Gas Mixer

Date 07-26-2012 23:38

Here's how I'd approach it. Nominal is just the label on the bottle, not exactly what's in it. Your measurements are "actual", not nominal (IMO if I'm understanding your post correctly). As an example, pull the gas specs from the local gas supplier. For any given mixed gas, they'll spec the actual composition of any given bottle to be the labeled nominal composition +/- some percent. For a bottle of 50% Ar - 50% He, it might be +/-5 %, or +/-2% or whatever, so while the nominal is 50/50, the actuals might be 55/54, 52/48 or whatever. To give you an idea of what sort of "universe" of tolerance you're working in, find a comparable commercially available gas mix and see what it's tolerance is. If the tolerance of your mix is within what they specify for theirs, then you have some good evidence for your non-conformance report to support the welds that have been produced are good. If the tolerance of your mix is outside what industry is using for a comparable gas, then you probably have some decisions to make. Basically, since there is little guidance with in the code for this, I'd use the industry accepted nearest standard as the yardstick. And then once this is all sorted out, it'd be a wise idea to write it down and roll it into your quality program/procedures.

By Metarinka (****)

Date 08-05-2012 05:43

from a technical standpoint, depends what the major and minor gas is. I'm guessing 84 He, 16 Ar? below 25% gas conductivity (penetration, bead shape) is very directly correlated to composition so small changes can make differences. easiest way is to make a few test welds cut them up and look at the macros to see what the penetration or heat affected zone looks like. In all reality if it's manual welding it probably wont' make a big enough difference.

By Solluz

Date 07-26-2012 20:00

When something goes awry like this, it's not uncommon to make a technical judgement call. In our business that's called writing up a "non-conformance" and making a disposition. If you did not observe any welding difficulties, you might simply make a disposition to "use as is". To support that decision you could do some additional inspections also. If you or anyone else in your organization feels uncomfortable with "use as is", you could consider also post-qualifying a procedure with the out-of-spec gas to justify its prior use. Regards, Sol

By weld7320 (*)

Date 07-27-2012 01:53

Sol and Mike, thanks a bunch for your insights. I'm going to go with both your thoughts and see what we can come up with. I appreciate the time.

By DaveBoyer (*****)

Date 08-05-2012 05:46

While on the subject, does anybody have any experience with a Smith gas mixer?

I have a few, but never messed with them.

By fschweighardt (***)

Date 08-05-2012 13:49 Edited 08-05-2012 13:51

Are you working to a code. You reference D1.1 (no edition). D1.1 will refer you to AWS A 5.32 for required tolerances. what you describe is outside the tolerance allowed for in the code, so then you have to make an engineering call.