HMT Frequently Asked Questions #3 Convective Heat Transfer Corrections, Boiling & Condensation, and Mass Transfer Fundamentals

From: Wang Chi-Hwa Sent: Wednesday, April 22, 2009 10:29 AM To: Tan Hui Ling Subject: RE: CN2125 Questions Dear Hui Ling: Thank you for your e-mail. Attached please find my response to your question: (1) For example 3, I mentioned during the lecture that the final values for q presented on the top of Page 310 needs some correction because the particular rho (density) and cp values indicated do not correspond to the bulk mean temperature and some correction is required for the Colburn equation 20-28. Equation 19.61 is a general energy balance equation, the reference temperature is dependent on the type of model (correlation) for the heat transfer coefficient. For example 2, it is based on Seider and Tate and hence bulk mean temperature should be used. (2) This case is dominated by forced convection and horizontal or vertical configurations do not matter. I think you are confused between natural convection and forced convection. Natural convection depends sensitively on the configuration (horizontal or vertical) but forced convection depends mostly on the "velocity" imposed to the system. (3) Yes. Please refer to page 339, WWWR, 5th Edition. (4) Part (a) is based on Figure 22.10(b), you can see the solution presented is based on flowing water in the shell (hot) and oil in the tube (cold). This part specifies water to be in the mixed, and hence shell is used. Part (b) is based on Figure 22.9(a). The question specifies the oil being the tube-side fluid and hence water has to be the shell side fluid. The answer is locked in this configuration. The dummy indexes are interchangable - The "h", and "c" do not mean high and low temperatures. In certain cases, the "c" stream can be hotter than the "h" stream. For any further questions, please contact me during my office hour to avoid delay in reply because of the huge numbers of e-mail received during these days.

Office and Phone Consultation Hours: (6) 4:00-4:45pm, 24 April. (7) 10:00-10:45am, 28 April.
Sincerely yours,

Chi-Hwa Wang CN2125

From: Tan Hui Ling Sent: Tuesday, April 21, 2009 1:19 AM To: Wang Chi-Hwa Subject: Re: CN2125 Questions Dear Prof Wang, For Chapter 20 WWWR pg 309 e.g 2: In e.g. 3, you mentioned that the last step invovling h calculationg is based on Tbulk instead. This I agree since the correlation used requires evaluating St at Tbulk. However, in e.g.2, when I subbed values into 19.61, using interpolation from appendix, I could not find out exactly what T the rho and cp values were evaluated at. In a previous email, you said these should be at Tbulk. But I think it should be at Tf instead? Since 19.61 was derived and is supposed to be used with Tf values. Hmwk 5, ID Problem 7.31: The orientation of the plate is such that air flows past a vertical (not horizontal plate). However, the suggested answer uses the correlation for horizontal flat plate. Should the diagram be changed so that the plate is horizontal or should another correlation be used that is applicable for parallel flow across a flat vertical plate? For Chapter 21 WWWR pg 339: If I have a composite of a condenser and a counterflow heat exchanger, then even though there is a phase change haldway, it should not affect my calculations, i.e. I don't have to do something extra to account for the phase change right? Simply continue to use q=UA(long mean T) and solve correct? pg 346 e.g. 2 You mentioned that H and C are dummy indexes and can be interchangeable with t and s. But what if in this question, I decide to set oil to be H and shell; and water to be C and tube? How do I know which I can assign to which fluid? I am quite confused on this part. Thank you. Regards, Tan Hui Ling NUS Undergraduate (Year 2) DDP, Chemical and Biomolecular Engineering & Business 91253274

From: Wang Chi-Hwa Sent: Wednesday, April 22, 2009 10:48 AM To: Huang Zhenbiao Subject: RE: CN2125 Doubts

Dear Zhenbiao: Thank you for your e-mail.

(1) Yes. You are correct. (2) No. Turbulent flow over flat plate, please use Equation 20-34, NuL = 0.036 Re(L)^(4/5) Pr^(1/3). The selection of suitable correlation depends on the particular type of flow. (3) Prandtl and von Karman analogies are more accurate for turbulent conditions. All four are models and we do not differentiate them for validity for either laminar or turbulent flow in general. (4) For the questions covered in the lecture, we were assuming smooth pipe. When there is a known roughness factor, you can use Moody diagram to predict the necessary skin friction coefficient. Sincerely yours, Chi-Hwa Wang CN2125
From: zhenbiao [mailto:u0705513@nus.edu.sg] Sent: Tuesday, April 21, 2009 11:57 PM To: Wang Chi-Hwa Subject: CN2125 Doubts

DearProfessor, Ihavesomedoubtsregardingsomeconceptsonconvection: 1) Whenwederivedtheequation ,weusedtoexactanalysisofthe laminarboundarylayer.Amirighttosaythatthisequationonlyholdsforlaminarflowon flatplate? 2) Ifwewanttofindforothergeometriesandturbulentflowconditions,canweusedto ? approximateintegralanalysis, 3) Inthelecturenotesfouranalogiesareshown,Reynolds,Colburn,PrandtlandvonKarman. AreReynoldandcolburnforlaminarconditionsonly?AnddoweusedPrandtlandvon Karmanforturbulentconditions? 4) Whensolvingthe4analogiesusingskinfrictionfactor,arewesupposetoassumesmooth pipelaw?Sincetheanalogiesarederivedbasedonthefactthatdragiszero? Thanksfortakingtimetoanswermydoubts! Regards zhenbiao


