Determin Ductile vs Brittle for Tensile test

Date 11-12-2009 13:18

Hojo
I am not sure you performed a tensile test.
Did you pull the test coupon apart?
Did you do it to a code?
If you do a proper tensile test there will be measurements taken of the coupon before you pull it into and after it is pulled into. The specimen will show a percent of elongation that will demonstrate whether the weld was ductile and stretched before breaking or if it snapped into, showing the test produced has resulted in a brittle condition.
Hope this helps more than hurts?

By Hojo

Date 11-12-2009 13:28

Thanks for the input.
Yes we did have a tensile test performed and it was done by an accredited lab years ago. It was performed to ASME Section IX. All they reported to us was the Width, Thickness, Area, Ultimate load, Ultimate stress and they reported it as "Broke in base metal". They have never reported % elongation for any of the test reports that we have. I have no other information. I guess that we will have to ask for the % elongation from now on.

By waccobird (****)

Date 11-12-2009 13:59

Hojo
I guess I should have mentioned reduced-section Tension specimens, as this will help cause the weld area to be the place of break and thus give a true indication of the weld hardness, not the base metal.
Good Luck.
Marshall

By Hojo

Date 11-12-2009 15:46

The tensile tests were reduced section tensile specimens.

Tkx

By waccobird (****)

Date 11-12-2009 16:04

Hojo
Then I suggest an All Weld metal tension specimen be pulled to confirm weld Ductile VS Brittle.
Marshall

By CWI555

Date 11-12-2009 17:05

http://www.steeluniversity.org/content/html/eng/default.asp?catid=151&pageid=2081271959

By MMyers

Date 11-12-2009 19:29

The funny thing about welds and percent elongation is that a transverse weld tensile is really at least three materials: weld metal, HAZ, and base metal. Each one will elongate differently, so percent elongation of the entire specimen is kinda meaningless from a welding procedure standpoint. I would say where the sample fractured is really your key here. It broke in the base metal, so you can't say anything about the weld with regard to mode of fracture. You really need to do weld metal impact tests for ductile/brittle stuff or some other all weld metal tests, but with impacts testing temperature comes into play which doesn't simplify things. I would be pushing them for a "why" they want this info.

By Hojo

Date 11-12-2009 19:36

This is a markup on a weld procedure by one of our customers. All we have stated is that the specimen "Broke in Base Metal" and they want the word Ductile inserted. This is for PQR's that were tested a number of years ago and I am certain that the lab does not have any information left other than what they provided to us on the report. I appreciate your info and agree that there are the three different zones that will all demonstrate different amonts of elongation and deformation until the specimen finally breaks.

By HgTX

Date 11-12-2009 19:49

If the contract doesn't include a code requirement for this info, they can't hold you to providing it. And if it broke in the base metal, what do they care what the mode was? That's the mill's issue, not the welder's issue.

Hg

By waccobird (****)

Date 11-12-2009 20:49

HgTX
The issue could be the customer is requiring them to answer a question, if they want to satisfy the customer the thing to do is answer. If it is being asked directed at the WPS then they are questioning the weld parameters being such that the weld is not ductile but brittle. HoJo has said the testing was a number of years back PQR's only last 5 years. If the weld is Brittle it may not suit their application so they want a guarantee/statement from someone that the parameters of the WPS produces a Ductile condition. Harder is not always better.
Hojo For a couple hundred bucks weld a Test specimen and have an all weld metal tension test done and have the testers determine it and answer the customers question.
Otherwise you may be doing the whole PQR thing and for my area it runs @ 700 to 1500 depending if charpy's are required.
Good Luck

By HgTX

Date 11-12-2009 21:17

That's my point. They shouldn't have to re-run any testing unless they failed to meet test reporting requirements to begin with. Particularly if AWMT isn't a contract requirement.

Even if they had reported ductile vs. brittle failure on the report, that would tell the customer nothing about the weld properties because the ductile or brittle failure, whichever it was, was not in the weld (which is usually seen as a good thing). I wonder if the manufacturer's C of C would have information on it that might reassure the customer in lieu of running an AWMT for no legitimate reason.

Hg

By HgTX

Date 11-12-2009 15:54

What code are they working to?

By Hojo

Date 11-12-2009 19:15

ASME Sect IX.

By 803056

Date 11-13-2009 20:13

The standard transverse reduced section tensile test does not include a determination for ductility because of the factors already noted. There are three (or more) areas that have different mechanical propreties. The standard RST test simply breaks the sample in the weakest region, i.e., unaffected base metal, heat affected zone, or the weld.

If the sample breaks in the unaffected base metal because it is the weak link, it will display the characteristics of the base metal. If the base metal is ductile, the broken sample will display the features of ductile failure. However, the width of the HAZ and the weld are usually too narrow to provide any information regarding the elongation experienced before failure even if one of them is the weakest link.

If ductility is the concern, an all weld sample should have been tested or simply the base metal tested.

The fracture will provide some evidence of brittle cleavage or ductile failure, but there is no measure of the failure mode.

To the best of my recollection there is not a requirement in AWS B4.0, ASME, or other welding standard that addresses the mode of failure in a transverse reduced section tensile test.

