ASME 2000 Citrus Engineering Conference

CEC2000
March 23, 2000, Lake Alfred, Florida, USA
CEC2000-4605

SANITARY WELDING STANDARDS

BY
Richard E. Avery, P. E.
Consultant to the
Nickel Development Institute
Londonderry, NH

INTRODUCTION
Austenitic stainless steel tube and pipe systems are a vital part of today's sanitary
(hygienic) processing facilities. Product contact surface welds in the tubular systems
usually cannot be ground or conditioned, so it is essential that the as-welded surfaces are
suitable for cleaning-in-place (CIP).

The American Welding Society AWS D l 8 Committee was formed in respond to the
request by the 3-A Sanitary Standards Committee for help in preparing welding standards
for use in the manufacture and construction of dairy and food product processing plants.
The 3-A Sanitary Standards Committees develop and promulgate sanitary design stan-
dards for dairy and food processing, packaging and handling equipment and systems.

Published with permission.

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 AWS D l 8.1 :1999 Specification for Welding of Austenitic Stainless Steel Tube and Pipe
Systems in Sanitary (Hygienic) Applications developed covers the requirements for gas
tungsten arc welding (GTAW) or TIG welding as it is also known, of austenitic stainless
steel tube and pipe 114 in. (6 mm) diameter and larger. Sanitary processing systems is
intended to include those systems handling products for human and animal consumption.
Such products include dairy, meat, poultry, vegetable, beverage, and other products
consumed by humans and animals.

The paper also includes design and operation guidelines that have proven useful to
engineers and users of austenitic stainless steel piping systems.

AWS D l 8 COMMITTEE ON WELDING IN

SANITARY APPLICATIONS
The committee is comprised of members representing producers of hygienic equipment,
users of hygienic equipment, and general or public interest for the dairy and food industry.
It was initially decided to limit the first document to the welding of tube and pipe systems,
which was identified by 3-A as the more urgent issue. Welding, of course, is used
extensively in the manufacture of a wide range of food equipment components such as
tanks, vessels, pumps, and valves, to name a few. However, most welds in such equip-
ment are accessible to visual inspection, and the weld surfaces can be ground and finished
to meet cleanability requirements. The D l 8 committee is now in the process of developing
AWS D l 8.3, Specification for Welding Vessels and Equipment in Sanitary (Hygienic)
Applications. Technical committees such as D l 8 are not closed groups and there is a need
for balanced representation. It is most difficult in obtaining user representatives because
they see equipment construction as outside their corporate interest. AWS encourages all
participants, but needs users for committee balance.

It was an early decision that the weld quality standards in the specification should be in
agreement with the current practices used by sanitary weld inspectors. One of the commit-
tee members arranged to make 36 weld samples in 2 in. OD, 0.065 in. wall, Type 304 or
Unified Numbering System (UNS S30400) tubing. The welds were all manual Gas Tung-
sten Arc Welds (GTAW) made without filler metal addition. A number of welds were
purposely made to include ID weld defects. The inside and outside surfaces were left in the
as-welded condition. The welds were "round-robin" examined by six welding inspectors
recruited from a list supplied by the 3-A Sanitary Committee Secretary and then examined
by the D l 8 committee. The inspectors were asked to first judge weld acceptance based
only on the outside weld surface, as would be the case of a field weld where the inside was
inaccessible for a visual examination. After the outside examination, the inside weld
surface was judged and reported.

Significant observations and results were:

There was usually 75% or greater agreement among the inspectors on weld
acceptance or non-acceptance
The inspection of the outside weld surface in most instances agreed with the

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 subsequent inside surface examination. In other words, an experienced welding
inspector can usually determine if the inside weld surface is acceptable by exam-
ining the outside.
From the exercise, the committee was able to identify the need for further work to
establish the final acceptance standards.

