Identifying a material grade based on PMI?? 1

[MetalFizz]

Hello great minds,

I have been sent a PMI report for a component (spacer) which forms part of a large motor (taken from offshore floating structure for repair). I have been asked to identify the material grade of the spacer so that a replacement can be purchased. I do not believe a PMI report is sufficient as it does not provide much information on the mechanical properties.

What would you suggest is the quickest and most low cost route to understanding which grade of material we have?

Thank you!

 

[weldstan]

Ask the manufacturer for the identity of the material.

 

[bobjustbob]

You cannot determine the grade with just a chemistry check. Depending on the method used for the PMI, you may not even get a full chemistry.

What type of material (steel, iron, copper, etc.) is it?

Bob

 

[MetalFizz]

I believe the vessel was built pre-1999 and the manufacturers information is not available.

@bobjustbob the PMI analysis has provided 3 'matches' with steel grades. But this is a heavy duty motor and I do not plan or considering these 3 matches as gospel without further testing.

See below:

 

[Tmoose]

As best I know PMI does not detect carbon.

If I don't know the % carbon I don't think I can know much about the steel.

 

[bobjustbob]

How was the PMI performed? If it was done using an XRF analyzer then all you know is that it is a steel with about 1.10% Mn. It will not give you carbon, and that is the most important hardening element in any of these steels.

There are methods that can give you carbon like using an OES. I believe there are handheld LIBS units that will give carbon as well but I don't have any experience with these.

If I had to figure it out, I would get a chemistry using an OES and the carbon with a combustion method (LECO) and then get a surface and core hardness. That would probably give you a good idea of the material and how it was heat treated. Depending on the hardness results I might want to look at the microstructure as well.

Bob

 

[EdStainless]

Bob is right on this one, OES or LIBS will give you C (and other light elements) and allow a real determination.
Most all of these devices also have higher precision mode for better accuracy. It usually takes 5 times as long to run.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[mrfailure]

1: I don't understand what the numbers next to Mn denote. Are they some kind of expression of %Mn or are they a confidence rating?
2: For what you have, you need to present all the results you got with PMI. If the material was 12L14 then you should have lead, while 1117 should have at least 0.08% sulfur. This is how you differentiate with at least these alloys. In other words, even without carbon analysis you can still differentiate between resulfurize and leaded steels and at least know what class you are dealing with.
3. Concur with Bob: you should use carbon/sulfur, OES, and/or ICP quantitative techniques to get the right composition (PMI is semiquantitative and not necessarily the most accurate technique). If you cannot get the sample lab tested, then you can at least use replication of the surface to determine if it is at least resulfurized and/or leaded.

 

[kingnero]

and if the sample is too small to perform tensile tests, get at least some hardness readings so that you can determine its metallurgical state.

 

[Dhurjati Sen]

Ask for the data sheet for the motor or pump. Find out material for construction of the spacer and place an order.
It is impossible to predict the material grade of a critical component by PMI or even chemical analysis plus destructive / non+destructive testing.

DHURJATI SEN
Kolkata, India

 

[kingnero]

It is impossible to predict the material grade of a critical component by PMI or even chemical analysis plus destructive / non+destructive testing.

OES + tensile (and/or hardness occasionally) is bread and butter for us though, and seems to work fairly well...

 

[Tmoose]

I think our decent, but Carbon blind, XRF gun accepts "steel" as an operator input and then tries to make a match to steel.

Some machines have spacers made of ductile iron. Without know/confirming the carbon carbon I think that that is a possibility.

A well intentioned designer XRF-ed a big broken casting and chose a good looking section for a few tests. It ended up being a nickel based weld repair.
The test was to determine the scrap value we'd pay the owner/customer so they would save shipping it back to them.

https://www.metalary.com/scrap-metal-prices/

I think they would have been delighted If we had paid them the nickel price .