From: Wang Chi-Hwa [mailto:chewch@nus.edu.sg] Sent: Monday, March 30, 2009 10:46 PM To: Ang Zhi Shun Joel Subject: RE: Querry about question 26.4 Tutorial week 8 Dear Joel: Thank you for your e-mail. The concentration of ethanol is actually given in the problem statement according to the two boundary condition. For instance, taking the higher concentration and the definition of 0.1 mole/m^3. You know there is 0.1 mole of ethanol together with roughly 10^6/18 moles of water. Isn't the mole fraction << 5% to satify our criteria for ignoring the bulk motion term? Sincerely yours, Chi-Hwa Wang CN2125

From: Joel [mailto:u0700274@nus.edu.sg] Sent: Mon 3/30/2009 8:50 PM To: Wang Chi-Hwa Subject: Querry about question 26.4 Tutorial week 8

HiSir, Iwouldliketoaskwhybulkfluxforethanolwasnottakenintoconsiderationfortheabove tutorialquestion. FromFicksequationNAZ=DAB.dCA/dz+yA(NAz+NBz), NBziszerosincethemediumofwaterorairisstagnant. However,NAzisnonzero,norisitdilute(sinceitisntmentioned) Thisbeingthecase,whatwouldbetheexplanationfornothavingtoconsiderthebulkmotion ofethanol? Thankyou! Joel
From: Wang Chi-Hwa Sent: Thursday, February 26, 2009 4:22 PM To: Pan Yue Subject: RE: Enquiry about CN2125 boiling correlation Dear Yue: In boiling and condensation, the situations are different.

(1) In boiling, hfg and liquid properties are evaluated at the saturation temperature. Vapor properties are evaluated at the film temperature. (2) In condensation, hfg and vapor properties are evaluated at the saturation temperature and liquid properties are evaluated at the film temperature. Sincerely yours, Chi-Hwa Wang CN2125

From: Pan Yue Sent: Thursday, February 26, 2009 3:00 PM To: Wang Chi-Hwa Subject: RE: Enquiry about CN2125 boiling correlation Dear Sir, In that case, does it mean that all liquid properties in the boiling and condensation chapter (regardless of the equation employed) are evaluated at Tsat? Or is this true only for Eqn 21.8? Thank you very much. Yours sincerely, Pan Yue

From: Wang Chi-Hwa Sent: Thu 26/02/2009 11:58 To: Pan Yue Subject: RE: Enquiry about CN2125 boiling correlation Dear Yue: Thank you for your e-mail. Equation 21.8: hfg and liquid properties are evaluated at the saturation temperature. vapor properties are evaluated at the film temperature. Sincerely yours, Chi-Hwa Wang CN2125

From: Pan Yue Sent: Wednesday, February 25, 2009 9:00 PM To: Wang Chi-Hwa Cc: Pan Yue Subject: Enquiry about CN2125 boiling correlation Dear Prof,

I have some questions about the correlation equations for boiling. Are the parameters (except hfg which is always evaluated at Tsat) normally evaluated at Tfilm? And for equation 21-8, kvf, pvf and miuvf are evalucated at film temperature, so what about density of liquid and vapor, heat capacity of vapor and surface tension of liquid? Please kindly advice on that. Thank you very much. Yours sincerely, Pan Yue -----Original Message----From: Wang Chi-Hwa Subject: RE: Tutorial 7 question Dear Tiak Peng: Thanks for your e-mail. Ya(Na+Nb) ~ 0 because Ya <<1, 1-Ya ~ 1.

For this case, Na is not equal to zero. If it is true, there is no mass transfer at all. Nb = 0 because you can assume the water or air film to be stagnant. Regarding the verification of 1-Ya ~ 1, you can evaluate the respective mole fraction for the liquid and gas phase to be less than 5%. Sincerely yours, Chi-Hwa Wang CN2125 -----Original Message----From: Pang Tiak Peng Subject: Tutorial 7 question Good day sir, I have a question with regards to tutorial 7, question 26.4 of WWWR. In your answer key, your differential form of Fick's equation had the term Ya(Na+Nb) canceled out, where A = Ethanol and B = Water. Is it because Na = Nb = 0 due to the fact that there was no reaction taking place, or is it Nb = 0 because no reaction was taking place, and Ya = 0 since it is diluted and hence the whole term Ya(Na+Nb) becomes zero? I had attended your tutorial on Monday, but could not catch what you have said with regards to this portion. Thank you for your time. Regards, Pang Tiak Peng -----Original Message----From: Wang Chi-Hwa Subject: RE: CN2125 question on flux Dear Remus:

Thanks for your e-mail. To evaluate the molar/mass flux, you need to use N(A)/n(A) where this includes Fick's first law as well as bulk motion contribution. Sincerely yours, Chi-Hwa Wang CN2125 -----Original Message----From: Wong Kang-Ren Remus Subject: CN2125 question on flux Good Afternoon Sir, I have a question regarding chapter 7 on flux. If we are asked to evaluate the molar/mass flux, do we use J(A)/j(A) where this only takes into account Fick's first law OR do we use N(A)/n(A) where this includes Fick's first law as well as bulk motion contribution? And why are they all calssified as Fick's rate equation since only J and j are a result of Fick's law?

