Thread: RCD tripping on long welding bead - cause?

Looking for some advice on what to check next....

welder is an older (20 yo) CIG buzz box transformer style arc welder. max current of 130A, and stepless adjustment down to 50A.

in the new shed (new RCDs) the RCD/MCBO trips whenever I try laying a long bead of weld - but the RCD is quite happy if I pause halfway through a rod.
rods are 2.5mm 7018s, running at around 90A.
If I dial the current down, I can run a full rod without tripping, but there are times when I want the higher current. It definitely is linked to time at full or high current, but it is tripping the RCD, not the MCB

I cleaned out the welder, blow the dust and rap out of the windings, and did a megger test between the windings and case - all good.
all that is inside the welder is the transformer, a thermal cutout, and a cooling fan. Each part tested fine for the megger and conductivity. The windings are enamelled copper.


thoughts on the cause of my observed problems?
anyone else experienced this?

Thanks,
Des

Are you using a extension lead, if so it probably time to replace it could also be the lead on the welder breaking down

I'm not an electrician, but does the acronym MCBO mean it's an overload breaker as well as an RCD? If so then is the welder tripping it out simply on over current? The time related symptoms you describe seem to suggest this. A while back something (can't recall what it was, but it could have been a welder) was tripping a breaker on one of the circuits in my workshop. I replaced the breaker with a new one, same rating, and the problem went away.

Graham.

At mens shed we had a similar welder that did the same thing.
We speculated that as these transformer welders age, repeated heating and cooling leading to the continual slight expansion and contraction might lead to a problem .
It works fine when it's cool but after running for a while and warming up can push the transformer coils around ( combined with possible less than optimal insulation in the odd hotspot) so's they touch something they shouldn't.

What we did was set up the electric testing gear so that it is right on hand and then run the welder until the RCD/MCB trips.
Disconnected from mains and IMMEDIATELY tested the circuit - don't give it a chance to cool down.
Sure enough the transformer failed a Meger test. Then when it cooled down it was fine again.

I am thinking along the lines of the thermal cutout tripping at the latter end of the rod, and the transformer coil creating a substantial inductive voltage spike that is shunted back through the mains cables to the switchboard. Any device operated from the switchboard with a mains filter or surge suppressor will then be trying to damp the spike by clamping it or shunting it to ground, thus tripping the RCB. Once the inductive spike has dissipated, there is nothing left for the average user's meters to find.

The only suggestions I can make to test this are:

1. Measure the transformer primary resistance with a multimeter with the unit unplugged (Active pin to Neutral pin on the plug) before use, and keep the meter handy. Use the welder in a manner that creates the problem, and when it occurs, immediately unplug from power and repeat the measurement. If the thermal cutout is in the primary side of the transformer, and there is an obvious open circuit second time around, it is a reasonable guess that the thermal is tripped and associated with the issue. If the thermal is on the secondary side of transformer, this test will not show anything as the primary circuit remains intact.

2. Does the welder last longer for the first stick of a session than for subsequent sticks? Here I am thinking that the transformer and thermal are close to ambient temperature after a decent break, and would take longer to reach a temp that trips the thermal than subsequent sticks would when the transformer and thermal would be close to the thermals auto reset temperature (probably about 20deg below the trip point). For example, 20 deg ambient, thermal trips at 70 deg and resets at 50 deg, starting from ambient means the system needs to rise temperature 50 degrees to trip thermal the first time, but subsequently only 20 deg (difference between trip and reset temps) for subsequent sticks.

3. Record mains voltage with some form of logger or storage CRO capable of handling large inductive spikes to see if that can capture evidence of a spike in the mains cable. Requires specialist skills and devices to do this safely, not in the realm of the typical hobbyist or factory worker, substantial potential for risk to personnel or equipment if done incorrectly.

Originally Posted by desbromilow

Looking for some advice on what to check next....

welder is an older (20 yo) CIG buzz box transformer style arc welder. max current of 130A, and stepless adjustment down to 50A.

in the new shed (new RCDs) the RCD/MCBO trips whenever I try laying a long bead of weld - but the RCD is quite happy if I pause halfway through a rod.
rods are 2.5mm 7018s, running at around 90A.
If I dial the current down, I can run a full rod without tripping, but there are times when I want the higher current. It definitely is linked to time at full or high current, but it is tripping the RCD, not the MCB

I cleaned out the welder, blow the dust and rap out of the windings, and did a megger test between the windings and case - all good.
all that is inside the welder is the transformer, a thermal cutout, and a cooling fan. Each part tested fine for the megger and conductivity. The windings are enamelled copper.


thoughts on the cause of my observed problems?
anyone else experienced this?

Thanks,
Des

New shed, new breakers? Give your sparky a call.

I would hazard a guess the the breaker (RCBO) is possible faulty or underrated for the task. Just for an example, if you are pulling a solid 15A from a 15A breaker, there there is a set time before they trip. Likewise, you can pull 20A though a 15 amp breaker and it might take a few minutes to trip. The exact times are specified, and I would have to look them up to confirm the exact numbers, and even then there is some tolerance.

I would first try fitting a new RCBO, if that fails then look at upgrading the circuit. Any sparky worth his salt would have run decent wiring if he knew the customer was using it to run welders or machinery, so hopefully it might be as simple as upgrading the breaker and the GPO.

I'd go along with BobL

It could be working perfectly normally but you are taking too much current for the breaker. It won't trip immediately. I brought a 130amp stick welder from UK with me. it has a 13 amp fuse in the UK plug and I just used a short uk extension lead with a 10 amp domestic plug on its input. That fuse never blew in UK, not ever. In NZ it went all the time and in the end i had to replace it with a nail because i used every UK 13 amp fuse I had. The 10A breaker in the distribution board never went, so make of that what you will. My point is those stick welders take a lot of current. mine was designed to run on UK mains which is probably 10v higher than NZ so that probably explains why the fuse kept blowing in the plug.

I am betting BobL is right.