Undermatched Filler metal tensile to base metal tensile

Date 04-07-2015 14:37

We are qualifying a PQR / Welding procedure joining two 100ksi tensile materials with 70ksi weld material.

My 2005 AWS B2.1 (par. 2.7.5) allows for the minimum tensile tests results to be:

1) the specified minimum strength of the base material, (or the weaker of the two base materials being joined)
2) The specified minimum tensile strength of the weld material when the referencing document provides for the use of weld metal having a lower tensile strength than the base metal
3) If the specimen breaks in the base metal outside of the weld or the weld interface, the test shall be accepted, provided the strength is not more than 5% below the specified minimum tensile strength of the base metal
4) If the base metal has no specified minimum tensile strength then failure in the base metal shall be acceptable.

We are compling to note 2 above, as our reduced tensile test results failed in the weld, and exceed the minimum filler metal tensile strength requirement.

Does AWS D1.1 specifically offer engineering the ability to qualify a weld procedure using undermatched filler material, or do we have to site AWS D1.1 4.2.1.2 using AWS B2.1 to accept the qualified procedure?
Thanks for any comments-

By SCOTTN

Date 04-07-2015 15:48

Although D1.1 is not intended to be used for steels with a minimum specified yield strength greater than 100 ksi, D1.1 allows the engineer to add to, delete from, or otherwise modify its requirements to meet the particular requirements of a specific structure, but they must be included in the contract.

By Bill M

Date 04-07-2015 16:24

Thanks-
Background: We are an OEM Equipment Manufacturing company and have been using this weld joint design for years in our product. We now are investigating an essential variable change and so we welded up some PQR sample plates for qualification. We enjoy the toughness and wear resistance of the base material, as well as the ductility of the weld joint and more forgiving and ease of welding with the 70ksi weld material. FEA (Finite Element Analysis) looks cool in the weld joint and we do not experience any legacy weld failures there.

SO, we sent the samples out for mechanical testing and the lab report said all the tensiles failed for both low tensile strength and for failure location in the weld joint. I tried to explain the weld procedure but he doesnt seem to understand.

By js55

Date 04-07-2015 20:16

Bill,
For tensiles your lab report does not have to specify whether or not the test passed or failed.
You can have him simply report the results.

By Bill M

Date 04-08-2015 16:04

JS-
I will request a report per your recommendation thanks!

By Tyrone

Date 04-10-2015 10:40

Hey Bill,
Did you model the weld at 70ksi in the FEA?
Tyrone

By Bill M

Date 04-08-2015 16:02

Thanks Al, and I agree. We did provide the written procedure to the lab. He remains firm in his belief that the 2010 D1.1 Code requires the reduced tensile test specimen tension test shall be not less than the minimum of the specified tensile range of the base material (par. 4.9.3.5). Hard not to agree with the guy, that is what it says. I was just wondering if the Code had other provisions for this type of application.

The same guy called the other day, because my PO said to test impact properties at -40 deg. He said he could not test the samples, because the PO did not say Fahrenheit or Celsius.

By 803056

Date 04-08-2015 16:21

Tell the lab to report the results of the test. Do not include a statement that it was tested to AWS or any other welding standard. You simply want the result of the tensile test.

In this case, you are using D1.1 as a method of qualifying the WPS. You could have used B2.1, ASME, or the Farm Code.

Sounds like the lab is being a horse's ass in this case.

Al

By Superflux (****)

Date 04-09-2015 19:20

Since not specified, -40 in Celsius would be, Uhmmm -40 in F.
-40C (or F) in Kelvins would be 233.15, but since there are no minuses on the Kelvin scale, it would/could be assumed that the requested temp be in either C or F.
Any good lab though could not make the test on such assumptions since surely somewhere there must be a minute fraction of a degree difference between -40F and -40C in Kelvins if one does the math up from an assumed Absolute Zero or backwards from water freezing.

On a side note, -40 is one of my favorite temperatures as there are some memorable approximates listed below:
BOTH F and C scales meet there (-40F = -40C).
It is the approximate boiling point of Propane. We did this experiment one night at the coal mine when it was -45F and "poured" a glass of propane outside the shop.
It is also the approximate freezing point of Mercury (-37.9F, 38.8C and 234.3Kelvins)
It is also the approximate freezing point of the Human posterior. As is often referenced with the following statements;
"It was 40 below and I froze my A!! off..."
And lets not forget the most annoying song from the 70s that said; "Well it's 40 below and I don't give a "F", gotta heater in the truck and it's off to the Rodeo..."

Back in the day, the old timers would say it dropped to 40 below and stayed there. Noticed by none was that the mercury froze at -39.
If someone back then said it was -50, then you know the story was BS.
I do remember in Feb of 1985 when the meteorological station in Peter's Sink. UT (on the mtn. above Evanston, WY) recorded -66F. City water mains at nine feet froze and burst.

By welderbrent (*****)

Date 04-10-2015 04:09

Now that was educational and scientific. Some posters could learn from that post.

He Is In Control, Have a Great Day, Brent