Thank You for your attention.

From: Wang Chi-Hwa Subject: RE: Queries Dear Jun Quan: Thanks for your e-mail. In this case, quiescent gas is used to differentiate the gas mixture above the diffuser. In this quiescent gas, there is still non-vanishing NA and NB (These two are reactant and product, respectively). Please take a look at Page 444 (example 2), only the inert gas will not have any molar flux (ND; inert He gas). In brief, there is no mixing in the quiescent gas but there could be non-zero molar flux. Solving mass transfer problem, you combine general mass balance equation for species A and then use the constitutive equation for NA to get the resulting differential equation. Sincerely yours, Chi-Hwa Wang, CN2125

From: Choo Jun Quan Subject: Queries Dear Prof Wang,

With respect to chapter 25 of the WWWR fifth edition, for example 1 on page 443, question given involved a layer of quiescent gas and thus there should be no convective term or bulk motion involved. However, in the solution, for the molar flux, the bulk motion term ie (Ya(Na+Nb) term was still being utilised. As such, should we eliminate this term instead of expressing Nb in terms of Na in this case. And in solving questions, should we also always start off by applying the general differential equation first rather than the Fick's second law as the latter being a simplied form of the general differential equation? Thank you for your kind assistance. With Regards Jun Quan

From: Wang Chi-Hwa Subject: RE: questions regarding heat transfer portion Dear Nicholas: Thanks for your e-mail. This change from (density of liquid) to (density of liquid - density of vapor) is to take into consideration for the buoyancy effect. Equation (8-12) is used to describe the laminar flow in an inclined plate. It was mentioned in the text book that "In the derivation of equation (8-12), the density of gas or vapor at the liquid surface was neglected". Sincerely yours, Chi-Hwa Wang CN2125

From: Lin Yongsheng Nicholas Subject: questions regarding heat transfer portion Dear Prof Wang, I was reviewing the heat transfer portion and I came across something I could not understand. In the section on film condensation, equation (8-12) was derived from fluid mechanics, force balance and then expressed in terms of film condensation thickness in equation (21-12). However, in equation 21-12, the density term in 8-12 was converted to the difference between the liquid density and vapor density. Why and how is this change possible? Thanks and regards, Nicholas -----Original Message----From: Wang Chi-Hwa Subject: RE: Questions regarding CN2125 chap 7

Dear Zhenhao: Thanks for your e-mail. (1) The sum of atomic volumes is only a rough estimate for molecular volume. This is essential when there is no information on the molecular volume. (2) Please find the following search from Google for your reference. Faraday constant From Wikipedia, the free encyclopedia Jump to: navigation, search In physics and chemistry, the Faraday constant is the amount of electric charge per mole of electrons. The Faraday constant was named after British scientist Michael Faraday, and is widely used in calculations in electrochemistry. It has the symbol F, and is related to the charge on an individual electron by F = NAe, where NA is Avogadro's number (approximately 6.021023 mole1) and e is the elementary charge, the magnitude of the charge on an electron (approximately 1.6021019 coulombs per electron). The value of F was first determined by weighing the amount of silver deposited in an electrochemical reaction in which a measured current was passed for a measured time[1]. Research is continuing into more accurate ways of determining F, and thereby NA. [Source: NPL Annual Review 1999] F = 96 485.3383(83) coulomb/mole Sincerely yours, Chi-Hwa Wang CN2125. -----Original Message----From: Hu Zhenhao Subject: Questions regarding CN2125 chap 7 Dear Professor, I have two questions regarding the lecture on week 7. (a) For the Fuller correlation, the diffusion volume is evaluated using table 24.3 (WWWR 4th edition). I notice that the diffusion volumes for simple molecules are not simply the sum of the constituent atomic diffusion volume, thus if we were to evaluate the diffusion volume of a species not given in the table, what should we go about doing so? (b) For Nernst equation, the units for the ionic conductances and Faraday's constant involve g-equivalent. May I know what does gequivalent mean? Thanks for answering my questions.

