Big linear bench PSU trips RCD at turn on or off

[oldway]

By following your reasoning, it would be the bipolar switch that is defective and not the capacitors ....

You are complicating your life for nothing .... remove these two capacitors and it is solved ....

The thyristors cause noise anyway .... it is not these two capacitors that will make a significant difference.

If you are looking for a low-noise power supply, do not choose a power supply with a phase-controlled pre-regulator.

[Gyro]

Just a thought - Put an NTC thermistor in series with the live line? I know that this shouldn't be an inrush overcurrent issue, but it might just calm things down a bit.

Another thought is excessive transformer primary to core capacitance, given that this supply will probably date from the days before 30mA RCDs.

[Freelander]

Just a thought - Put an NTC thermistor in series with the live line? I know that this shouldn't be an inrush overcurrent issue, but it might just calm things down a bit.

Another thought is excessive transformer primary to core capacitance, given that this supply will probably date from the days before 30mA RCDs.

Not a good candidate for an NTC solution. 'On' *stabilised* resistance and high current don't mix well. Ok in lower power applications.

[floobydust]

Could OP unplug the unit's incoming power, short the input wires once (to discharge Y-caps) and simply measure capacitance/resistance between L, N, to GND and see if there is imbalance.

[Kleinstein]

I can imagine are three ways the inrush current could be a reason for tripping the RCD:
One is that there might be a more or less weak connection between N and PE somewhere in the circuit - especially if the RCD was added later / lately to an old installation. With high load current the N voltage reaches enough voltage drop to drive enough current. Under Light load this might not be enough to trip the RCD.

During turn on and maybe also during turn off, there might be a significant DC current spike (   ) from the SCR circuit part. Some RCDs don't like DC current that much and can trip from this, even if it flows the "correct" way. Inrush currents with large transformers can also reach really high values (e.g. 20 times nominal current) - this might trip the RCD by an unwanted side effect too.

Fast switching transients can also cause enough higher frequency noise on the line and this can give rise to higher than normal currents through the EMI suppression caps - the caps responsible may be even in other gear at the same circuit.

[Chris Wilson]

Just to close this, I replaced the caps with new ones of the correct rating. No change. I didn't want to remove any caps as I use it to power a 1kW TX on 136kHz, and there's a real risk of noise. So I got hold of a 100mA trip RCD to replace the 30mA one, and fitted it yesterday. Problem ceased immediately. Many thanks for all the help and discussion.