AWS D18.1:1999
D l 8.1 :I999 addresses welding qualification and visual examination requirements prior
to postweld conditioning. The specification does not cover the "how-to-do-it" aspects of
making small diameter tubular welds in austenitic stainless steels. Welding procedure
details are covered very completely in other documents. (1)

QUALIFICATION
Traditionally, welding qualifications have been generated for particular industries or end-
use applications such as high-pressure piping systems. Two widely use qualification
standards are ASME Section Vlll and AWS B2.1. (2), (3) The food industry has been lax
with regards to welding qualification requirements, often allowing welding qualification
details to be the option of the fabricators or constructors. Admittedly, food industry process
equipment welds do not represent the safety risk associated with high-pressure piping
welds, but some degree of formalized procedures and documentation can be very benefi-
cial to the owners. The D18.1 specification addressed Procedure Qualification and Perfor-
mance Qualification and introduces a new concept of a Preconstruction Weld Sample
(PWS). The plant owners or their representatives have the option of imposing all or none of
these items.

PROCEDURE QUALIFICATION
D18.1 requires that a written qualified Welding Procedure Specification (WPS) be made
for each type of weld used in a sanitary plant application. The WPS can be in any format
provided all the welding variables are recorded. Usually it is most convenient to use the
format of ASME Section Vlll or AWS 82.1. A detail description of welding variables is
beyond the scope of this presentation, but briefly they include such items as:

Joint design - a change in groove type or increase in root opening

Base metal - a significant change in base metal thickness

Filler metals - the addition or deletion of filler metal or the type of filler metal
Shielding gas - a change of torch shielding or backing gas
Electrical characteristics - a change in current type or in pulsing current

Test welds are then made following the written WPS and the welds examined by visual
and mechanical tests. The visual examination is made using the acceptance criteria to be
described later in this paper. Weld mechanical tests are tensile, root bend, and face bend
tests made in accordance with ANSIIAWS B4.0, Standard Methods for Mechanical Testing
of Welds.

PERFORMANCE QUALIFICATION

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 The Performance Qualification is simply a test of the welder's ability to make acceptable
welds in accordance with the qualified WPS. The test welds are evaluated by a visual
examination and by bend tests. The Performance Qualification remains in effect indefi-
nitely unless the welder is not engaged in the welding process qualified, e.g., the GTAW
process, for a period exceeding six months, or there is reason to question the welder's
ability.

PRECONSTRUCTION WELD SAMPLES (PWS)

One of the main objectives of the D18.1 specification is to provide guides to evaluate the
inside weld surface (when the weld is not accessible for an internal examination) by
examining the outside weld surface. Welders have different characteristic "signatures,"
that is the weld surface features often vary from welder to welder. To record the welder
variability, three PWS welds are prepared by each welder. When the outside and inside
weld surfaces of each weld meet the agreed-upon acceptance standards, one PWS is
given to the welder, one to the welding supervisor, and one to the Owner's representative
for reference when there is an acceptance question on a production weld.

VISUAL EXAMINATION REQUIREMENTS

The specification requires that the outside surfaces (non-product surfaces) of welds
should be visually examined at the start, the termination, and the width of the weld should
be consistent with each welder's PWS. Welds that do not meet the acceptance criteria by
external examination are to be examined internally with a borescope or other suitable
device and meet the below acceptance standards.

VISUAL EXAMINATION ACCEPTANCE CRITERIA FOR ALL WELDS

The acceptance criteria for the internal and external surfaces of all welds are:

Welds shall be full penetration.

Welds shall not contain cracks, undercut, crevices, pits or embedded or protruding
material.
Offset or misalignment shall not to exceed 10% of the wall thickness as shown in
Figure 1.
VISUAL EXAMINATION ACCEPTANCE CRITERIA FOR
NON-PRODUCT CONTACT SURFACES
These criteria are developed to provide confidence that the inside weld surface is
acceptable without making an internal examination. The values were developed based on
the "round-robin" examination described earlier and the experience of committee mem-
bers.