From: Wang Chi-Hwa Subject: RE: qn on 2125 chapter 6 Dear Guan Qun: The subcooling term in the derivation of vertical plate was used and the correction value is limted. ie. the submcooling term is much smaller than the latent heat term). In the horizontal plate mentioned, only a pseudo-stead solution is used. This is already a rough solution. As a result, the subcooling is not included in this rough solution. Sincerely yours, Chi-Hwa Wang

From: Tan Guan Qun Subject: qn on 2125 chapter 6 Dear Sir For the example 1 in boiling and condensation chapter about condensation on a horizontal plane, how come we do not need to account for the subcooling effect of the condensed liquid (from Tsat to T, the bulk film temp) compared to the case in the vertical plane? Are we assuming that the the condensate film temp is the same as that of the water vapor in the horizontal plane case? During what situations do we need to consider the subcooling effect in heat flux balance? Thanks Guanqun

From: Wang Chi-Hwa Subject: RE: Questions regarding cn2125 Dear Xinling: Yes, this is simply a terminology problem. Sincerely yours, Chi-Hwa Wang CN2125 From: Chen Xinling Subject: Questions regarding cn2125 Dear Sir,

Is arithmetic mean bulk temperature equivalent to bulk mean temperature? Best Regards, Chen Xinling From: Wang Chi-Hwa Subject: RE: Some questions on chapter 21, 22 WWWR (4th ed)

Dear Leslie: (1) Equation 21.20 (derived by Nusselt). Equation 21.22 (derived by Colburn). Both are okay. (2) Equation 21.20. All condensation correlations uses values based on Tf except latent heat term and density of vapor. The latter two are based on saturation temperature. (3) Equation 22.11. This is a special case where delta_T (inlet) = delta_T (outlet). (4). The dummy index is interchangable. The readings are almost the same. (5) Yes, in the calculation for log mean temperature difference, indices 1 and 2 now refers to stations of the heat exchanger. Sincerely yours, Chi-Hwa Wang CN2125 Co-instructor
From: Yeo Qi Fu Leslie Subject: Some questions on chapter 21, 22 WWWR (4th ed) Dear Prof Wang, Please assist me on the following queries that I have. 1. May I know what is the difference between eqn 21.20 (which is for laminar condensation of vert. plate) and eqn 21.22? Eqn 21.22 seems to be also for laminar since it is to be used when Re is below 2000. (critical Re is mentioned to be 2000) 2. All condensation correlations uses values based on Tf except latent heat term? 3. In eqn 22.11, may I know is the delta_T the difference in temperature of the hot streams inlet n outlet or the cold streams inlet and outlet? 4. For eqn 22.12 and 22.13, you mentioned something about the indices being dummy. I understand that either fluid can be C or H but may I know if S must correspond to H, T must correspond to C? it seems like I will get a different answer if I put S C and TH. 5. Lastly, when I am calculating Delta_Tlm for shell, tube exchangers, its on the basis on counterflow, this means that the indices 1 and 2 now refers to stations of the heat exchanger and not the respective fluids inlet n outlet right?

Thank you for taking time to go through these questions. Cheers, Leslie From: Wang Chi-Hwa Subject: RE: possible error? Dear Leslie : Thank you for your e-mail. Attached please find my response to your e-mail.

(1) Yes, you are right. The unit for should be corrected as m^2/s.

(2) On a separate, the rough check of the flow condition can be either inlet or bulk mean. This rough check is imposed at the beginning of solution just to tell whether the flow is turbulent or laminar. Both ways are rough estimates because only the converged bulk mean temperature sets the state of correlation. Ideally, the flow condition does not change from laminar to turbulent during the course of passage and heating (or cooling) through the pipe and only one set of correlation is used throughout the entire problem. Sincerely yours, Chi-Hwa Wang CN2125 Co-instructor
From: Yeo Qi Fu Leslie Subject: possible error? Dear Sir, I refer to example 3 of todays lecture notes.in the calculation of Reynolds number (to check if laminar or turbulent), the publisher has used thought momentum diffusity should be in m2/s.

= 1.478 105 Pas . Why is the units in Pa.s? I

Another question, what temperature is this Re number (to check laminar or turbulent, even before any correlation is chosen) calculated based on? My guess is the bulk mean temperature of air, which is 290+360/2 = 325K. At 325K, according to appendix I for air, under the column of , I got 1.8 x 10^-5 m2/s using interpolation. So which is correct? Thank you, Cheers, Leslie -----Original Message----From: Wang Chi-Hwa Subject: RE: Question regarding tutorial problem 19.12 Dear Tanny: Thank you for your e-mail. As communicated in tutorial session, film temperature is a more popular choice for boundary layer theory derivations (as in this problem). Both bulk mean temperature and film temperature are used for heat transfer correlations. Either temperature can be used for illustration on the development of boundary layer quantities in this question. Both are okay. Chi-Hwa Wang

-----Original Message----From: Barry Tanny Subject: Question regarding tutorial problem 19.12

Dear Dr. Wang, Would you mind explaining why in solving question 19.12 Tutorial 4 fluid temperature is used to evaluate fluid properties instead of film temperature as stated in notes/textbook? Thank you for your attention. Regards, Barry T. From: Wang Chi-Hwa Subject: RE: Question regarding WWWR Chapter 20 example 2 and homework question 20.28 Dear Wei Lun: Thank you for your e-mail. Yes, equation 19-16 is based on only one thermal resistor in the radial direction. In solving both example 2 & Question 20.28, you can combine the two thermal resistors in the following way: In example 2, hi <<ho (i.e. ho ~100 hi), so 1/U~1/hi+1/ho ~ 1/hi. In question 20.28, hi ~ho (comparable scale), you can evalaute the overall (lumped) 1/U ~1/hi+1/ho. Sincerely yours, Chi-Hwa Wang From: Ng Wei Lun Subject: Question regarding WWWR Chapter 20 example 2 and homework question 20.28 Dear Dr. Wang, I have a question regarding WWWR Chapter 20 example 2 and homework question 20.28. Both questions involved solving using equation 19-16 from the textbook. However, in example 2, the h in the equation is defined as the heat transfer coefficient of the interior fluid, h(interior). The h(exterior) of value 11400 was not used in solving the question. While in the solution that you've provided for homework question 20.28, the h is defined as 1/ {1/h(interior) + 1/h(exterior)}. Can you please englighten me about why the h in equation 19-16 are defined differently in the 2 questions?

