Fillet WPS Qualification for D1.6

Date 07-16-2013 14:46

Don't forget to weld a standard V-groove coupon that will be sliced and diced for determining the mechanical properties and soundness of the joint. Look at clause 4.1.7 (D1.6:2007 as a reference).

A WPS that is only qualified by welding a T-joint that is macroetched tells you nothing about the mechanical properties of the combination of base metal and filler metal. You could weld CS to SS with a coat hanger and pass the macroetch. I don't know that I would want to hang my reputation on that weld unless I knew it was capable of meeting the tensile strength and ductility requirements.

Best regards - Al

By jwright650 (*****)

Date 07-16-2013 15:21

Al, testing the mechanicals seem to be something good to consider for sure, a bit above the D1.6 code requirements for fillets, but a good suggestion non-the-less.

OK, Thinking along the lines of ensuring good mechanical properties between the CS and SS using my FCAW 309LT1....would one side of the V-groove joint be CS and the other SS? and what would I use for the backing? CS or SS? D1.6 doesn't seem to give any guidance on testing this unless qualifying a PJP or CJP(I don't have any CJP or PJP anywhere on the project, so I have to sell this extra testing to the bean counters).

By 803056

Date 07-16-2013 20:59 Edited 07-16-2013 22:48

Hello John;

I guess it is time to get philosophical about this and reexamine why one would the spending the time and money to qualify a WPS. Before expending time and ink, let’s consider the following:
1) Is there an existing qualified WPS that involves welding stainless to carbon steel previously?
2) Have the mechanical properties been verified for the combination of filler metal and base metals?
3) What are the purpose and function of the welded component you are fabricating?
4) Are the design loads and the unit stress of the weld in question known?
Assuming the answer to questions 1 and 2 are “no,” you need to verify the proposed WPS will be capable of producing the mechanical properties expected by the designer. You need to know the mechanical properties of the weld before you can determine whether the WPS is adequate to fulfill the needs of questions 3 and 4.

When I review the qualification requirements of a WPS in Clause 4.1, I see that only CJP and PJP groove welds can be used to qualify the WPS. Both the CJP and PJP test coupons qualify for fillet welds, but clause 4.1.7 specifically states that a fillet weld has to be further qualified by performing the macroetch tests defined in clauses 4.3.2 and 4.4. The fillet weld tests described in Clause 4 only determines if the weld is fused to the root and it can meet the visual acceptance criteria, i.e., it “looks pretty.” It does not determine whether the weld will produce the required strength and it does not determine whether the weld is ductile or if it is as brittle as plate glass.

The contractor has an obligation to meet the requirements of the code and the project specifications. The contractor has an obligation to take whatever additional steps are needed to protect their interests both from a financial prospective and their reputation as a quality fabricator. With the aforementioned in mind I may be overly conservative. However, there must be a certain rational applied to how and why a WPS is qualified.

In this case I agree that the 309 should produce the required results, but that is because I have qualified several WPSs using different austenitic stainless steels with several different low alloy steels. Would I feel as confident if it was proposed that ERNiCrMo-3 be used to weld CS to SS? No, I have no doubt that it will produce a fine looking fillet weld, but I have no experience with that combination and I have no firsthand knowledge that it will produce the desired strength or ductility. I would test it by welding up a CJP groove welded coupon to verify the mechanical properties are acceptable. This is not a situation where the SWAG method is best practice.

The CJP can be backed with either the carbon steel or the stainless steel. I would use the same base metal specifications for my welded sample as I would be using for production. I would used the same low alloy steel welded to the same alloy composition of stainless steel required for production.

Remember, it could be your family or mine that is affected by a failure should one occur. We, as CWI, have a responsibility to do what we believe will ensure the safety of the people using the product even if it is over and above what the code sets as a minimum, base line, requirement.

Best regards - Al

By jwright650 (*****)

Date 07-17-2013 11:15

Al, I'm in agreement with you but...

>"would one side of the V-groove joint be CS and the other SS?"-quoting my earlier posting

What do you think that I should do with this situation for the groove weld to test the soundness and mechanicals?

Use all SS? or a combination of CS and SS in this groove weld like the fillet will be consisting of? I'm hung up on how to proceed with the soundness testing, since the D1.6 code don't give any direction for anything like this for a CS/SS fillet weld.

