how many digits after the dot are required in chemical composition ? 5

[stanislasdz]

Hi,

My question seems to be "simple" but I can not find an answer.

The specification of our costumer in a steel alloy composition is :

0.55 Cr % Max

But our chemical analysis shows the value of 0.552 % of chromium.

• . If we around the value of 0.552 to 0.55 we met the requirements.

• . If we dont around the value of 0.552 we don't met the requirements.


• Specific question: what is the right thing to do in this case ?
• General question : how many digits must be required in the chemical composition specifications ?

Thanks in advance ?

 

[EdStainless]

There are ASTM specifications that addresses both of your questions.
Depending on the product specification there is a required number of significant figures.
There is also a tolerance on a product chemistry.
I am sorry that I can't give you spec numbers but I am not in my office.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[stanislasdz]

Thanks EdStainless

I 'am waiting.. when you came back to your office send me the ASTM number of the spec

Stan

 

[CoryPad]

The standard is ASTM E 29 Practice for Using Significant Digits in Test Data to
Determine Conformance with Specifications.

You need to round your data to match the same significant digits as the requirements. Then, you can determine conformance. Your example is in conformance.

 

[arunmrao]

First be clear if you are comparing product analysis with bath analysis. If so, you are then making a mistake. There is a tolerance value given by ASTM for different alloying elements and in different ranges.

Rounding off to the nearest significant decimal place,would depend on its relation with physical interpretation. For eg C 0.031% is significant as compared to 0.035% . You will not be able to round off.

But Cr 0.31% or 0.35%,should not really be a matter of concern.

I have not failed. I've just found 10,000 ways that won't work." — Thomas Edison
_____________________________________

 

[stanislasdz]

Arunmrao said:

First be clear if you are comparing product analysis with bath analysis. If so, you are then making a mistake

Why ? we always give the analysis of the bath to the costumer. On products the macrosegregation can give wrong indications !

Arunmrao : if you said it depends on physical proprieties I can agree, but it's a subjective interpretation and can be out the specifications.

 

[EdStainless]

and we always demand a product analysis because that is what we are working with.
This is from a stainless spec
8.2 A product analysis tolerance of Table A1.1 in Specification A 480/A 480M shall apply. The product analysis tolerance is not applicable to the carbon content for material with a specified maximum carbon of 0.04 % or less.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[SMF1964]

Add to this discussion the QA limits of the instrument used perform the test. Just because the computer system spits out 0.552% does not necessarily mean that the instrument can accurately measure a variation of +/- 0.001%. What do the detection limit/accuracy/precision studies show for the instrument?

 

[salmon2]

I went through the same question before. The standard we were using has a table having max values for several chemical elements and also it defines a table for +/- tolerance of each element. Some people said max is max which means no tolerance, sharp. Some people said: hold on, it got a tolerance here...

Finally today, I got the answer for my long time puzzle question.

By the way, the heat/batch analysis could be slightly from product analysis. But you should use product analysis since that is the finished product you got as others said already.

 

[deadrange]

Stanislasdz:

Having been in your predicament before I know exactly how you feel. The way I look at things when it comes to chemistry is to go back to the customer specification. If the customer has spec'd out Cr to 0.55 wt % as you have indicated then anything up to 0.554 is acceptable. However if the specification is written at 0.550 then clearly we have to follow the guidelines written.
That is just my impression on the matter.

 

[arunmrao]

dead range et al
One should check if the customer has specified a part to be supplied conforming to some standard.

If he does and also adds in a tabular column the chemical and mechanical properties which might not necessarily be what is specified. In such an event,which acceptance norm should be followed,



I have not failed. I've just found 10,000 ways that won't work." — Thomas Edison
_____________________________________