WPS Vs. PQR Essential Variables

Date 02-04-2010 16:58

You need to pick which PQR is supporting your WPS, and then work with that PQR and the table to figure out how far the WPS can be from the PQR. You can't mix and match bits of multiple PQRs to qualify a single WPS.

Or am I misunderstanding your question?

Hg

By Stephen32 (*)

Date 02-04-2010 17:49

I have been led to believe that you can have multiple PRQs for each WPS. As an example a PQR done in a 3G position is only good for the 3G position. You would still have to qualify a 2G to be able to do production welding in the 2 and 1 positions (using the same parameters as the 3G). The same would be true for any material that is not listed in table 3.1.

Have I confused myself? Maybe I have been looking at this too much and have started to chase rabbits...

By HgTX

Date 02-04-2010 17:53

The only time you'd have multiple PQRs would be if you're using different procedures within the same joint, which really is more like combining multiple WPSs but I think most people just list it on the same piece of paper which for most people means "one WPS". So you might have one set of parameters for your root pass, backed up by one PQR, and then another set of parameters for the rest of the joint, that sort of thing.

But you definitely can't just mix and match, pull your travel speed off one PQR and your current off another and your volts off a third. The PQR shows how all those parameters work together.

Hg

By Stephen32 (*)

Date 02-04-2010 19:24

I do agree that you cant mix and match. However I disagree as to having only one PQR for WPS. As stated in my previous post, under some circumstances you would have to have multiple. Sometimes as many as 10 or more.

Case in point: We use a material called Domex. It is a 90 KSI steel, also it is not listed in Table 3.1. Our Engineers have authorized the use of E71T-1 FCAW for the welding of A572, A36, and several other non-loadbearing steel "brackets or connectors" to these Domex webs. Now due to the strength of this steel, we have to have a material specific PQR (in 2G and 3G positions) for each of these, depending on the location of installation.
We are an API certified facility that builds drilling rigs and work-over rigs from the ground up. This in turn means that the specific WPS must be posted on all of the shop drawing (there are thousands of drawings). To cut down on the total number of WPSs I have been trying to consolidate and re-certify our WPSs with new PQRs.

My intent is not to offend, but to get better and learn from some of the people here.

By HgTX

Date 02-04-2010 19:36

If you ran your 2G PQR at 11 ipm to give a WPS range of 9-13 ipm and you ran your your 3G PQR at 8 ipm to give a WPS range of 6-10 ipm, you can do whatever positions are qualified by the 2G PQR at 9-13 ipm, and you can do whatever positions are qualified by the 3G PQR at 6-10 ipm. You can't just stack the ranges and say you can weld both 2G and 3G at 6-13. That's mixing and matching, because you're applying the 6 ipm you obtained from a 2G test to a vertical position you qualified with the 3G test.

Or am I still not getting it?

Hg

By Stephen32 (*)

Date 02-04-2010 20:20 Edited 02-04-2010 20:30

Yes you are getting it. I am with you on all that you said in last post. The question remains: Can it be on the same WPS or does it need a whole new WPS for the 2G PQR?
Key phrase in the code " shall require requalification of the WPS".

Ok I reread what you said. I think I know what you are saying. On the WPS where ever you have the parameters listed are you saying to basically list everything for the 2G/1G on one line or column and then list everything for the 3G on another line or column?

By HgTX

Date 02-04-2010 20:23

Sounds to me like you're trying to stuff multiple WPSs onto one piece of paper.

If you have a way of tracking on your WPS "for this kind of pass, do this, and for this kind of pass, do that", then you're okay. If you can't find a way to say that on the WPS without leaving open the possibility that the welder might misinterpret and do some mixing and matching, then separate all that same info into different WPSs.

It will probably be the least confusing if you separate them out by position. But that's just what I think sitting over here not looking at either your paperwork or your part.

Hg

By Stephen32 (*)

Date 02-04-2010 20:41

Hg I really do appreciate your input into this. Again my intent was not to disagree but to learn and to get better.
Does anybody else have any input on this?

By jwright650 (*****)

Date 02-04-2010 20:42

re: parameters on multiple lines

I do this to list several electrode diameters, where applicable.
ie. show parameters for each diameter on seperate lines.

