Hi Kam,
What we do is check the output yearly to satisfy the aditor that we are in fact running within the ranges shown on our WPS's. They usually want you to have a written procedure stating how often you check it and how you went about it. This also helps your welders know how much to adjust if it needs tweaking to stay within the WPS. Also note that after the cables get hot the variables will change some, so we check them after they have already been run. Check your stick machines close to the stinger to avoid the drop in the cable. This gives a little more accuracy to your readings. Calibrated power supplies won't make a better welder out of you, but it will prove to the auditor that you are within your WPS.
They go through my shop and look around then spot someone to question as they are lifting their shield. Questions to the welder usually are: Can you show me the WPS for the weld you just ran? What are you going to look for once you chip off that slag? How much preheat did you have in that joint before you welded it? While the welder is answering him, he is looking for the tempil stick crayon marks where the welder checked it before he welded it. Then he asks the welder after four hours, say you haven't used all of those rods in your pouch, what are you going to do with the ones that are left? Lots of questions to prove that you have trained your welders to read WPS's and know something about the job they are doing. It never fails they always can find the newest welder in your shop every time, so train them before you put them into production work. At least when he gets through you will have some idea where your strong and weak points are for the next year's audit.
Hope some of this helps you to get going in the right direction. I'm sure there are a lot more others can add that they have experienced in these audits.
John Wright

Personally speaking, I too agree that "calibrating" welding machines doesn't do much to help or hurt our weld quality. At least not for the parameters and processes we normally run. Especially for D1.1 work where parameters must be within manufacturer's specs. If you look at them, most wire spec sheet ranges vary way beyond what is practical to use. (Basically, if you can make it spark, you can weld with it. OK, a little exaggeration there but I think everyone knows what I mean.)
Further, we are using mostly semi automatic FCAW. The power supplies are constant voltage but amperage can vary a bit depending on many factors. The travel speed is very hard to calibrate (doggone welders won't stand still so I can put a clamp meter on 'em). I understand the need to verify heat input in many situations, but it isn't a big issue in what we do (not speaking for other shops here).

Having said all that, because we are a bridge shop, we have to "calibrate" welding machines every 3 months as a minimum. All I do is to set the parameters at mid-range for our most common D1.5 WPS, and use a calibrated volt-amp meter to compare readings. I mark the differences on a self stick label on the machine wire feeder. The wire feeder and power supply serial numbers are recorded in case anyone plays musical chairs with the equipment. The label tells the welder how far off the voltage readout is and how many amps the clamp meter reads at a specific wire feed speed. The only readouts provided on our equipment is for voltage and wire speed so this gets them in range. Our meter is calibrated yearly and traceable to NIST by the meter manufacturer.
Because we seldom use SMAW, all I do is to set the machine for about 100 amps DC+, burn a 1/8" 7018 rod, and read the "actual" amperage. I indicated "actual" because the meter dances around and I pick the middle of the extremes. Same type of sticker, just that I record what the machine was set at and the amps obtained. Per D1.5, prequalified SMAW doesn't need a PQR if you are within manufacturer's amperage ranges and they never list voltage or travel speed, just amperage.

I dislike "calibrating" welding machines becauses it seems like a lot of work for something that adds a dubious value to our welding program. But rules are rules. (There may be a better word to use than dubious but it's all I can think of at the moment) If anyone has a better way to go about this, I'd sure like to hear it. Anything will help.

I'm sure some people will disagree with me, but realize we only deal with A709 grades 36, 50, 50W and the equivalent structural steels. Not with the higher grade stuff.

Time to leave it, sorry for rambling.
Chet Guilford

It's possible to use a welding grid power supply (like you find in shipyards) as a giant resistor.....I don't know the details, but you can feed the welding machine output into it, at a steady state and measure the current/voltage of the power supply to be calibrated with a calibrated multimeter. You'll need a 50mA shunt too for catching the amperage.

It lets you avoid the dancing needle problem.