.
1. (6) Explain the terms Cracking pressure, Full flow pressure and Pressure
Override. (You can make a graph on the last page if you like)
2. (8) Explain why a hydraulic system with a closed center DCV and a constant
delivery pump is very inefficient from an energy usage standpoint
rad
6 N
s . The pressure upstream is 6.78 * 10 m 2 and the downstream pressure is
m3
N
3.58 * 10 6 2 . The flow rate through the motor is 2 *10 2
m
s
2
6 .7 8
25
3 .5 8
a) Calculate the hydraulic power in the motor, include the unit.
Cylinder 2
p1
p2
The cap end and rod end areas of the two cylinders are equal and the Cap end areas are
twice the rod end areas:
AC1 = AC 2 = AC
(1.)
AR1 = AR 2 = AR
AC = 2 * AR
(2.)
In order to move the pistons to the left, the forces F1 and F2 have to be equal and the
check valve needs to open which requires a pressure of p1 .
c) What pressure do we need to generate at p 2 to move the pistons? Write your answer
as a function of p1 .
6 (6) Draw the symbols for 1) a direct acting PRV and 2) an FCV with temperature and
pressure compensation
PRV
FCV
4
b) Indicate in your graph what area is used to compute the Indicated Mean Effective
Pressure pIME .
c) Show using the units in your chart that the area from question b indicates work.
b) At full power the engine consumes 3.8 kg/s of heavy diesel fuel. The heating value of
this fuel is 41.2 MJ/kg. Calculate the Fuel Power into the engine, include the unit.
d) When the engine is running at full power, the ship can reach a speed of 12 m/s.
Calculate the force (usually called thrust) that the single propeller exerts on the ship at
this speed (assume no loss in energy). Include the unit.
9) (10)
Draw a complete schematic of the following hydraulic arrangement (use the scratch
paper if you need to).
An hydraulic system contains a tank from which the unidirectional gear pump draws
fluid. At the pressure side of the pump, a PRV is connected that diverts flow back to tank
in case the systems pressure rises above 3MPa. The exit line of the pump feeds into the Pconnection of a 4-3 lever operated tandem center DCV with spring return. In addition,
an in-line Pressure Compensated FCV is mounted between the pump exit line and the
P-port of the 4-3 valve. The T-connection of this valve is connected to tank.
At the A and B connections of the valve a dual acting dual rod cylinder is connected.
10) (6) Is this particular DCV a good choice for driving the cylinder? WHY?
PHYD =
PROT
PLIN =
PEL =
328
Bibliography
[1] T. Grift, J. Walker, and J. Hofstee, Aerodynamic properties of individual fertilizer
particles., Transactions of the ASAE, no. 1, pp. 1320, 1997.
[2] T. Grift, Spread pattern analysis tool (spat), part 1. development and theoretical examples., Transactions of the ASAE, no. 6, pp. 13411350, 2000.
[3] T. Grift, J. Walker, and D. Gardisser, Spread pattern analysis tool (spat), part 2.
examples of aircraft pattern analysis., Transactions of the ASAE, no. 6, pp. 13511363,
2000.
[4] ASAE, Standard 341.3 procedure for measuring distribution uniformity and calibrating
granular broadcast spreaders., 1999.
[5] R. Mennel and A. Reece, The theory of the centrifugal distributor, part iii: Particle
trajectories., Journal of Agricultural Engineering Research, no. 1, pp. 7884, 1962.
[6] T. Grift and J. Hofstee, Testing an online spread pattern determination sensor on a
broadcast fertilizer spreader., Transactions of the ASAE, no. 3, pp. 561567, 2002.
[7] G. Kweon and T. Grift, Feed gate adaptation of a spinner spreader for uniformity
control., Biosystems Engineering, no. 1, pp. 1934, 2006.
[8] T. Grift, G. Kweon, J. Hofstee, E. Piron, and S. Villette, Dynamic friction coefficient
measurement of granular fertiliser particles., Biosystems Engineering, no. 4, pp. 507
515, 2006.
[9] T. Grift and J. Hofstee, Measurement of velocity and diameter of individual fertilizer
particles by an optical method., Journal of Agricultural Engineering Research, no. 3,
pp. 229233, 1997.
329