Varaiation in material properties 1

[hartmanb]

We are running independant testing on materials that we recieve to check the accuracy of our vendors' MTRs. Obviously we don't expect every number on the original MTR to match the results of the independant testing 100%, but how much variation should I expect?

For example, we sent a 4130 bar specimen for testing, the yield strength came back at 72500psi, while the mill's MTR shows 83000psi. The UTS, elongation, and reduction of area all came back lower than reported on the MTR as well. FYI the bar diameter is 6".

 

[unclesyd]

What condition was the bar in?

Where was your sample cut from, near the edge or the center?

 

[hartmanb]

The condition is normalized, hardened, and tempered. Both the mill's MTR and the independent labs certificate state the testing was performed at the half radius.

 

[Wrenchbender]

Too much variation. Something is wrong assuming this bar has been hot worked from a cast ingot. Make sure both labs took the specimens in the same direction, e.g., both longitudinal or both transverse. Then time to take a closer look, literally, with metallography.

 

[hartmanb]

Good point Bestwrench, the original mill MTR is longitudinal, and the lab was transverse. Is there a general rule that I can use to relate the two, or do I need to send another specimen requiring longitudinal?

 

[EdStainless]

The yield may be off a lot because of test methods but the UTS and elong should be closer than that.
Did your lab pull multiple bars? single tests don't say much.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[hartmanb]

Ed, the UTS, elong, and ROA are much closer than the yield. They are all about 5%-7% low, compared to the yield being around 15% low.

We are using drops from our machine shop to send to our lab, so unfortunately we don't have alot of extra material to send off for multiple tests.

 

[metengr]

I disagree, the yield and ultimate tensile and elongation can vary considerably because specimen orientation will effect mechanical test results, period. This is why you should follow and established test protocol for comparing results.

 

[arunmrao]

This is why you should follow and established test protocol for comparing results.

Agree with your comment metengr. Just for reemphasizing and reminding us this basic tenet a star comes your way. Thanks

 

[Wrenchbender]

No admonition from me, hartman, 'cause I been there, done that too, and sure it won't happen again. To answer your question about a general rule relating the two directions - no, nothing quantitative that i know. Depends on the chemical purity of the steel 'heat' and the grain size & morphology. (Readings should get closer with higher purity, and smaller equiaxed grains.) If you trust both labs, you should have no problem for now.

 

[EdStainless]

Can you go back and ask for more data from the mill? My guess is that they reported an average of some number of tests. It would be good to see what range they actually see.

I believe that Hart said that these were all longitudinal mid-radius samples. I am not concerned about that aspect.

I work with tubing and the specs there don't require many samples. We usually pull double the required number so that we can see what variation within a lot there actually may be.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[btrueblood]

Ed,

No, he said the mill did longitudinal samples, but the other lab did transverse samples. Metengr is right, the pull tests won't necessarily match on rolled bar stock in the two opposed directions.

 

[hartmanb]

We are trying to verify the honesty/integrity of the mill, so asking them for more samples wouldn't be a good solution. Thanks to everyone for the input, I will have to get another test coupon to have a longitudinal test performed.

 

[redpicker]

We are running independant testing on materials that we recieve to check the accuracy of our vendors' MTRs. Obviously we don't expect every number on the original MTR to match the results of the independant testing 100%, but how much variation should I expect?

You won't get a definative answer to your question because you are mis-understanding the MTR.

The mechanical properties on the MTR do not represent the minimum, maximum, or even average properties of the material. Rather, they merely demonstrate that the material meets the specification requirements. You should expect your re-test results to meet the original specification requirements, but that is all you can expect.

For example, if the specificaiton the material was produced to required a minimum yield of 70,000 PSI, then both the MTR and your tests are in agreement. If, on the other hand, the material specification was a minimum yield of 80,000 PSI, then your test would indicate that the material is not in conformance.

Consider the producer. He heat treats the bar and takes a test from the end (since that is where he can get a sample without cutting the bar in two). He knows that the properties from the end will be higher than from mid-length, but his experience tells him that if he gets better than, say, 80,000 PSI, then any test from mid-length will be above 70,000 PSI. Therefore, he certies the material as meeting the specification and reports the actual result.

Now, consider the same producer, but his requirements are now 80,000 PSI minimum yield and he pulls the same test and gets the same result. If he uses that test to certify the material meets a minimum 80,000 PSI, then he has a problem with process control and he is producting material that does not meet the specification requirements.

Now, a lot of warehouse suppliers will take the MTR and make the assumption that it is a certification of the minimum properties, but is is not. Most producers are very careful to make sure the sepcificaiton they are certifying to is on their MTR. This is why.

Now, if you are asking if it is suprising to have 6" solid 4130 bar that has been normalized, quenched and tempered to have a yield strength of 83,000 when tested at mid-radius on the end of the bar and have the yield, UTS, and % El all show less when tested mid-lenght, then I would say, no, not suprising at all. In fact, that is what I would expect. 4130 is not very deep hardening and in a 6" bar, so I would expect the "end effect" from quenching to be rather pronounced.

rp

 

[unclesyd]

4130 steel has anisotropic properties aligned with the direction of rolling or forming. The greatest variation is seen in impact testing though elongation and reduction of area can be affected, not so much as the tensile. The hardness can be affected especially the larger the bar.

The statement above about the purity and morphology holds very true as the size, orientation and elongation of the MnS inclusions plays a big roll in the physical properties of 4130.

Also mass effect plays a big roll in physical properties of 4130. As stated above everyone has to be on the same page.

 

[TemperedMartensite]

Hartmandb stated, "We are trying to verify the honesty/integrity of the mill, so asking them for more samples wouldn't be a good solution."

Where is this mill located? In the United States fraudulent mill certs is a ticket to a 8' x 8' jail cell. I would be shocked to find out a mill knowingly misrepresented their certs.

 

[redpicker]

In the United States fraudulent mill certs is a ticket to a 8' x 8' jail cell. I would be shocked to find out a mill knowingly misrepresented their certs.

I am shocked, shocked to find out gambling is going on here - Captian Renault

What criminal statutes would apply? Certianly there are liability issues, but how are you going to throw a mill in jail?


rp

 

[unclesyd]

I don't know rules now but at one time one was to civil and criminal penalties for incorrect information of anything tied to the Form 20.
WE had an incident where a company was building or repairing rail tank cars for us and was supplied supplied 304SS instead of required 316L SS. The leaking of one of the tank cars precipitated a full scale investigation by the government. I don't recall anyone going to jail but some hefty fines were paid as well as a considerable sum to our company.

http://www.phmsa.dot.gov/inspect-enforce

 

[TemperedMartensite]

Anytime you purposely mispresent information on a contract you are committing fraud. Mill Certs are, in essence a contract, they state that certain conditions were reached and the specifications were adhered to.

Who within the mill would go to jail? Start with the cert auditors and the quality control personnel. If it is seen to be systematic through the organization - then company presidents would be charged.

Has it happened recently - from a mill? Not to my knowledge, since the penalties and risks are too steep. However, there are a couple of (and one very high) profile ceses of distributors getting caught on the wrong side of the law as they attempted to *cert-up* material. Federal indictments flowed and all involved are awaiting trial.