By 803056

Date 02-05-2010 13:37 Edited 02-05-2010 13:43

Different codes have different requirements, but in general terms the ranges can be whatever is supported by one or more PQRs. AWS has been doing just what has been said can't be done for years with their Standard WPSs.

There is nothing that I know of that prohibits a company from qualifying several PQRs, one for each position or several for the same position, to increase the permitted ranges listed on the WPS.

As an example; one PQR is qualified for stringers using high travel speeds and low amperage and low voltage. A second PQR is qualified using weave beads, high amperage, and higher voltage. The two PQRs can be combined into one WPS permitting a wide range of travel speed and a wider range of amperage and voltage than would be permitted by table 4.5.

Likewise, when qualifying a welding procedure for impact toughness it is not uncommon to qualify one PQR for high heat input and a second for low heat input. The data of the two qualified PQRs are combined into one WPS with the ranges validated by the two separate PQRs.

A third example is where multiple PQRs are qualified, each for a different position and then combined into one WPS for all positions. The ranges do not have to be divided into ranges for each specific position.

As long as there are PQRs to support the parameters listed on the WPS, the document is valid. The WPS can be more restrictive than the ranges supported by the PQR, the WPS can list the ranges for each position, or it can lump them together.

Once again, this is where the individual writing the WPS has to consider the type of work being fabricated and what objectives are to be met. Is the WPS for general applications where wide ranges are needed to provide sufficient latitude to meet varying field conditions or are tight controls required to ensure a specific joint in a specific position is to be welded the same way every time?

AWS SWPSs are supported by several PQRs to support a wide range of variables. Here is just one example taken from a SWPS for FCAW using E70T-1/E71T-1:
Dia. 0.045 inch
Voltage: 24-28
Amperage: 180-280
WFS: 230 - 440

As long as the ranges are supported by one or more PQR, the ranges are valid.

Best regards - Al

By Stephen32 (*)

Date 02-05-2010 15:15

First off I would like to thank you gentlemen. This is what makes all of us get better.

John, I appreciate your input and will probably use your approach on some of the instances.

Al, what you stated was my original take on the matter. Certain joints and or certain materials MUST have very tight tollerances on all variables. However other joints or materials can be much less restrictive.

Hg I thank you also.

By 803056

Date 02-05-2010 16:05

On a personal note, I like to keep the parameters tight enough that a new or inexperienced welder can set up the machine using the parameters listed as optimum and begin to weld.

I am not a fan of the the SWPS because they are too general. They follow the mind set of ASME and are not "welder friendly". How many welders do you know that knows what P number an ASME SA-105 is?

Best regards - Al

By HgTX

Date 02-05-2010 16:06

Al, are you seriously saying in your 3rd paragraph that those two PQRs could be combined to allow, for example, weave beads with low amperage and low voltage?

Hg

By 803056

Date 02-05-2010 16:24

Unfortunately, most codes will allow you to do just that.

This is where the welding engineer has to have sufficient experience to differentiate between what the "codes" allow versus what is realistic.

How many times have I made the comment that I can write a WPS that meets the requirements of the code, yet it would be useless to the welder?

We have to be careful to differentiate what is permitted by "code" and what is personal opinion. Many of us have developed a sense of what does and what does not work in the "real world" based on our experience. The codes allow us to utilize that experience to formula welding procedure specifications that best suit our needs. Those needs are dictated by job conditions, customer expectations, and knowing the limitations of the welders that will be using the WPS.

Different codes have different requirements as do customers. I have customers that insist on a separate WPS for each joint on the specific job. You guessed it, the client is DOT, and they require a separate WPS for fillets and a separate WPS for each groove type, i.e., V-groove, J-groove, etc. That is a customer requirement, not a code requirement.

As a TPI that is saddled with reviewing WPSs submitted by various contractors I have to be careful to keep my opinion separate from code requirements and limitations. My opinion is reserved for those occasions when it is requested. I do not offer an opinion during a review unless it is specifically requested.

Nothing is black and white; there are many areas of gray that require us to tread carefully.