Regards, Wei Lun

From: Wang Chi-Hwa Subject: RE: Convection Correlations

Dear Zewen: Thank you for your e-mail. (1) Yes, Whitaker correlation for convective heat transfer coefficient has only week dependence on the solid phase (s, the surface temperature of solid). You will see one ChE experiment on this. (2) From Whitaker correlation, you can find "h". Subsequently, this 'h" will affect the transient temperature of the solid sphere. (3) Yes, correlations are best fitting to existing experimental data. Not exact solutions. Sincerely yours, Chi-Hwa Wang From: Zhang Zewen Subject: Convection Correlations Dear Sir, I noticed that many of the correleations taught in the previous lecture do not take in to account properties of the object the convection is passing through.Example for Whitherket correlation, the equation does not have any properties of the sphere in it (eg density of sphere, thermal diffusivity of sphere etc). Compared with Chapter 3, unsteady conduction, we often consider the material used. Does this means the correlations are not that accurate, and are only approximations (since they are empirical anyway)? Thanks Zewen

From: Wang Chi-Hwa Subject: RE: questions on convective heat transfer (chapter 4)
Dear Puay Fang: Thank you for your e-mail. Attached please find my answer to your questions. (1) As communicated in the tutorial session, you might choose either film temperature or bulk mean temperature for the answer. The objective of the question is the development of various quantities with increasing x. In general, convective heat transfer coefficient prediction based on theory is mostly by film temperature while experimental correlations are based on either film temperature or bulk mean temperature.

(2) This question just a proof that different orders for models will give us different coefficients. The derivation (postgraduate level material) is used as a tool for understanding the subject. Do not worry about the detailed steps. (3) Yes, please use either interpolation or extrapolation from the standard sources (WWWR and ID). (4) Yes, please use the film temperature for integral analysis. Sincerely yours, Chi-Hwa Wang

From: Heng Puay Fang Subject: questions on convective heat transfer (chapter 4) Dear lecturer, I have a few queries on chapter 4. They are as follows: 1) (tutorial 4 problem 19.12) In this example, we used the given air temperature (310k) to obtain the values of Pr, k etc from the appendix. Is the temperature given the bulk mean temperature? Why is the bulk mean temperature used, instead of the film temperature? Is there a standard for what temperature to used in questions? 2) (tutorial 4 problem 19.15) I have some difficulties in understanding the solution to this question. Is it required for us to know how to carry out the derivation, as presented? Or is this question just a proof for us that different orders for models will give us different coefficients? 3) (tutorial 4 problem 19.2) For glycerin, the data provided in WWWR book stops at 100 degree fahrenheit. Are we suppose to extrapolate to obtain the data required? What is the method used to obtain the answer? 4) (lecture 4 notes) In applying the results of the exact analysis of the laminar bounadary layer, it is costomery to use the film temperature (as stated in the lecture notes). For the approximate integral analysis, we are suppose to use the film temperature too? (It is not specified in the lecture notes). Thank you for clarifying my doubts!! Regards, Puay Fang ___________________________________________________________________ From: Wang Chi-Hwa Subject: RE: Homework 4 Dear Attia:

Thank you very much for your e-mail. This equation [dT/dx = 2(q/A)/(rho*V*Cp*D(1)) = c(q/A)] comes from energy balance in a small control volume in the channel. You can see this by the following: (i) dT*rho*V*Cp*D(1) = energy gain by fluid after passing through the control volume. [D = distance between the two plates] (ii) 2 (q/A) * (dx*1) = heat transfer from the "two" plates. Please note D(1) = cross sectional area for fluid flow (per unit width in the transverse direction) 2*(dx*1) = area for convective heat transfer from the two plates. If you have any further question on this, please feel free to visit me to clarify on this. Sincerely yours, Chi-Hwa Wang

From: Amalina Bte Ebrahim Attia Subject: Homework 4 Hello. I have a question regarding Problem 19.8(WWWR) of Homework 4. In the solution given, dT/dx = 2(q/A)/(rho*V*Cp*D(1)) = c(q/A). I am wondering where does this equation come from? I thought we have to use equation 19-15 which will reduce v_x dT/dx = 0. Or is another form of energy balance equation used? Thank you.