Maximum allowable concavity shall be 0.006 in. (0.1 5 mm), as shown in Figure 2.
Maximum allowable convexity shall be 0.012 in. (0.3 mm), as shown in Figure 3.

Manual welds shall have a minimum weld face of 2T where T is the tube wall
thickness, as shown in Figure 4.

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 The weld face shall be uniform in width. The minimum face width shall be at least
75% of the maximum face width, as shown in Figure 5.
The start, termination, and width of the weld shall be consistent with the welder's
PWS.
VISUAL EXAMINATION ACCEPTANCE CRITERIA FOR
PRODUCT CONTACT SURFACES
The acceptance criteria for the inside weld surface follows:

Maximum allowable concavity shall be 0.012 in. (0.3 mm). (The outside concavity
of the weld results in a convex condition on the inside weld contour. There is a
greater tendency for the inside of the weld to "sag" causing a high convex contour,
hence the limit of 0.006 in. [0.15 mm] on the outside surface.)
Maximum allowable convexity shall be 0.012 in. (0.3 mm).

Oxide islands greater than 1116-in. diameter shall be unacceptable. No more than
four oxide islands shall be present in any weld. See Figure 6.
The weld surface shall not contain excessive heat-tint oxidation or discoloration

It is not possible to adequately describe and convey various degrees to heat tint or weld
discoloration verbally, although many welding specifications attempt to do so. AWS D l 8.1
and a special version, AWS D18.2:1999, have a color photograph showing ten different
degrees of heat tint. A black and white version is shown in Figure 5, but a color photograph
is needed to depict the different degrees of oxidation.

A word of how the sample was made: Ten full penetration autogenous welds were made
on the outside of a 2-in. (50 mm) 316L stainless steel tube. Welds on 304L tubing showed
no significant difference from 316L. Oxygen was added to the pure argon backing purge
gas to obtain the oxygen levels shown in welds No. 1 through 10.

The illustration should be used as a reference to identify the degree of heat-tint oxide by
number and not to specify oxygen limits in the backing gas. The acceptable degree of heat
tint can vary with different service environments. The cost involved in obtaining very low
levels of heat tint should be considered when specifying such levels.

The amount of heat-tint oxide and its appearance can be influenced by factors other
than oxygen, such as:

High levels of moisture in the backing gas can increase the degree of heat tint.
Contaminants, such as hydrocarbons, moisture, and some types of particulate on
the surface prior to welding, can affect discoloration levels.
Hydrogen in the argon backing gas can significantly reduce the amount of heat-tint
oxide.
- The metal's surface finish can affect the appearance of heat tint.

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 OXIDE ISLANDS AND HEAT-TINT OXIDE
Oxide islands may occur on either the weld face or the inside surface of stainless steel
GTAW welds. The color may vary from gray to a very dark color. The islands are usually
tightly adhering and would not become detached in normal service conditions. The welder
or welding operator usually has little influence over the formation of oxide islands through
technique variables. When analyzed, the islands are usually oxides high in silicon, calcium,
and similar elements. Some base metal heats seem to be more prone to the formation of
oxide islands that other heats. In most service applications, including the food industry,
occasional oxide islands do not present a weld quality problem.

Heat tint is an oxide, primarily of chromium, that forms on the weld and the heat-affected
zone in the presence of oxygen. The discoloration may vary from straw yellow, deep
golden, rose, blue or gray, to almost a black color. When the heat tint is tightly adhering, it
does not break off to cause a product contamination problem. In some environments such
as those involving chlorides or unclean water, the corrosion resistance is reduced in the
heat-tint area and pitting or crevice corrosion is likely.

Heat tint can be avoided by limiting oxygen in the backing purge area of pipes to very
low levels, e.g., 25 ppm of oxygen or less. This can consistently be accomplished in orbital
welding where very accurate joint fit-ups are required and high purity purging can be
obtained, but it is difficult with normal manual welding conditions.