Best regards - Al

By Hojo

Date 11-16-2009 16:48

Thanks Al, I am just trying to address comments on about 13 weld procedures that have been sent back to me by a customer and they have written the word Ductile in the PQR's where we have just stated "Broke in base metal" for the tensile tests. I have been writing and qualifying weld procedures for 19 years here and this is the first time I have ever had anyone do this. All of our procedures are reviewed by our AI and he has never had a problem with it either.

Thanks for the response.

By jarcher (**)

Date 11-18-2009 01:19 Edited 11-18-2009 01:21

The only code ramification I'm aware of as to location of the break occurs in ASME. A 5% variance is allowed if the break occurs in the ductile base metal as opposed to the weld or the HAZ. If for example QW 422 said that the UTS was 100 K and your tensile broke at 98K in the ductile base the tensile would be allowable, whereas if it broke in the weld metal or the HAZ the tensile failed. The engineer that I use to review WPS's always writes "ductile base metal" in WPS's he develops to distinguish between heat unaffected base metal and the HAZ, which is, of course, base metal too. If the tensiles develop the full UTS, then break location makes no difference to whether the tensiles were acceptable or not. My engineer friend, as a matter of course, describes breaks in the heat unaffected base metal as being in the ductile base whether needed by the results or not. Sounds to me like your customer's man has lost sight of the reason for using the phrase "ductile base metal," and has made a fetish of its use, assuming you developed the full minimum UTS in the tensile tests.

By waccobird (****)

Date 11-18-2009 10:59

jarcher
Thank You
I knew there was a lot I was missing and you said some of it.

Marshall

By HgTX

Date 11-23-2009 19:23

They can't just write that in and neither can you. If you have no evidence of a ductile fracture (in the base metal or anywhere else), it can't be written on the test report.

It sounds to me like they don't know what "ductile" or "brittle" mean, and they think "ductile" means "broke in base metal" and "brittle" means "broke in weld". If that is the case, they are just plain wrong.

Sometimes the client is wrong. What you choose to do about it is something the code cannot help you with.

Hg

By jarcher (**)

Date 11-23-2009 20:23

If the original lab report says the the tensile broke in the ductile base, he is free to revise the PQR to reflect the statement and revise the WPS to agreement with the PQR. I've had to wage an educational campaign with the lab our company uses to report the exact location of the break as discussed above. If all his tensile pulls are above the minimum UTS, the location is moot in ASME anyhow, and, as far as I know, AWS testing has no provision for variance from the minimum tensile no matter what the location of the break.

By HgTX

Date 11-24-2009 17:53

"broke in ductile base" is still not the same as "ductile fracture".

Hg

By jarcher (**)

Date 11-24-2009 18:52 Edited 11-24-2009 18:57

Absolutely. But I don't think it was about ductile fracture at all. I really didn't understand what the first post Hojo made was referring to, but he clarified when he responded to Al enough that I could grasp his problem. As I posted in response to js55, the only place I find ductile/brittle fracture examination commonly used in testing is for Charpy specimens under %Shear, and that's not required by either ASME or AWS either, its simply a standard examination for the test. As I also mentioned to js55, it may be common in other regions to ask for ductile/brittle examinations of the tensile specimen breaks, but its not here and its not required insofar as I am aware by either ASME or AWS. As I responded to Hojo, assuming his tensiles met the minimum or above, there's no need for inclusion of the break location on the PQR as its not an essential or nonessential variable for any process. The fact that 13 PQR's were returned leads me to believe it is a customer problem rather than a problem on his end. Thirteen PQR's that are under minimum just by enough to be qualified by the 5% variance allowed is rather unlikely. Hopefully Hojo will come back and tell us what the resolution was.

By js55

Date 11-24-2009 20:46

How bigga regions we talkin about here?
The wheat fields of Kansas or maybe all of North America from the Chihuahuan Desert to Ellesmere Island. :)

By js55

Date 11-23-2009 21:10

If my understanding of the original post is correct then all they are looking for is a visual examination of the fracture to determine if it appears ductile or brittle, and to have this reported on the PQR. Not uncommon for tensile specimens. This is consistent with Al. And consistent wih Hg, it is not a code requiremnt in any code I know of.
At this point in time you cannot re-create the fracture appearance.

By jarcher (**)

Date 11-23-2009 22:45 Edited 11-26-2009 15:37

I don't have Section IX on hand, as I'm on vacation this week, but I'll give a you reference, as time permits, to the variance I spoke of when I return to work. In my experience the only place appearances of fractures are reported is with charpy specimens under %Shear, but I recognize there may be regional variations about whats typically reported when not required by code. As I've taken pains to make clear, if tensiles meet the minimum in ASME, location doesn't need to be reported, but if your specimen is up to 5% under the minimum tensile strength, it has to be reported in order for the tensile to be acceptable.

I called someone to get the reference instead of waiting until I got back to work. The reference was QW 1.53b (page 5). I'm not sure whether he was using 2007 or 2010.

Ah, 2007, my misunderstanding, orders for 2010 are being taken, but it doesn't ship until mid 2010. I hope it meets its deadline better than D1.1 did in the last edition.