----- Original Message ----From: LIM YUEH YANG Subject: Re: Re: CN2125 _ question for Eqn 20-27

Dear Sir, As discussed in the previous email, for Example 3 found on Pg. 325 of WWWR textbook, I have made an attempt to evaluate the convective heat transfer coefficient (h) at the bulk mean temperature (Tbulk = 325K), instead of the film temperature (Tfilm = 349K). I am not sure if they are correct. The suggested workings are as follow: St = 0.00428 (obtained in previous working) v = 24 m/s (given) Based on Appendix I, Pg. 726 of WWWR, Specific Heat Capacity at constant pressure, =( )

= 1007.6 Density of air, = = 1.08695 Thus, h = = 1.08695 = 112.5

Yours Truly, Yueh Yang ----- Original Message ----From: Wang Chi-Hwa Subject: RE: CN2125 _ question for Eqn 20-27

24

1.0076

0.00428

(i.e. 7% different from value evaluated at Tfilm)

Dear Yueh Yang; Thank you for your e-mail. (1) The St should be evaluated at bulk mean temperature based on the correlation condition. (2) Yes, you are right, the "h,rho,v,cp" should be evaluated at the bulk mean to evaluate St. This condition is used in the calculation for "h". This means you should correct for the mistakes for rho and cp values in example 3. (3) Other students raised a similar question on this earlier. They told me that the last two lines of WWWR Page 325 should be corrected. Can you calculate and tell me how different is the h value obtained? (From 105 W/M^2-K to ?)
Sincerely yours, Chi-Hwa Wang

From: LIM YUEH YANG Subject: Re: CN2125 _ question for Eqn 20-27 Dear Sir,

Sorry to trouble you. I wish to seek clarification for Eqn 20-27 of the WWWR book. As mentioned in the book, it says that St is evaluated at T(bulk), while Re and Pr used are to be evaluated at T(film). However, since St is expressed as Re and Pr, how come St can be evaluated at T(bulk)? Further more, for the Example 3 (Week 5 Notes), I realise that when evaulating h (using St, rho, v, and cp), the values of rho and cp are interpolated at 349K (which is the film temperature for that example) instead of T(bulk) of 325K. As such, St can be expressed in terms of "h,rho,v,cp" method OR "Re,Pr" method, but all these seems to be evaluated at T(film). Hence, may I enquire where does the T(bulk) comes into the picture for the evaluation of St?

Thanks and Regards. I apologise for any inconvenience caused.

Yours Truly, Lim Yueh Yang CN 3A From: Wang Chi-Hwa Subject: RE: Queries on Tutorial 7 Dear Kelvin: Thank you for your e-mail. (1) The Lenards Jones parameters ( and ) for HCl (a polar coumpound) are both given in Appendix K (page 743). The correction to by equation 24-45 (WWWR) has already been taken into consideration in the preparation of Appendix K by the authors. (2) The correction to D, on the other hand, depends on the value of delta for both SiCl4 and HCl. Since for SiCl4 (nonpolar) = 0, AB = (A * B )^0.5 = 0, then D = D0 (to be found in

Appendix K as per normal cases). Sincerely yours, Chi-Hwa Wang

From: Lim Jincong Kevin Subject: RE: Queries on Tutorial 7

Hi Dr Wang, I attended your tutorial class today and have some queries on tuorial 7 q24.10 (WWWR). 1)I would like to ask do we really need to do the correction in order to get the omega because in your suggested solution, there was no correction.

2) If no correction was done (as shown in the suggested solution), does it mean we can make do with the Hirchfield equation without correcting the omega value for polar compunds? Thanks Regards, Kevin
-----Original Message----From: Wang Chi-Hwa Subject: RE: Chapter 20 questions

Dear Nevin: Thanks for your e-mail. (1) Fluid is inside the enclosure. By conduction only, q = -k T/ x=-h (T). Nu = hL/k=[k/x]*L/k; For spheres: WWWR Page 316 NuD = 2 + 0.43 RaD(1/3) L= x -> Nu=1. Ra -> 0, NuD = 2. h T = k |dT/dr| = k* T/(D/2) (2) Eq 20-10. -> h D/ k = 2 = NuD
(3) Yes. (4) EQUATIONS 19-25, 26: LAMINAR FLOW. EQUATION 20-31& 32 -> TURBULENT FLOW. (5) Yes. Chi-Hwa Wang -----Original Message----From: nevin nar [mailto:nevnahh@yahoo.com] Subject: Chapter 20 questions Hot

Conduction only cold

Dear Prof, I've got a couple of ques to ask. Pls bear with me yar? Thks. 1) For rectangular enclosures, is the fluid inside the enclosure or flowing past it? Also, why izzit that for case 1, when teta=180 degrees (upper surface is heated), heat transfer is by conduction ie Nu=1? 2) For tut 5 q1, which eq is better to use? izzit eq 20-10 or eq 20-11? This is for the case of horizontal cyclinders. 3) For eg 2 on pg 324, are we assuming steady state for heat transfer?

4) On pg 326 of WWWR, for the case of flow parallel to plane surface, they apply colburn analogy for turbulent flow in boundary layer. Why isn't Reynold's analogy used since on pg 303, they mentioned that for turbulent flow, Pr is unity and Reynolds analogy applies in the absence of form drag? 5) Lastly, I think the Cp values for tut 5 q 20.26 is wrong. The values should be multiplied by 1000. Thanks for your time. Rgds, Nevyn CN 5

-----Original Message----From: Wang Chi-Hwa Subject: RE: assignment Q19.20 Dear Guangxing: Thanks for your e-mail. This depends on the correlations used.