Heat tint can be removed by an acid pickling treatment, electropolishing, or mechani-

cally by grinding, sanding or abrasive blasting. The most effective pickling solution to
remove heat tint is a nitric-hydrofluoric acid solution. ASTM A380 and other documents
provide details on pickling of stainless steels to remove heat tint, embedded iron, and other
surface defects (4) (5). There are also commercially available nitric-hydrofluoric pastes
that are easy to use on boldly exposed surfaces, but in most instances could not be used
inside small diameter tubes. Ammoniated citric acid is fairly effective on light heat-tint, and
can be made more effective by an ammonium bi-fluoride addition (6). Nitric acid solutions
are standard stainless steel passivation solutions, but they are not designed to remove
heat tint, deeply embedded iron particles, and other surface defects that are often intro-
duced during fabrication. With any type of acid treatment, it is essential to completely
remove the acid by a thorough water rinse.

When heat-tint is removed mechanically, it is essential to use grinding disks or abrasives

that have never been used on carbon steel in order to avoid embedding iron particles on
the stainless steel surface. Free iron particles will corrode, causing rust spots to form, and
give the appearance that the stainless steel is rusting. References (4) and (5) provide
details on the removal of free iron. Wire brushing, even with stainless steel wire brushes,
distorts and disturbs the surface, causing it to have lower corrosion resistance. Wire
brushing is not nearly as effective as a fine grit abrasive treatment to remove heat tint or
other surface defects.

DESIGN AND OPERATING CONSIDERATIONS

Welding is only one factor that goes into making a stainless steel piping system that will
give good service. It is well recognized that stainless steels have excellent corrosion

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 resistance in many environments, particularly in the food industries where cleanability is
also of primary concern . However, it may be less well recognized that stainless steels
perform best when the surface is kept clean and free of deposits. High velocity flow rates,
as high as 100 ftlsec or higher, are beneficial because they maintain clean surfaces.
Conversely, stagnant conditions have been known to contribute to corrosion.

Basic design guides for stainless steel piping systems include:

Slope horizontal lines so they drain completely.

Avoid dead legs and sections that cannot be drained during shut-down or stand-
by.
Provide inspection or wash-out ports on horizontal runs to allow flushing of sedi-
ment.
Design for maximum flow rate consistent with pressure drop to reduce sediment
deposits.
At some stage in the construction or operation of food plants with stainless steel
components, ordinary water is involved. This may be water used for hydrostatic testing,
system flushing or cleaning, or other operational processing. Therefore, it is useful to
summarize some of the guidelines in water handling, which is covered in more detail
elsewhere. (7) (8)

CHLORIDES
The chloride level of the water is an important factor in determining the resistance of
stainless steel to crevice and pitting corrosion. Recognize, though, that there are other
important interacting factors that may have a major role, such as the presence of strong
oxidants, crevice geometry and pH. Laboratory trials and support service experience
suggest in most cases natural, raw and potable water with a pH of 6.5 to 8:

Crevice corrosion of 3041304L is rare below about 200 mg/l of chlorides.

Crevice corrosion of 31 6/31 6L is rare below about 1000 mg/l of chlorides.

In fully de-aerated water, much higher chloride levels can be tolerated.

CHLORINE
The addition of oxidants such as chlorine, up to some limiting concentration, can be
beneficial to stainless steels in preventing microbiologically influenced corrosion (MIC).
The residual chlorine as it leaves a potable water treatment plant may typically be 1.8 mg/
I, which is an entirely safe level for Type 304. In fact, levels as high as 25 mg/l of chlorine
for 24 hours is a standard disinfecting treatment. There is another important effect of
chlorine. In moist vapors above the water line, chlorine can reach concentrations that stain
and even pit 304 or 316 stainless steels. This is almost always more of a cosmetic problem
than structural, but pitting can occur with long-time exposure in un-ventilated areas.

