Chemical analysis tolerance 1

[MattG44]

My first thread so please go easy on me!
So we have a raw material (stainless steel) with an element tolerance of 3.0 to 3.5%, the Mill cert supplied records the element at 3.51" which exceeds the tolerance range in the specification. Our supplier is arguing that the material is compliant based on:
It is not outside tolerance at 3.51% as the specification is only 3.0 -3.5 ie to one decimal place.
Therefore under SI rules anything up to 3.54 would be rounded down to 3.5 ie acceptable to specification

My stance is the tolerance is absolute and as such the material supplied would be considered non conforming.
Would be grateful for any opinion on this.

Matt

 

[HPost]

Stand your ground...

It's outside of the tolerance and that's all there is to it.

The lack of zeroes after the 5 doe not given them license to supply duff material. If the tolerance was 3.54%, it would say 3.54%.

 

[Don56]

Depending on the general governing specification I might agree with your supplier. If this was ASME material, say a carbon steel governed by SA-20 and we were discussing thickness under tolerance we are limited to 0.01" under nominal thickness. There are multiple interpretations that allow the material to be up to 0.014" under as the ASTM E29 rules for rounding apply. Check the governing specification for the stainless you have.

 

[rb1957]

nah, 3.51>3.5.

but on the other hand if it has properties that meet the spec then why not ?

another day in paradise, or is paradise one day closer ?

 

[Nescius]

Just a question...What is the certainty of the 3.51% measurement?

 

[EdStainless]

No, the mill is likely correct.
-They are not doing themselves a favor by reporting the extra decimal place. What they report should match the specification. But even then the value rounds to an allowable one, you can't claim that the spec says 3.50 when it doesn't.
-Is it a heat (melt) analysis or a product (check) analysis? Every spec system that I know of offers additional tolerance for a product check.

That said, we require product checks on all incoming material, and we require that they meet the specs with no additional tolerance. Some of our purchasing requirements are even tighter. I have customers that I know will test and I want some room. I also differentiate between elements that I really care about (N, C, Mo) and ones that are of little significance (Mn, Ni).

A range of 3.0-3.5 is very tight, If this is the Mo in a duplex grade take it, be glad that they are shaving it low.

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

 

[MattG44]

Thanks for the feed back so far.
to clarify the affected element here is Ni, the 3.51% is the melt analysis.
Unexpectedly the product check analysis has come out at 3.6% for Ni which is concerning (I have been informed that product check has a permissible extra tolerance of +/- 0.07%)

 

[EdStainless]

Who did the product check?
We have had very poor results with many professional labs. they just don't have good enough standards for working with stainless steels.
You need to look at he check tol again.
If this material was bought to the AMS spec then the Ni would 3.50% and product check tol would be 0.07% on top of that resulting 3.57% max.
But if it was bought to ASTM then the Ni would be 3.5% and the tol is listed as 0.07%, but it would have to be rounded giving 3.6% max.
You can't hold someone to tighter resolution than the governing spec.

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

 

[MattG44]

Product test was carried out by an accredited lab by Optical emission spectrometers (OES)
, which test process would be you suggest?

 

[EdStainless]

I would like to see the statistics.
Do they test this alloy for you all of the time?
Can you send them 6 samples from different heats and have them check the Ni?
We did this, but we used 26 samples. Most labs came back with a spread that was more than 3x the mill results (heat and check) and the labs showed different samples as the high and low ones.
Like I said, I am picky.
I don't really care if they use XRF or OE. XRF can have surface finish issues, and OE depends on the excitation method.

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