(1) WWWR page 306: "Film temperature" is used for Reynolds and Colburn Analogy for heat transfer properties estimation. (2)Page 324 Bulk mean temperature is used for Sieder-Tate relation in determining heat transfer coefficient. (3) Page 325 top portion, Re is first calculated using entrance temperature for rough estimation. The authors have changed the value of Re subsequently in applying Equation 20-27 for calculation of the outlet temperature. Revised Re = 0.0127*24/0.0000205=14868. The correlation of Eq 20-27 requires that ReD and Pr be evaluated at film temperature and St be evaluated at bulk temperature (WWWR Page 323). Sincerely yours, Chi-Hwa Wang -----Original Message----From: Li Guangxing Subject: assignment Q19.20 Dear AP Wang, For Q19.20 , are we supposed to evaluate the Re No and fluid properties at the bulk mean temp or the film temp? Why is it that Chap 19 eg 1 uses film temp but Chap 20 eg2 uses the bulk mean temp for 2 similar egs on internal flow? Please advise me. Thank you.

Regards -----Original Message----From: Wang Chi-Hwa Subject: RE: question Dear Kean Leong: (i) The values of Hfg and Hfg' are close when the heat effect for sub-cooling is not significant. This is true when the excess temperature (Ts-Tsat) is not large. However, the subcooling effect is still included in the film condendation calculation for thickness and heat transfer coefficient (i.e. WWWR 21-18) & (21-19). (ii) The clarification for this point (0.68 vs 3/8) is explained clearly in WWWR 347, the last four sentences above example 1. "Rohsenow performed ....". (iii) Yes. (iv) Horizontal plate problem should be pseudo- steady-state rather than steady state because the film thickness change with time. Example 1 (WWW347 gives you good illustration). This is different from horizontal cylinder because the later case, fluid in the film will flow away. (v) When either p or R is zero, from Figure 11.10- 13, you can see the value = F ~ 1. Sincerely yours, Chi-Hwa Wang CN2125 Instructor

-----Original Message----From: Ch'ng Kean Leong Subject: RE: question

Dear Dr Wang, I have some doubts and questions on boiling and condensation. Kindly help me to clarify. Thank you. (i) As suggested by ID, pg 619, eqn (10.33), m = q/Hfg'... where Hfg' is given in eqn (10.2) as Hfg'=Hfg+0.68Cp(Tsat-Tw)..... --shouldn't this be more appropriate than using m=q/Hfg... when cooling effect is neglected? (ii) The formula Hfg'=Hfg+0.68Cp(Tsat-Tw) is given in WWWR also, page 347, last paragraph. For exam purpose, should we 3/8 or 0.68? (iii) In example 1 pg 347, are we neglecting the cooling effect in order to simplify calculation?

(iv) To determine h for horizontal plate, can the formula for horizontal cylinder be used? since for vertical plate and vertical cyclinder the formula is identical.... (v) Could you kindly explain the paragraph in ID pg 655 which reads like this: "Heat exchanger behaviour is independent of the specific configuration. Such would be the case if one of the fluids underwent a phase change " Thank you so much for your time and attention. Sorry for refering u to various pages here and there..... ;o) regards kean leong.

CN2125 Heat and Mass Transfer Quiz #1 (March 1, 2010) Venue: MPSH5. Time: 10:00-10:50am. Covering Range: Steady and Unsteady Heat Conduction + Convective Heat Transfer Correlations Student Name: Tutorial Group: _____ Matriculation Number:__________ Yr:_______

1. A 1-mm-thick copper plate is sandwiched between two 5-mm-thick epoxy boards that are 15cm x 20 cm in size. Determine the %Error involved in the total thermal resistance of the plate if the two thermal contact conductances are ignored. The thermal conductivities are given to be k = 386 W/mC for copper plates and k = 0.26 W/mC for epoxy boards. The contact conductance at each of the interface of copper-epoxy layers is given to be hc = 6000 W/m C.
2

Hints:

(i) Assumptions to be made are given by: 1 Steady operating conditions exist. 2 Heat transfer is one-dimensional since the plate is large. 3 Thermal conductivities are constant. (ii) The thermal resistances of different layers for unit 2 surface area of 1 m are to be calculated. (iii)The total thermal resistance for the five-element thermal circuit is given by Rplate+2Rcontact+2Repoxy (iv) %Error = 2 Rcontact/Rtotal *100% .