By 803056

Date 07-17-2013 14:07

The way I have qualified these WPSs in the past is to use plate or pipe matching the specifications of the materials used in production. Things can get a tad tricky when the specification dictates product form as would be the case of an ASTM A992 or A53, etc. In those cases, with the concurrence of the Engineer, I match the mechanical properties and the chemistry of the materials used in construction. It also makes a difference when working with a code that may dictate the use of the same material specification to be used in construction in which case little latitude is permitted.

Now that the materials are out of the way, I use one piece of low alloy or carbon steel and the other piece is the same stainless alloy that will be used for construction. Each is beveled to provide the appropriate groove angle. The WPS can be qualified with or without backing. I usually opt for CJP without backing to avoid that complication unless backing will be used in production. If backing is to be used in production, I use the same backing material for the PQR. In your case you are not using a groove weld in production, so I would opt for the CJP without backing. Instead, I would simply weld one side and back gouge the opposite side and complete the weld from the second side.

I should mention once again that I match the chemistry of both base metals if there is an option of using pipe or plate. Whether I use pipe or plate is dependent on production needs and the applicable code. In your case if you can use plate that matches the chemistry and specification that will be used in production life is not as messy. This is one situation were you cannot simply pick any carbon or low alloy steel from one group in AWS and call them "equivalent". Likewise, you cannot simply pick a stainless from group A or B and consider them to be equivalent. The chemistry, heat input, and dilution will affect the Ferrite Number and the propensity to develop solidification cracks.

Establish your mechanicals using a grooved test assembly and then qualify the fillet welds using the T-joint. Then you are good to go and no one can question whether the production welds have the required mechanicals needed to safely transfer the design loads.

Before spending time and money to qualify the WPS, pass the proposed WPS by the Engineer for his concurrence.

Best regards - Al

By jwright650 (*****)

Date 07-17-2013 14:15

Thanks for your detailed replies Al. :cool:

By 803056

Date 07-17-2013 14:18

Any time John. It isn't often that I have an opportunity to offer something constructive to an expert.

Best regards - Al

By jwright650 (*****)

Date 07-17-2013 14:36

SS is not something we do very often....this shop is just not set up to keep everything clean/iron free and we don't have any tools to work with the SS...a down right pain in the butt trying to manage the material flow in the shop when you have to dedicate a welding machine to SS when every table in here is set up for CS. So for these 500 tons of material, I'm trying to just deal with it and get passed the headache without running our shop hours up just moving materials all around due to this.

The engineer says the A240 Type 304 SS is to be a thermal break from cold to hot to minimize sweating inside the building. I had no idea that SS wasn't as thermally conductive as CS. I do know it is very sensitive in that a simple tack weld will pull a plate way out of alignment and that distortion is something else to deal with.

By 803056

Date 07-17-2013 15:28 Edited 07-17-2013 15:33

The austenitic stainless steels are a different breed to say the least.

Here is a comparison of some of the physical and mechanical properties:

304 Stainless    AISI 1020 carbon steel (as rolled)

UTS 90 ksi 65 ksi
YS 42 ksi 48 ksi
Elongation    55% 36%
Mod. of Elasticity 28 x 10^6    29 x 10^6
Resistivity 7.2 x 10^-5 ohm-cm 1.59 x 10^-5 ohm-cm
Coeff. of Expansion    9.39 micro inch/inch-F 6.5 micro inch/inch-F
Conductivity    112 BTU/pound-F    360 BTU/pound-F
Liquidus    2649 F 2760 F (approx.)

Looking at the thermal conductivity, you can see the engineer's point, the stainless is only 1/3 as conductive as the carbon steel. The stainless is not a stiff (modulus of elasticity) and the electrical resistivity is 3 times higher than carbon steel.

Al

By jwright650 (*****)

Date 08-01-2013 17:46

Hi Al,
Just wanted to update you regarding the testing. I submitted the ASTM A240 Type 304/A36 Fillet Weld sample (5/16 max single pass-3/8 smallest multipass "T joint") for macros and bending, along with a B-U2a ASTM A240 Type 304/A36 combination welded with E309LT1-1 for tensiles.

Bends were acceptable.
Macros were acceptable.
Tensiles were reported as Ductile / BOW on the A36 side and was noted that AWS D1.6 did not require mechanicals, but customer requested them to prove that the combination of material was acceptable.

Thanks for your help, Al.

By 803056

Date 08-01-2013 21:35

Happy to help.

Al