Variations in chemical analysis

Date 01-08-2008 00:03

I would take a look at your consumable's classification and then take a look at the applicable specification to see what the allowable range is for that classification. As for the analysis being off by 0.1 or 0.01 and its impact it really will depend on the element...

By CWI555

Date 01-08-2008 00:49

E29 "Standard Practice for Using Significant Digits in Test Data to Determine Conformance with Specifications"
.1 or .01 can make a big difference depending on the element in question, but as a mimimum, the difference between absolute and rounded values needs to be understood.
Either or, if the analysis is off spec, it's a mute point from my perspective. If the material as spec'd is out, is should not make it to the welding stage.

Regards,
Gerald

By dmilesdot (**)

Date 01-08-2008 12:25

Here is where I am coming from, we require a chemical analysis on the all weld metal tension sample from a PQR. For SAW we have a chart that lists the elements that we are looking for and either a maximum or a range that these elements must fall between. The two that are the most problematic are Copper (Cu) and Nickel (Ni). Our acceptable on Nickel is 0.40-0.80 and the Copper is 0.30-0.75. Occasionaly I will get a PQR in with the Nickel at .38 or the Copper at .29. By our code (NYS SCM) I cant accept these because they are out. But what does this small percentage really mean to the weld and or base metal? How are these ranges determined? I understand that there has to be a cut off, but how is this cut off determined?
Thanks
Dave

By CWI555

Date 01-08-2008 18:45

While it has been many moons since I've had any dealings with nys I recall them being slightly more restrictive in regards to allowances than what is usually allowed. Each of the givne alloying constituents will have a specific range in which it will do it's job. I believe those values initially came from empirical test, experience, and failure analysis, but now it is a definitive range. In the ranges you are working with.. I doubt very seriously it would have an adverse affect on the weld given most of those required values will have a safety factor engineered in. (usually a fairly substantial one)

If your going to question NYS SCM on this, your going to need more than a few forum post to back yourself up. I think you would have better luck convincing chuck schumer all guns are good and every american should carry one.

By js55

Date 01-18-2008 17:51

LMAO!!!
Gerald is right.
Two problems here. One is easy, relatively, the other not so easy.
Will that little bit of variance in Ni or Cu be a problem? Its hard to imagine it would. But you have to engineer it. In other words, get some data to support it.
The other problem? Convincing a bureaucracy. See Gerald above.