MICROBIOLOGICALLY INFLUENCED CORROSION (MIC)

The corrosion of metals by substances produced by microorganisms is well known and
can be a problem in many industries. The subject of MIC is covered in detail in Reference

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 (8). Some of the practices for prevention of MIC follows:

For hydrostatic testing, ballast, settling and run-in procedures, use the cleanest
water available, i.e., demineralized, steam condensate, potable, etc.
Regardless of water quality, drain, dry and inspect to assure dryness immediately
following a hydrostatic test, i.e., within 3 to 5 days.
Eliminate or at least minimize crevices in fabrication.
Avoid heat tint in pipe welds with good inert gas backing procedures. Where
unavoidable, remove heat tint scale by grinding, abrasive blasting, pickling or
electropolishing.
Slope horizontal pipelines and heat exchangers to make them self-draining.

SUMMARY
AWS D18.1, Specification for Welding sf Austenitic Stainless Steel Tube and Pipe
Systems in Sanitary (Hygienic) Applications has recently been made available to the food
industry for technical support in welding piping systems. The particular areas addressed
are Welding Qualification and Visual Examination Requirements. By implementing the
specification provisions, it is expected that improved stainless steel piping systems will be
realized by the food industries.

REFERENCES:
1) AWS D l 0.4, Guide for Welding Chromiurn-Nickel Stainless Steel Piping and
Tubing, American Welding Society, 550 N. W. LeJeune, Miami, FL

2) ASME Boiler and Pressure Vessel Code, Section IX, Welding and Brazing Re-
quirements, ASME P. 0. Box 2900, Fairfield, NJ
3) ANSIIAWS 82.1 :1998, Specification for Welding Procedure and Performance
Qualifications, AWS

4) ASTM A380, Standard Practice for Cleaning, Descaling, and Passivation of

Stainless Steel Parts, Equipment and Systems, ASTM. 100 Barr Harbor Drive,
West Conshohocken, PA

5) Tuthill, A. H., Avery, R. E. , Specifying stainless steel surface treatment, NiDl

Technical Series No. 10 068, Nickel Development Institute, 214 King Street West,
Suite 510, Toronto, ON M5H 3S6

6) Private communication - Patrick H. Banes, Oakley Specialized Services, Inc.

7) Avery, R. E., Lamb, S., Powell, C. A., Tuthill, A. H., Stainless steel for potable
water treatment plants, NiDl Technical Series No. 10 087

8) Kobrin, G., et al, Microbiologically influenced corrosion of stainless steels by

water used for cooling and hydrostatic testing, NiDl No. 10085

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Figure 1 - Maximum misalignment
10% of tube wall thickness

0.006 in. f
Figure 2 - Non-product contact surface -
Maximum concavity

0.012 in. f
Figure - 3 Non-product contact surface -
Maximum convexity

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 T f
Figure 4 - Minimum face width for
Manual welding - 2T

MAXIMUM WELD
I'FACEWIDTH-W

4 MINIMUM WELD
FACE WIDTH 0.75 W -
Figure 5 - Minimum face width for manual welding -
75% of W where W = maximum face width

WELD TERMINATION
WITH OXIDE ISLAND
ON INSIDE OF TUBE

Figure 6 - Maximum acceptable diameter of an

Oxide island is 1I16 in.

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Figure 7 -Weld Discoloration Levels on
Inside of Austenitic Stainless Steel Tube

The amount of oxygen in the backing gas was

measured to be as follows:

NO.1 - 10 ppm NO.6 - 500 ppm

NO.2 - 25 ppm NO.7 - 1000 ppm

NO.3 - 50 ppm NO.8 - 5000 ppm

NO.4 - 100 ppm NO.9 - 12 500 ppm

NO.5 - 200 ppm NO. 10 - 25 000 ppm

Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 03/27/2019 Terms of Use: http://www.asme.org/about-asme/terms-of-use