Best regards - Al

By HgTX

Date 02-05-2010 17:00

See, my interpretation of the code is that you can't do that. If you can do that, then you could run one PQR with base metal A and filler metal B and a different PQR with base metal C and filler metal D (with C and D being nothing like A and B), and write a WPS with base metal A and filler metal D. The point of the test is to put certain variables together and show that they work in concert. And I don't see anything in the code that allows taking some variables from one PQR and some from another. Calculating allowable heat input from the current in PQR 1 and the voltage in PQR 2. Or aluminum base metal from PQR 1 and steel filler metal from PQR 2. At what point does logic and common sense kick in? You might as well not run any tests at that point, because none of them mean a thing if all you pull out is one variable in isolation. And I don't see anything in D1.1 to support that practice.

It comes down to what "a" WPS is. If you stuff every kind of weld you might possible want to do at any time onto one piece of paper and call it "a" WPS, things get very messy indeed. But to me, that's not "a" WPS. That's a bunch of WPSs crammed onto one piece of paper. Each type of pass represented on it needs its own qualification (or prequalification).

Hg

By 803056

Date 02-05-2010 18:48 Edited 02-05-2010 18:54

Your point is not without merit, but the code does allow considerable latitude in these matters. Consider the following:

Procedure A is qualified using GTAW for open root grooves on pipe in the 6G position using A106 pipe. Procedure B is qualified using SMAW on pipe with backing in the 6G position using A53 pipe. The parameters and test results are documented as PQR A and PQR B respectively.

WPS AB is written to use GTAW for the open root and second layer. The fill and cover are to be deposited using SMAW. WPS AB is supported by PQR A and PQR B. Perfectly legitimate and it is common practice for mechanical piping.

Are you suggesting that there should be two separate WPSs of the application noted above or that it isn't permitted?

Another scenario: Aluminum alloys welded with GTAW using a common filler metal 4043. The first successful test is documented as PQR C using GTAW, Argon, ER4043 filler metal, 6061-T6 base metal using a bevel groove. The second test is documented as PQR D using GTAW, Argon, ER4043, using 5052 base metal with a double V-groove. All the tests are acceptable, the differences between the two being the groove type and the M number of the base metals. Both PQRs are used to support one WPS listing all M22 and M23 base metals and all the groove details depicted in AWS D1.2.

Once again, I ask you, this is not acceptable?

My last example: Procedure E is 3/32 inch thick carbon steel plate welded in the flat position with 3/32 E7018 at 23V, 90 amps, 4 ipm. Procedure F is carbon steel plate is 3/8 inch thick welded in the flat position using 5/32 E7018 at 27V, 8 ipm, and 180 amps. Both are documented as PQR E and F respectively. The fabricator elects to write the WPS for 3/32 inch to 3/4 inch using 3/32 to 3/16 electrode with the ranges of 22V to 29V, 70 to 285 amps, 3 to 10 ipm listed.

Are you saying that he can't do that?

Are there better ways of documenting the WPS? Only in your opinion which may differ from that of the fabricator or even in my opinion.

Is he wrong in the manner in which he does his documentation? Not unless he is violating a particular requirement of the code he is working with.

Your opinion is just that. It is your opinion and it may be an acceptable manner to meet the minimum code requirements. However, there are other ways to meet the code requirements that may be just as valid in someone else's opinion. No two companies document their welding procedures in the same way. As long as they address the essential and nonessential variables listed by the applicable code, they are golden. Whether it is the best way in your opinion or my opinion is just so much wasted hot air.

What you did not do in your scenario is keep the essential variables in mind. If the base metals have different P/M/or S numbers, but the filler metal is the same F number, and if the test results validate the practice, both WPSs can be combined. There is no doubt that the individual documenting such complexities better have a good understanding of what can and what can't be combined.

Last example and then I'm moving on. Lets say the contractor qualifies a procedure, PQR-M, using 6061-T6 base metal using ER4043 filler metal. A second procedure, PQR-N, is qualified using 6061-T6 base metal with ER5356. The filler metals have different F numbers. A third procedure, PQR-0, is qualified on 5052 using ER4043. A forth procedure, PQR-P, is qualified using 5052 base metal with ER5356. Again, the base metal M number is the same, but the F number of the filler metal is different.

