Current PMI Capabilities (Carbon Steels)

[PipingEquipment]

Hi,

The company I work for is looking at purchasing a PMI gun, I have been in contact with many suppliers but due to our specific needs the companies we have narrowed the choice to are Olympus, Niton, Bruker (in that order) and because of our need to distinguish carbon steel we are looking at the top of the line gun for each supplier.
I know great strides have been made in PMI limits of detection, due to rapidly advancing technology, allowing hundredths of a percent detection in light elements possible which it looks like to me finally makes the chemical analysis of carbon steels possible. I know carbon cannot be 'seen' still but all the other alloying agents should be able to be detected and this should allow confirmation of carbon steels possible, right? Due to our limited experience with PMI guns in general I was wondering if any of you have an opinion of whether handheld PMI guns are able to distinguish between, say, SA-516-70 vs A-36? We also deal in Stainless, Chrome-Molly, Titanium, Nickle, and other alloys but most of the time it is carbon steel and we are looking to use this tool to build our QA/QC, including checking incoming materials and checking the materials of welds, individual items (fittings, valves, welding rod, ect) and finished products that we make such as tanks, pressure vessels, piping lines, and other items like skids, and equipment.
Although carbon steel is one of the lowest cost metals, it is what we use upwards of 90% of the time thus my interest in this area specifically, and the type of carbon steel we use are structural, pipe, flanges, fittings, valves and plate of ASME/ASTM standards including pressure vessel grade quality (SA-xxx).


I look forward to your responses and thank you ahead of time!

-Kyle

 

[unclesyd]

Though I've been out of this aspect of the business I've seen No Joy in analysis of the Carbon Steels. In the lower alloy range they tend to lump them togather. The higher alloy materials do quite-well with and experienced operator using the proper standards.
On the lower alloys material we depend on the manufacturers to mark the steel by stenciling, not paint or crayon. Comparison has to to be made with the actual data sheet supplied, not typical. We have always take the right of first refusal on any materials.
In addition to MTR we have some hardness spec's on some materials and if possible will remove a sample for testing in the lab in lieu of field test.

On of the big problem with QC on materials especially plate is that will in the laydown yard someone cuts the stenciled area off and doesn't remark it.
We also stick with one or two suppliers and keep a good communication channel open.

 

[metengr]

I concur with unclesyd. PMI is a great sorting or QC tool to segregate carbon steel from low alloy steel or from stainless steel, and non-ferrous materials. Beyond this, I would not use PMI to re-certify material for code use or reliability-related use.
A sample from the unknown material would need to be removed to determine the full chemical composition and if necessary, hardness testing or even mechanical testing in the lab. This is the proper way to certify unknown materials. Once again, for simple sorting purposes to control inventory or to aid in welding carbon versus low alloy, PMI is fine.

 

[stanweld]

A-36, and A-516 cannot be distinguished by XRF instruments. Arc spectrographic, PMI instruments could be used to distinguish fine grain melting steels like A-516 from coarse grain melted steels. Understand that A-36 may be fine or coarse grain melted as well as a number of other carbon steel specifications.

 

[JR97]

Another area of difficulty is coated welding rod. If you PMI the bare end, your chemistry will appear incorrect. The weld generated from the weld rod contains alloy elements from the flux. You really need to put bead on plate to determine if the correct rod chemistry.

 

[PipingEquipment]

Thank you for the info guys.
I did a little case study of MTRs we have gotten and compared the low and high numbers that occurred and it appears there really is no way to determine between SA-36 and SA-516-70, by the machine alone. However we will be able to check them against an MTR and make sure they are in the range they are suppose to be and that vendors are sending the right stuff.
One rep told me yesterday that the carbon steels are the holy grail of the handheld XRF industry and they have MIT grads working on it for nearly a decade and still not able to crack it.

 

[stanweld]

XRF is useful in sorting micro-alloyed carbon steels containing small additions of Nb, V, Mo, etc. to improve toughness and tensile strength. We used it 40 years ago in our steel plate mill with great cost effect.

 

[XL83NL]

For handheld solutions PMI guns using XRF technology are an option, but as pointed out by unclesyd and metengr, they're not really what you want when you're trying to detect small differences in C-traces.
Not sure if it's been mentioned here [maybe by stanweld?], but another option you may look into is PMI through Optical Emission Spectrometry (OES), which can detect C, S and P traces. From what Ive read, it's capable of measuring the difference e.g. between a 304 and 304L. Although it has a 0.005% less reliability, it may be useful for base materials. Disadvantage of OES is that is leaves a small arc on the tested sample, and therefore may not be useful for testing welds. Im also not sure if it can distinguish A36 from A516-70 by means of e.g. integrated databases for material comparison.