I attached a qualified weld procedure to a new job. It was within the essential variables of the code, AWS D1.3. The part is a washer like cap welded (GTAW) to the end of a 0.750" diameter tube. The welder welds a weld about 3/4", flips the part, welds another weld about 3/4", flips the part, and repeats until the weld is complete. The weld time isn't very long. It's a series of starts and stops.
An auditor asked me why the welder wasn't welding the part per the WPS's amps and volts? I didn't have a good response. I took her suggestion and agreed to create or modify the WPS to accommodate the electrical differences.
A few days later, I observed the welder weld some parts, and modified the amps and volts on my WPS. The welds are welded within the ranges, except for the starts and the stops. Also, the first weld is cooler than the third. The displayed amps and volts on the welder don't have a chance to zero out because the time between welds is so short. The welder is less than five years old, a Miller Syncrowave, that is calibrated annually. Basically, the amps and volts are bouncing all over the place. It's really a visual guess.
I'm wondering what will I do if the auditor receives my corrective action, decides to see for herself, schedules a follow up visit, and the amps and volts don't match.
What is a more scientific way to measure the amps and volts for a WPS?
Is it wrong to record the amps and volts on my PQR's, and WPS's by averaging the readings?
I stand by the welder, while he's welding, and record the amps and volts, over and over, during the weld (for a PQR). I'll typically have four or five readings. Then I'll average them, and enter on my PQR. Is this wrong?
Any help would be greatly appreciated.
Keith
First of all, under D1.3 volts are NOT an essential variable for SMAW or GTAW. And for good reason. Starts and stops will do exactly as you have witnessed. Also, even if it were the variable states "mean" arc voltage.
As for amps, since the machine is CC you should be able to read somewhat stable numbers and still also apply the "mean idea".
I have a great Fluke clamp on type of meter and it has a setting that gives me the average.
Would you place the clamp meter near the torch or near the welder, or would it matter?
Thanks for the help.
Keith
amps will the the same throughout the circuit
I always check as close to the welder electrode/ torch as possible to account for any loss in the cables.
Imagine a gut with 20' of lead and another guy with 500' of lead. The longer lead may loose more and thus have to turn the machine up to account for it. The dial setting may appear high, but whats coming out the end should be within your range.
If you wanted to get fancy you could use an oscilloscope to determine the maximum and minimum welding parameters while the welder is welding and list the full range on the PQR. It may be over-kill, but if you are forced to contend with an auditor that lack real-world experience you may have little recourse. As you observed, the welding parameters fluctuate while welding because the welding arc is dynamic. The actual conditions are constantly changing, i.e., the arc length changes due to the human element as does the resistance across the arc, thus so do the voltage and amperage. Since the arc "on time" is brief, the arc never really becomes stable.
You will have to be careful of how you connect the voltmeter to the machine. If you do not hook it up before the high frequency oscillator, you will toast the meter.
Al
As a note, if this is an inverter based machine, you may see the switching noise on the output if you use a scope that's fast enough (most are). In short, you won't easily be able to get a "number" for voltage unless your scope displays RMS voltage readings on the screen. A volt meter that does RMS is better for this, but if your procedure is pulsed, you will run into issues getting good voltage readings due to the pulsing. I don't have a good solution for this aside from spending money on a DAQ, but these are some of the possible road blocks if you go down this road on an inverter based machine.
Forum
