Diluted Weld Deposit Chemical Analysis 2

[Rideout]

This may be a basic question but, I am fairly new to welding chemical analysis.

I am using an XRF spectrometer to measure weld deposit on a 3/4" NPS A182-TP316L socket weld (as per client specs).

I am wondering what code and table the analysis should conform to.

Section II Part A is for Pipe and fittings, Section II Part C is for Undiluted Weld Deposit, and Section IX A number 8 is for procedure qualification.

I don't know which is applicable to the completed weld in this case.

Thanks

 

[metengr]

Rideout;
What exactly is the customer specification - to verify a 3xx series filler metal was used for the socket weld? Otherwise, XRF is a qualitative analysis and should be used as such for typing or sorting of metals, not for reporting chemical composition results.

 

[Rideout]

Yes, it's to verify the welding rod used.
I'm just wondering also, how much will the dilution from the parent material effect the analysis of the weld?

 

[metengr]

Depends on the size of the fillet weld. Typically, for 1/4" size or greater fillet welds, the face of the weld should provide good results with minimum dilution effects.

 

[Rideout]

So here's my situation, I have A182-316/316L material welded with a WPS for P8 to P8 and F6/F5 and A# A8. The rods used were ER316L and E316L-xx.

When the weld cap is scanned with XRF spectrometer what material specification should the Cr, Ni, and Mo results be within, Sec 2 C or Sec 9 A8 chemical composition.

I can't seem to get a definitive answer from the client or anyone else within my company.

Help would be greatly appreciated.

 

[SJones]

Consult API RP 578, Clause 6 and start with ASME II SFA-5.9 in the centre of the weld cap until something goes wrong - then it's up to the owner.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[metengr]

What I would do, is have a small coupon made of bead on plate using the stated filler metal. Using this coupon, XRF the weld surface and use this as a standard for comparison with what you will be checking as produced. I have used this method before with good results.

 

[Guest102023]

If 'XRF spectrometer' refers to a PMI device then you are not performing chemical analysis, you are doing alloy identification. Even with that you cannot report the approximate readings for individual elements (as my former employer fraudulently insisted on doing).

 

[SMF1964]

Brimstoner, Please explain: you cannot report the approximate readings for individual elements? The software in my portable XRF unit provides a breakdown of detected elements. If these values are not accurate, how is the system differentiating between, say T9 (9Cr-1Mo) and T91 (9Cr-1Mo-¼V)? Or are you saying that because the elemental list is incomplete (no carbon, silicon, nitrogen, lighter elements that don't make it to the detector, etc) it is not a complete chemical analysis?

 

[SMF1964]

To answer the original question, we've seen iron dilutions up to 20% Fe content in the weld overlay made using a nickel alloy that contained less than 6% Fe in the original wire. Since it was a weld overlay, it's not a matter of getting compositional data from the surrounding base metal because the detector port is larger than the weld bead.

 

[deco0404]

Rideout,

I have to agree with brimstoner here, I do a lot of PMI work using xrf equipment both in power plant and chemical plant. One of the main problems I come across quite a bit is identification of 316 elements. You say that the filler is 316L, but the chemical composition of 316, 316L & 316H are all pretty much identical apart from Carbon content, which xrf (unless you have the latest and greatest versions) do not pick up. Also if your equipment cant tell the difference between 9Cr & T/P91 you might have a problem with the equipment.

I would also agree with meteng here in that the effect of dilution rates are marginal at best.

Declan