UTS Testing Anomaly

[Chris Morgan]

Hello,
I’m new to this forum and have an issue with parts failing tensile strength tests that I’m hoping someone can help me with.
We have two of lots of parts made from 7149-T73511 extruded bar stock that have been rough machined and then sent out to be annealed, formed and heat treated back to T73 condition. The material for both production lots is from the same mill heat lot. The two lots were annealed & heat treated in separately.

One lot yielded good tensile test results after heat treat to T73 and the other lot has significantly lower results but only in the “LT” grain direction. The tensile strength tests for the "L" direction is very similar between both the failed and passed lots. Our heat treat supplier is saying they did everything right and they need to start over.

See attachment for actual results.

Question 1: Does it make sense to have a large variance in only one grain direction ("LT") especially when compared to another lot that utilized the same starting material heat lot?
Question 2: Other than a testing process failure, what else if anything could cause the “LT” test results to be so much lower compared to the “L” direction?

Thank you in advance for looking at this!

 

[EdStainless]

So these came out of the same size extrusion, and were machined into the same size finished parts?
Were all of the extrusions made at the same time?
A heat lot is like telling me which hemisphere you are from, it doesn't narrow it down much.
Is there a controlled process on the extrusions?
Is there an 'LT' requirement on the extrusions?

I don't think that that second lot were made the same way. Different extrusion temp, or reduction ratio, or thermal treatment, or something.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Chris Morgan]

The extrusions and machining are the same.
The extrusion for both lots were extruded at the same time in a single lot.
The extrusions were made and machined in a controlled process. The extrusion as received from the mill includes rolled lot number marking and this was verified prior to starting machining. All parts were serialized for process control and trace-ability.
The requirement for tensile verification after heat treat is on the engineering drawing and it requires a test coupons be taken along the length "L" and along the width. There is actually no specific mention to "LT" (or long Transverse) in the drawing testing requirements. The "LT" reference is how the test coupons were identified by the lab.

We are generating new coupons from the failed lot and will be sending them out to rule out a testing anomaly. If the results match, this puzzle will continue as to why we have so much variance but only in one grain direction.

 

[EdStainless]

Not knowing the shape of the parts, is it possible that they could take LT coupons that have different orientations wrt the original extrusion?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Chris Morgan]

Good question and I should have noted this earlier. The engineering drawing identifies a specific location and orientation at the end of the part for the coupons. The coupons were taken from the same location on both lots.

 

[isoca]

How many tensile tests have been performed? Have you observed the same difference in all tests?
Were the samples taken from the same bar section (back end, front end or from the middle?

 

[Chris Morgan]

There were two specimens tested from the same end of (1) part in the lot. We are currently running a second set of tests to validate the first test. Test results are due Monday (1/27/20).

 

[Chris Morgan]

So we received the test results back on the re-testing and the parts passed. We tested two sets of specimens and both resulted with nearly identical results. Our supplier has now conceded that there was a testing anomaly despite their previous insistence that we needed to pay for re-processing of the parts which would have been $14,000!

For those who provided input to this issue, thank you for your time.