My understanding of critical flow devices is that the flow rate does not change with a change in the downstream pressure, as long as the presure ratio across the orifice is ~ 0.53 (depends on the gas mix).

The flow rate does depend directly on the upstream pressure - so if your gas supply as at a fairly constant pressure, this should work OK

I am facing the same problem. The minute someone sees porosity their first reponse is to crank up the flow rate. It is not unusual to see 70 cfh. I was curious as to how the Harris surge guards worked and how much they cost. The company installed critical orfices before I started there. They seem to work OK, but several have already been drilled out. We have reinstalled flow regulators on some of the equipment. The only thing I do not like about the flow orfices is you lose the ability to control the gas flow rate. Also, the gas supplier decided on the flow rate (naturally) and chose the orfice size (no one at the plant new any better before I came). I am not sure of the best solution, but if anyone knows a method that works well, let me know.

About 15 years ago, I spent a considerable amount of effort on the same problem. Basically, flow meters and restricting orifices are the same thing, except that the flowmeter is an adjustable orifice with a flow indicator, but the problems are the same. They are very dependent on both upstream and downstream pressure. Even a distance of only ten feet through the small tubing and orifices found in most welding guns will result in a measurable drop in flow due to back pressure. And, you've probably already figured out there's no way most gas manifold systems will deliver constant pressure at every outlet. The only way you can use flow meter readings or restricting orifice sizes reliably is when using identical equipment under identical pressures, a situation difficult to find in most production situations. Since welders can't see what the gas is doing, they have to be able to hear it or feel it, or be able to blow-dry their hair with it. At one time I tried mounting flowmeters right on the wirefeeders to allow them to see how much was flowing and eliminate the variable of the hose back pressure out to the feeder. I tried orifices, too, but they would just drill them out and replace them, so you couldn't tell what they had. The bottom line is that if a welder has a problem that he thinks is gas-flow related, he will find a way to adjust it to where he wants it. If he is turning out good welds, the cost of the gas is a lot less than the cost of labor, so there's not a lot of harm in it. If he is turning out bad welds, then it may be gas-flow related, or it may be some other variable in the procedure, maybe even one that is inconsistent or related to some obscure variable such as gun angle or electrode extension. I once found the source of porosity in an aluminum welding setup, for example, was an oversized inside diameter on a contact tip which allowed gas to blow right in to the arc plasma and disturb the transfer. Other sources of porososity can include mill scale, "weldable" primers (that's an oxymoron), coatings on the filler wire, and even elements (or a lack of them) in the base metal. I finally concluded that the best check was a daily fillet break test, since we couldn't control all the variables all the time in production, and a rigorous follow-up on required x-rays to find out who wasn't doing a good job and get them retrained. The answer to your problems is probably in training, not equipment. It is not the sheilding gas that needs to be controlled.