D1.5 PQR all weld metal tension retest per section 5.20

By hbardy Date 01-28-2013 20:12

I have a 4G PQR test plate that failed the all weld metal specimen per the AWS D1.5 2010. Clause 5.20 allows for a retest of two specimens from the same test plate or a new test plate conforming to the same specification. The contractor elected to run a new test plate was ran and both specimens passed the tension test. My dilemma is that I have two different sets of parameters due to two test plates being ran. Which set of parameters should be used to write the WPS? This appears to be an rare occurrence from what I have been told and could possibly be due to a testing lab anomaly. Any help from someone with experience in this matter is appreciated.

By 803056

Date 01-28-2013 20:19 Edited 01-28-2013 22:00

The WPS is based on the parameter of the successful test.

If the parameters of the two tests are different why would you base the WPS on the parameters used for the test plate that failed?

The failure could have been because the test weld had a discontinuity or because the mechanical properties of the test weld were insufficient. Hydrogen could have also played a part in the failure. Did you check the filler metal specification to see if it includes a requirement to age the test coupon before tensile testing? Oops!

Al

By hbardy Date 01-28-2013 21:41 Edited 01-29-2013 17:41

The filler metal used is Lincoln NR-203 1% Ni on A709-50W. No aging is required for this filler metal and base metal aging is prohibited by D1.5 unless required by the WPS, which none is required. The first test plate passed all bend and CVN testing as required by the D1.5. The tension tests from the first plate failed the elongation by 1%. The test was welded manually so there is a slight variance between the parameters in the two plates.

Since only the tension tests were taken from the second test plate and both specimens passed, in my mind I have two test plates that passed some of the criteria given the clause that allows for a retest. So in your opinion which plate is the successful test since in essence they both passed, or am I missing something?

By 803056

Date 01-28-2013 22:04 Edited 01-29-2013 01:30

If the code does not specify I guess it is your choice.

I believe there is a presumption on the part of the code that both plates, i.e., the original and replacement, are welded using the same parameters.

Al

By hbardy Date 01-29-2013 13:24

I totally agree that the code presumes the same parameters for both plates. Obviously we ran the same voltage and wirefeed speed as the other test and monitored amperage during both. The issue we found is with matching the travel speed, especially from pass to pass,doing it manually therefore effecting heat input. We had a very experienced welder doing the test and did the best we could to match the welding parameters.

By waccobird (****)

Date 01-29-2013 15:18

hbardy

Did you develop your parameters through formulas or from the experienced welder?

I would have to ask is this welder the only one using the procedure?

If not I would say that instead of looking at the difference in test results as an anomaly, I would take it that my heat input was too high and change my parameters such to achieve the required results consistently from all welders.

Marshall