Poorly designed automotive spot welder that suffer inrush current.

[mOSkit]

Good Morning all

I bought a spot welder / dent puller. Of course for the price it has to be made in china.
The stuff suffer from inrush current, and almost every time I try to do something it pop the brakers. I guess this is because the transformer have to be saturated on each trigger rather than staying charged?
I already went with the route of C type 32amp with very little improvement. (maybe 1 weld every 10 attempts)

I do beleive that the verisotor (20D471k) supposed to act as inrush limiter as its just before the transformer.
the SCRs are CLA50E1200HB.

Can some one suggest how can i modify this board to better manage the inrush? Unfortunetly i don`t have schematic,

Thanks for any help.

[wraper]

Varistors do not limit inrush current.

[mOSkit]

Appologies, i thought i`ve seen an article where the varistor is used as ICL.

Any idea what to do? Maybe I should use NTC Thermistor inline?

Thanks

[moffy]

A simple inrush limiter is a series resistor with the mains input, suitable power and value, that after a short period is shorted out with a relay. The resistor limits the inrush current  and is then bypassed by the relay, but this only works for switch on type inrush, if you are tripping the breaker after each weld then you have more serious issues. Some Circuit Breakers have integral earth leakage sensors, if that is the case then you need to find out if it is tripping from over current or earth leakage current.

[moffy]

I cannot see any large capacitors that would cause a large inrush, and the transformer looks small for welding, especially with the small diode bridge after it, it looks like the power supply for the logic. It couldn't be switching the non isolated mains through the SCR as the welding pulse could it? or is there more circuitry that isn't shown?

[mOSkit]

Sorry i could not upload more than 4mb of photos.
This is the whole unit. the PCB is the only thing that time trigger the transformer.

[Kleinstein]

Ideally they would have a design to start the transformer on the correct have wave, depending on the time it was turned off. This would however be a tricky part to add later - it would be the correct way to design from the start. Even though made in China there is a chance it is designed this way and just failing, e.g. with a braken part somewhere - still hard to tell and fix.

[moffy]

Wow, that is a heavy winding on the output of the welding transformer. From what little I can see, and I could be wrong, the SCR fires the mains across the welding transformer primary which is connected to the electrodes by the high current secondary, which is effectively short circuited at the weld point. The only thing limiting current is leakage inductance and series resistance of the windings and electrodes. If correct, the device is dangerous and not readily modifiable, but I am making a number of assumptions based off a photo and could well be wrong.

[mOSkit]

That`s what i thought it happening. The SCR will only fire the output on 0-180 wave.
Not so sure what to do with this thing, but yeah.

the PCB kind of work ok on when i disconnect the transformer, but i cannot determine anything more.
the primaries measure 0.5ohm and this kind of match other type of welding devices.

[Kleinstein]

A ring core transformer is not really suitable for working with 1 half wave only. One should still have a 2nd SCR to also use the other half wave.

By nature there will be quite some pulsed current. So a slower blowing fuse is likely needed to operate such a device.

[mOSkit]

You can see that there are 2 SCRs on the board.
I used C type MCB. The standard domestic is B type, the C type supposed to take care of the inrush because it 5-10 times the rating.
That improved but not as much.

[wraper]

Does it actually sense mains voltage to start energizing transformer at peak voltage to minimize inrush? I see no optocoupler, so the only thing could be sensing smaller transformer output.

[Circlotron]

Maybe when you hit the weld button the transformer primary is being energised at the zero crossing of the AC input. This is the worst situation for transformer inrush current. It will saturate and pull LOTS of current for a number of cycles. Best option it to power the transformer on at the peak of the AC voltage.

[mOSkit]

I think this is the case now. i replaced the breaker with 40 amp D curve.
It weld, but now the household plug 13 amp fuse gets hot after 6 shots and pops. What a junk

[moffy]

I think this is the case now. i replaced the breaker with 40 amp D curve.
It weld, but now the household plug 13 amp fuse gets hot after 6 shots and pops. What a junk

It is more than junk it is positively dangerous, a fire in the house/shop wiring is a real possibility.