We have 6061 (M23) qualified with both ER4043 (F23) and ER5356 (M22)
We have 5052 (M22) qualified with both ER4043 (F23) and ER5356 (M22)

(M numbers are for base metal groupings per AWS D1.2 and B2.1. S and P numbers are from ASME, and S numbers from NAVSEA publications)
(F numbers are for filler metal groupings per AWS D1.2 and B2.1 and ASME. They are similar to the A groups found in NAVSEA publications)

There is nothing that says a single WPS can't be written for both M22 and M23 base metals welded with either F22 or F23 filler metals as long as there are supporting PQRs that validate the combinations. The combinations have been qualified by testing and the mechanical properties (all combinations) meet the code requirements. Unless the parts are to be anodized and a color match is a customer requirement, there is no reason I know of that supports the need for four separate WPSs.

Once again, the person better have a thorough understanding of the code requirements that apply and what can be combined and what can't be combined. The combinations have to take into consideration the essential variables of each PQR. In one of your examples you are trying to combine F numbers and M numbers that haven't been tested or supported by "good" PQRs.

Am I saying that the amperage used for 3/32 electrode is the same as that used for 3/16 inch diameter electrode? Hell, no. Even I'm not that silly.

Bottom line: know the code you are working with and know your audience when developing a WPS. Too many WPS creates a paperwork nightmare, too few can cause confusion.

Best regards - Al

By HgTX

Date 02-05-2010 20:00

Case 1:

"Procedure A is qualified using GTAW for open root grooves on pipe in the 6G position using A106 pipe. Procedure B is qualified using SMAW on pipe with backing in the 6G position using A53 pipe. The parameters and test results are documented as PQR A and PQR B respectively.

WPS AB is written to use GTAW for the open root and second layer. The fill and cover are to be deposited using SMAW. WPS AB is supported by PQR A and PQR B. Perfectly legitimate and it is common practice for mechanical piping.

Are you suggesting that there should be two separate WPSs of the application noted above or that it isn't permitted?"

To me that's combining two WPSs in one joint, but that distinction doesn't really matter; I may be wrong in my terminology. In this case, you're using PQR A, in its entirety, for the open root and second layer, in their entirety. Likewise for PQR B and the fill and cover. Each type of pass is independently qualified by its own PQR. You're not taking bits and pieces of PQRs A and B and mixing them together to create an entirely different type of pass from what was tested in either PQR. So if your WPS adequately indicates that *this* PQR qualifies *these* passes and *this* PQR qualifies *these* passes, I'm fine. If your WPS doesn't distinguish between the first two passes and the rest, and just says "weld this joint with SMAW or GTAW, based on PQRs A and B", then you have a problem. If your choice of how to document your procedure captures what needs to be captured in order to make the distinction, then you have no problem. I don't care how many pieces of paper you use or how many numbers you assign to it as long as you separate what needs to be separated in some way. But I am in agreement with you that, assuming it's written to convey what needs to be conveyed, that this should be allowed.

Case 2:

"Another scenario: Aluminum alloys welded with GTAW using a common filler metal 4043. The first successful test is documented as PQR C using GTAW, Argon, ER4043 filler metal, 6061-T6 base metal using a bevel groove. The second test is documented as PQR D using GTAW, Argon, ER4043, using 5052 base metal with a double V-groove. All the tests are acceptable, the differences between the two being the groove type and the M number of the base metals. Both PQRs are used to support one WPS listing all M22 and M23 base metals and all the groove details depicted in AWS D1.2.

Once again, I ask you, this is not acceptable?"

Again, I think what you really have there is two WPSs combined on one piece of paper, but that's not really what this discussion is about.

If the *only* difference is the base metal and the groove, not a problem. You're not mixing and matching because everything else is the same and there's nothing to mix. If, however, you had a particularly low-heat set of parameters on PQR C and a high-heat set of parameters for PQR D, do you think you've adequately demonstrated that the low heat of PQR C will work for welding the type of base metal you used for PQR D? I don't know the answer to that; I don't deal with aluminum. But for steel, if you used a low-heat procedure on one type of base metal or filler metal and then high-heat procedure on a completely different type of base metal or filler metal, there's nothing there to tell me that the low heat would have worked on both. That's the whole point of Table 4.5. "You did X in the test. There's only so far away from X you can go for that test to still be relevant. If you deviate enough from the test, that test no longer represents what you plan to do in production and you need to show me another test."