Give your brief answer here: Analysis: the thermal resistances of different layers: 1 1 1 Rcontact = = = 0.00017 oC / W 2 o 2 hc A (6000W / m C )( Am ) A L 0.001m 1 R plate = = = 2.6 10 6 oC / W 2 o 2 kA (386W / m C )( Am ) A L 0.005m 1 Repoxy = = = 0.01923 oC / W 2 o 2 kA (0.26W / m C )( Am ) A The total thermal resistance is
Rtotal = 2 Rcontact + R plate + 2 Repoxy

1 1 (2 0.00017 + 2.6 10 6 + 2 0.01923) = 0.03880o C / W A A Then the percent error involved in the total thermal resistance of the plate if the thermal contact resistances are ignored is determined to be 2R 2 0.00017 % Error = ( contact 100)% = ( 100)% = 0.88% 0.03880 Rtotal Which is negligible. =

2. In a manufacturing facility, 2-inch-diameter aluminum balls initially at 250oF are quenched in a water bath at 120oF for a period of 2 min at a specified rate 120 balls per minute. If the convective heat transfer coefficient is 42 Btu/h.ft2.oF, determine (a) the temperature of balls after quenching and (b) the rate at which heat needs to be removed from the water in order to keep its temperature constant at 120oF. The thermal properties of the balls are constant. The heat transfer coefficient is constant and uniform over the entire surface. The thermal conductivity, density, and specific heat of the aluminum balls 3 are k = 137 Btu/h.ft.F, = 168 lbm/ft , and cp = 0.216 Btu/lbm.F Hints: (a) What is the value of Biot modulus (Bi)? Is Bi <<0.1? (b) Can the expression T T = e BiFo (WWWR Eq. 18-9) be applied to determine the temporal variation for T0 T temperature of the aluminum balls? Give your brief answer here:

Question 2

(A) By the Method of Log Mean Temperature Difference

LMTD =

(T2 T ) (T1 T ) (250 120) (152 120) = = 70 T2 T 250 120 ln ln 152 120 T T 1

qTotal = hA(2)(120)( LMTD ) = (42)(4 )(

1 2 ) (2)(120)(70) = 61575 Btu / hr 12

(B) By m*Cp*T

qTotal = 1034(60) = 62040 Btu / hr

3. Determine the convection heat transfer coefficients for the flow of air and water (both to be heated) at the velocity of 2 m/s in an 8-cm-diameter and 7-m-long tube when the tube is subjected to uniform heat flux from all surfaces. Use the fluid properties at 25oC. The properties of air at 25C are k = 0.02551 W/m.oC, = 1.562 x 10-5 m2/s, Pr = 0. 7296. The properties of water at 25oC are = 997 kg/m3, k = 0.607 W/moC, = 8.937 x 10-7 m2/s, Pr = 6.14. You may make the following assumptions: (i) Steady flow conditions exist. (ii) The surface heat flux is uniform. (iii) The inner surfaces of the tube are smooth. Please verify that both flows are turbulent and the Dittus and Boelter equation (WWWR 5th edition, Equation 20-28, NuD = 0.023 ReD0.8 Pr0.4) can be used to predict the heat transfer coefficients for the two respective cases. Give your brief answer here: Question 3 (a) for air, the Reynolds number is (2m / s )(0.08m) VD Re = = = 10,243 v 1.562 10 5 m 2 / s Which is greater than 10,000. Therefore, the flow is turbulent and the entry lengths in this case are roughly Lh Lt 10 D = 10(0.08m) = 0.8m which is much shorter than the total length of the tube, therefore, we can assume fully developed turbulent flow in the entire duct, and determine the Nusselt number from hD Nu = = 0.023 Re 0.8 Pr 0.4 = 0.023(10,243) 0.8 (0.7296) 0.4 = 32.76 k Heat transfer coefficient is k 0.02551W / mo C h = Nu = (32.76) = 10.45W / m 2 o C D 0.08m (b) Repeating calculations for water: (2m / s )(0.08m) VD Re = = = 179,035 8.937 10 7 m 2 / s v

hD = 0.023 Re 0.8 Pr 0.4 = 0.023(179,035) 0.8 (6.14) 0.4 = 757.4 k k 0.607W / mo C h = Nu = (757.4) = 5747W / m 2 o C D 0.08m Discussion: The heat transfer coefficient for water is 550 times that of air. Nu =

Grader's Remark (Qiao Jian, g0800017@nus.edu.sg; 6516-1536) Question 1: About two third of the students got the full mark of this question. The mistakes are as shown follow: some students calculated the area of side view instead that of cross section.; the thermal resistance of the contact part should include the boundary of both side of the copper plate, some students only calculated one side. Question 3: Most of the student got the right Re number of water and air, only some students made the mistakes of treating the as viscosity of air and water, some others use the length of the tube to calculate the Re number instead of diameter. About two thirds of the student write the equations of calculating the Nu number, but only half of them then got the right value of Nu and then used the Nu number to calculate the heat transfer coefficient. In the last step, when calculating heat transfer coefficient from Nu, some of the student also use the length of the tube length instead of the diameter. Graders comments: (Alireza Rezvanpour; email: rezvanpour@nus.edu.sg; 65167161) Question 2: In the first part of the question, most of the students knew how to use the equations to solve the problem, but nearly one fourth of the students got the wrong answer when calculating the Bi number and T(t). In the second part of this question, about half of the students used hAT to calculate the amount of heat transfer between the balls and water. During this process, the temperature of balls keep changing, so this method is not suitable, instead of hAT, mcpT can be used to solve this question.