Can anyone suggest a suitable dummy load for this purpose? Needs to be able to handle up to 200A for at least half a minute. I suppose the ideal thing would be an arc light bulb out of a cinema projector, but I'd prefer something more readily obtainable aready lying around the house/garage. I want to test current draw under load and see if there's any intermittent breakdown in output (which I suspect there may be with these newer inverter welding sets).
Are you allowed to run it into a short circuit? That way the only heat dissipation is in the equipment.
Obviously short circuits occur during MIG welding every time the wire hits the grounded metal. At that point, the wire is sputtered, the circuit becomes momentarily open until the wire feed makes contact again and this is continually going on at x cycles per second. As far as I know, anyway. The situation with ARC/MMA is fundamentally different as plasma is the dynamic resistance. I don't know the answer but discussing it does at least help to clarify my thinking and might give you or someone else here a better idea of the problem...
Most welders have an open circuit of 20V so at 200A we're dissipating
4kW and require a resistance of 0.1 ohms.That's some load.
Now if you could get a few of these:
You'd start with an article discussing the electrical characteristics. Which you already know some of them.
Here is an example of a load bank (this looks low-tech and probably for non-inverter welders). At least the scale of the box and the vent hole for the fan cooling on the side, hints at what sort of size is involved. Some of these load banks are intended for trailer-style welders. (Back home, the gentleman next door worked for the highways department and repaired their trucks, and he had a good sized trailer-mount welder.)
You could start with some wirewound resistors.
You can construct a heat tunnel, using four heatsinks with inward facing funs. Fasten a 120mm fan to the end of the tunnel, Drill and tap the heatsinks. Screw down the aluminium power resistors to the heatsink. Some thermal paste between heatsink and power resistor helps (Arctic Silver).
The hardest part of the project, is connecting the resistors together. The resistors are, after all, only 0.1 ohms and you will be connecting ten in parallel to hit 0.01 ohms. If the welder is 100A 100W, that load would work. If the welder was 200A 400W, that load would work (the resistors will get damned hot), One reason you are selecting a large number of resistors, is so they can be reconfigured to create other resistance values, according to welder characteristics.
To be a decent welder, the power numbers are probably different than that. Maybe the damn thing puts out a kilowatt. Make sure of your power numbers before incinerating something.
The resistors are available in more powerful values, but the price is proportional to the increase in power rating (and consequent reduction in delta_T heat rise above ambient). The 10W resistors are expensive and piddly (you might use them without heatsinks, still with a fan, but at the price, no way!). The 100W resistors are expensive. And so on. The seller sometimes tries to "move" components not selling well, or components where they have decent quantities. I got a real good deal on ceramic caps once, and regret not buying a lot of them while I had the chance.
I used to get power resistors for 1/8th the price of what I see today. The local electronics store used to sell them, and it made buying power resistors "painless". But it still doesn't solve the infrastructure issue. You have to make sure you're doing a good job of dumping the heat, into the air. The wire gauge to the ends of the resistors must be "extra fat". The electrical load I constructed has a fan too. But my electrical load is not in the same class as yours, so my wiring is a joke.
The one I show above, is the "fancy" version. Because it is encased in Al, and has screw holes, it begs to be bolted to a heatsink. There are cheaper ones that rely on their own casing for cooling, but then you need more of them. If you're not careful in your selections, the solder could even begin to melt. Some are also fuseable, and the material in the resistor will melt (it's part of the design). So read the datasheets carefully.
The above resistor is 200ppm or so, so not nearly as bad as the tempco of copper wire.
If the "resistor bank" is 0.01 ohms, the wire interconnect to the ends of the resistors must be very good quality, otherwise "the cabling is the resistor" and not the resistors. Copper has a high tempco, and if the copper resistance is greater than the resistor resistance, the tempco of the copper interconnect dominates the resistance behavior. (Resistance of load will change during operation, just like light bulbs change radically and nichrome has a tempco.)
Another thing to remember, is at currents above 50A, you start to feel magnetic deflection effects. This means that some of your materials need to be stiff enough to resist twisting or torque while powered. We got a demo of that at a local technical school during open house. They had some long cables with 100A running through them, and when the power was switched on and off, the cables (heavy cables) would deflect. Still not enough to be a danger, but still a visible effect. In engineering school, there were calcs highlighting the effect, but a visual demo makes a more lasting impression. These are not wires that were "coiled", so no multiplicative effects were present.
Paul
Perhaps a daft idea, but could you get a similarly specced welder which has been dumped because its control circuitry is faulty, but the transformer is ok, and use it in reverse? Then maybe measure the voltage across that transformer's primary when your welder's output is connected to its secondary. I assume you'd get around 230V, so what would happen if you put a 3kW heater across it to load it? Or maybe a 3kW + 1kW if Fredxx's mention of 200A @ 20V output is right? If the 230V (or whatever) didn't change much with load, wouldn't that suggest the 200A output spec is more-or-less as stated?
that's about 13m of 2.5mm so 6.5m of T&E with the phases shorted at one end? in a bucket of cold water if smoke is a worry ;)
Where's Andy Moir when you need him?
5-6 cheap electric showers, in effect water-cooled resistances?
Owain
Might need mains voltage for those.
ETP copper is usually taken as 1.7x10^-8
17 / 2.5mm = 7mohms/m so as you say ~13/14m. 4kW is an awful lot to conduct through PVC to water in a bucket. I think I might treat the cable to be a consumable in the experiment.
Interesting! I can't say I've noticed any cable twitching when welding using MMA at up to 320A, but then I wasn't looking at the cable at the time! I'm wondering if a trough of salt water would work? I seem to vaguely recall some serious radio hams in the US used them for RF dummy loads for their big linear amps.That might work if the electrodes were spaced close enough to emulate the resistance of a plasma 'ball' as it were. :-/
I should have made clear I looked upon the whole cable - copper as well as PVC - as sacrificial
and also that right length is tricky as resistance will increase with temperature so if crucial to maintain 200A for some seconds best shorter
- perhaps 10m
Is it AC then on the electrodes? I thought it might be chopped dc. Brian
Starter motor on a relatively seized up non starting car/lorry?
I think DC is more commonly used, but it depends on what the welder is being used for. Many welders seem to have switchable AC/DC output.
We need one of those takeapart videos.
For a look-see.
Paul
400mm of 2mm diameter stainless steel rod will give you 0.1 ohm. Stick it in a bucket of cold water and it should be good for a few minutes.
I'm glad someone did the sum, I was thinking much the same myself (but with steel: hadn't thought of using stainless).
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.