Case 3:

"My last example: Procedure E is 3/32 inch thick carbon steel plate welded in the flat position with 3/32 E7018 at 23V, 90 amps, 4 ipm. Procedure F is carbon steel plate is 3/8 inch thick welded in the flat position using 5/32 E7018 at 27V, 8 ipm, and 180 amps. Both are documented as PQR E and F respectively. The fabricator elects to write the WPS for 3/32 inch to 3/4 inch using 3/32 to 3/16 electrode with the ranges of 22V to 29V, 70 to 285 amps, 3 to 10 ipm listed.

Are you saying that he can't do that?"

Yes, I'm saying he can't do that. To weld 3/4" plate with 3/32" electrode he's pulling the plate qualification from PQR F and the electrode diameter from PQR E. He's done nothing to demonstrate that that combination ought to work (assuming qualification testing actually has merit and that it's not the case that pretty much anything will work with anything). That's every bit as logical as pulling his base metal (brass) off PQR E and his filler metal (titanium) off PQR F. Okay, I exaggerate, but that's what you get when you follow the "one from column A and one from column B" approach to its extreme.

To me, Table 4.5 says that if you've run a test, here's how far away from that test you can go. There's nothing in the code that says you can run a separate test to qualify each essential variable independently and to me that is a terrible misapplication of the code's intent.

And, of course, every word I've ever typed on this forum, short of direct quotations from documents, is only my opinion. My opinion only counts if you're trying to get me to approve something.

Hg

By 803056

Date 02-05-2010 21:21 Edited 02-05-2010 21:31

You are running afoul of ASME Section IX because the parameters you are concerned with are nonessential variables. They can be changed without supporting evidence of a PQR.

I am not disagreeing with your thought process from a welder's view point, but strictly speaking as a "code guardian" your argument would hold no water if the applicable code is ASME Section IX.

Heat input only plays a major roll when notch toughness is a concern. Whether you elect to weld heavy plate with small diameter electrodes or large electrode as is the case with PQR C and D is of little concern. It will take longer with small rods. It will require more weld passes with small rods, but the number of passes has little influence on the mechanical properties such as tensile strength or ductility. As long as the two PQRs used base metals having the same P, M, or S number (it did, both SA105 and SA106 are P1), it is perfectly legitimate to combine the PQRs into one WPS. It is standard practice for many applications.

I understand where you are coming from and I am usually on the same side of the fence you are, but there is a difference between meeting code requirements and meeting the customer's needs. As a consultant you have to be aware of both. As a TPI reviewing the welding documentation package you have to keep your personal opinions to yourself and do the review based on what is and what is not acceptable per the applicable code. When you interject your personal opinion into the process you have made yourself a consultant instead of a TPI.

I am going to admit that I screwed up on the aluminum procedure, I forgot to include a PQR for 5052 to 6061 with ER4043 filler metal and 5052 to 6061 using ER5356 filler metal to make the combination legitimate. So, one WPS supported by six PQRs. Its all a part of my reduction in paperwork enspoused by the US Government. You've seen the statement I'm refering to on all the IRS forms. Back on subject, with the six PQRs I can weld 6061 to 6061, or 6061 to 5052, or 5052 to 5052 with either ER4043 or 5356, all with just one WPS. For this exercise I would use a tabular format to describe the combinations. The variations and permutations are endless! Ah ha ha ha! (mad evil laughter)

Best regards - Al

By HgTX

Date 02-05-2010 21:27

I assumed we were talking about D1 codes, this being the "D1 Questions" board. I've never dealt with ASME Section IX and don't know what its requirements are.

It still doesn't make any sense to qualify each line of the D1.1 Table 4.5 essential variables with a completely independent test. In my opinon, of course.

Hg

By 803056

Date 02-05-2010 21:33 Edited 02-05-2010 21:39

Many of the comments are valid for D1.2 as well as B2.1 or ASME Section IX.

I now return to Stephen's original question and I still say the response is yes. He has supporting PQRs that allow him to weld in multiple positions and the ranges are a compilation of the ranges used when qualifying the individual PQRs.

I like Johns approach to separate the positions and list the parameters based on the applicable PQRs, but there is nothing that says I have to do it that way.

As a matter of fact, I typically write a WPS for welder performance qualification with very wide ranges to see what parameters work the best. Then I revise the production WPS by narrowing the range for the parameters to reflect the lessons learned while qualifying the welders.

Best regards - Al