PMI - XRF: Dealing with third party field inspections

[Geof]

I love handheld PMI and I hate handheld PMI.

A recurring theme over the last ten years for me - dealing with customers using a third party inspector to perform PMI in the field.

My company manufactures and supplies cast pipe products in various alloys.
We have two handheld XRF analyzers that we use to verify materials match the MTR's provided, very important with some of the expensive alloys we work with.

The issue - after we've done this and provided documentation with our products, the customer or end user will very often also perform field PMI.
Usually we never even hear about it - no issues. However, several times a year, we end up in a situation where the field readings don't match up - and the pain begins.

Yesterday, a customer provided a third party inspection report. This was covering four units cast from the same heat number and shipped out one or two months ago.
(13) of (16) shots were judged "acceptable". Three read low on a couple of elements, and they are concerned.

I noted in an email response that all shots were taken from the same Heat of material and requested that they re-inspect the spots in question using a longer observation time - they had done 15 sec shots while we typically take 1 minute. Cleaning/grinding is also recommended to get below any surface corrosion or contamination.

Bottom line - how do you handle this situation:
* Single lot of material
* Certified MTR's.
* Our own inspections provided.
* But three out of sixteen shots taken by someone out in the field that we can't speak with put the materials in question.

Kind of a rant - but wanted to see if anyone else has had to deal with this and how they handled it.

Thanks,

 

[EdStainless]

PMI is for alloy type verification not the determination of chemistry.
Internally we have procedures for both.
The type verification is a 10-20 sec shot.
If someone wants chem the test time is based on actual x-ray counts (of the lightest element, often 200sec or longer), requires surface prep, and also requires before and after shots on two reference standards of the same grade.
When we hear that a customer wants to use a third party I ask if they have reviewed and approved the procedure, and offer ours as an example.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

Three read low on a couple of elements, and they are concerned.

That is because the 3rd party inspection company has presented the results in a manner that suggests equivalence to exact chemical analysis. These companies over-hype PMI because it is a very lucrative line of business, and have trained their clients to expect more than PMI is capable of doing.
'
PMI is for alloy ID only.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[weldstan]

It has been my experience that third party performance of PMI with XRF has been generally acceptable for alloy ID only and in some cases has not even been acceptable for that. I have used third parties to perform PMI with portable light spectroscopy equipment with argon and vacuum shielding to perform exact chemical analyses as compared with laboratory equipment. I have not compared the portable equipment analyses to wet chemical analysis of same specimens.

 

[bobjustbob]

Part of the issue is the lack of knowledge most people have about measurement in general and chemical measurement techniques in particular.

Many, when they see a test report they immediately believe that that number is correct, and any other test by another piece of equipment should yield the same number. Anyone who has spent time with a spectrometer, XRF analyzer or combustion type carbon/nitrogen analyzer understands understands that that is not the case. Calibration, sample prep, and the operator can all affect the final number. ASTM has product analysis specifications for a reason. Unfortunately there is nothing to do but be the educator when these instances come up.

I remember flying down to Houston because a CF3M valve was checked with a portable OES analyzer on the surface and showed a carbon out of specification. We had to meet at a third party where a piece was removed from the valve and ran on a LECO to check the carbon before they would accept the part.

 

[arunmrao]

PMI for cast pipe products is not a good practice. I hope, they are not tracking C and Si elements.Several times, during my investigationI have noticed that if the surface is contaminated, it could lead to erroneous results.

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[Geof]

Thanks for the feedback.

Does anyone know of a good white paper or article discussing this subject? Something that I can present to the end user?

All the literature that I am finding is either very technical or hyping up the capabilities of the method. I’ll dig into it more When I get back to my office.

 

[priyam07]

If the issue is bothering so much, how about using a spectrometer in house and send the results to your client. As said XRF is for alloy ID you should not be questioned based on the levels the client sees. IMO the client's concern is not valid after you send MTR results unless they used a third party lab and verified the readings.

 

[mrfailure]

PMI is a portable X-ray fluorescence analyzer (XRF) that is semi-quantitative in nature (like EDS in the SEM) and among its limitations it is unable to measure light element content (also like EDS), even if you see a reading for such an element like carbon. This is why, as our colleagues mentioned, it is
usable only as an alloy identifier, and not for conformance. Surface preparation is important here - I usually use a Dremel tool to remove surfaces scale and contaminants. Otherwise, results will be inaccurate and inconsistent.

 

[Geof]

All,

Thank you for the responses. I've been out due to Hurricane Florence.

Customer is actually starting to understand. With four pieces cast in the same heat, he didn't realize that the inspector was basically testing the same material over and over again.
Seen in that light, 13/16 15-sec shots confirming the material grade and the MTR is a pass, not a fail.

He's still looking for some more assurance - A statement on our letterhead.

This is not really resolved yet because he still has the end user to deal with, but we can hopefully get there.