Why does SMPS trip the RCD?

[jdev99]

I have  a Qidi Q1 Pro 3D printer with a heated chamber, which actually worked very well, up till tonight.

I was starting a new print job tonight, and wanted to heat up the chamber, when the printer caused the RCD (earth leakage device) of my house to trip.
I connected the printer to an isolation transformer and notice that the voltage between Live and earth was 120V and between Neutral and earth was also 120V.
The mains supply voltage (in the UK) is 240V.

Is this "normal" to have 120V AC split between Live and Neutral to earth?
What can be a possible cause for the RCD to trip in this case?

I did contact the Qidi support for answers also.

The SMPS is a 350w, 24V model: CZL-350D-24. (Czchenglian)

[tooki]

I have  a Qidi Q1 Pro 3D printer with a heated chamber, which actually worked very well, up till tonight.

I was starting a new print job tonight, and wanted to heat up the chamber, when the printer caused the RCD (earth leakage device) of my house to trip.
I connected the printer to an isolation transformer and notice that the voltage between Live and earth was 120V and between Neutral and earth was also 120V.
The mains supply voltage (in the UK) is 240V.

Is this "normal" to have 120V AC split between Live and Neutral to earth?

When running on an isolation transformer, yes, that is not surprising.

Take a look at the schematic of a typical mains input line filter: https://www.we-online.com/en/components/applicationguide/appguide_mains_filter

See the two caps at the right, one from line to earth, the other from neutral to earth? With no earth present, as when using the isolation transformer, those form a capacitive voltage divider. So 1/2 of mains voltage is what you’d expect to find.


Unfortunately I don’t really have any suggestions on what might have caused the RCD trip. I had one randomly a few weeks ago in my kitchen, but no idea what from, and it hasn’t recurred…

[jdev99]

Tried to be clever by using the isolation transformer, to avoid tripping the RCD again, just to confuse myself with the 120V split. Should have known, thank you for the clarification. The minute I saw that schematic you linked it was 100% clear. 🤦

[coppercone2]

small power supplies have ghastly inrush currents sometimes. I have a 12A 12V switch on a supply that is 1.5A output maximum (~150W) and it literally caused sparks to fly out of the rocker switch on turn on, it like a 15 or 20A rocker.  I saw sparks shoot out the front through the seam. it makes me think the inrush is gigantic

I never saw behavior that bad before lol

[tooki]

small power supplies have ghastly inrush currents sometimes. I have a 12A 12V switch on a supply that is 1.5A output maximum (~150W) and it literally caused sparks to fly out of the rocker switch on turn on, it like a 15 or 20A rocker.  I saw sparks shoot out the front through the seam. it makes me think the inrush is gigantic

I never saw behavior that bad before lol

Yep, charging the input cap of a switching power supply can pull a lot of current, especially if by random chance contact is first made when the AC waveform is near its peak.

But what does that have to do with an RCD? That’s not an overcurrent device.

[SteveThackery]

But what does that have to do with an RCD? That’s not an overcurrent device.

Exactly. I wonder if the OP used the wrong term and meant the circuit breaker.  No current to ground = no RCD trip.

[coppercone2]

I think it get problem from EMI

[AVGresponding]

Could be a failing "Y-class" cap; a lot of cheaper gear has caps that aren't properly X- or Y-class rated, and may be failing leaky

[Siwastaja]

a lot of cheaper gear has caps that aren't properly X- or Y-class rated

A lot might be exaggeration; note it is an actual and serious criminal offense to purposely import and sell such devices which might end up in jailtime; in reality of course rarely that, just large product recall.

I have read about such cases and recalls but don't remember personally ever seeing an example. Even semi-shoddy supermarket chains selling random China crap try to do minimum amount of testing to ensure they don't kill their users, which a wrong type of safety-critical Y capacitor does.

If you self-imported mains-powered random China crap from Aliexpress/Ebay then you of course are responsible for killing yourself with it. But I wouldn't say "a lot of cheaper gear". I mean, it is similar to saying "a lot of budget cars have no brakes": the situation is not that bad IMHO.

[Brumby]

Could be a failing "Y-class" cap; a lot of cheaper gear has caps that aren't properly X- or Y-class rated, and may be failing leaky

^ ^ ^ This was my thinking. ^ ^ ^

Since these typically run in pairs (phase to earth and neutral to earth) if one holds good and the other fails leaky ... there's your imbalance.

[AVGresponding]

a lot of cheaper gear has caps that aren't properly X- or Y-class rated

A lot might be exaggeration; note it is an actual and serious criminal offense to purposely import and sell such devices which might end up in jailtime; in reality of course rarely that, just large product recall.

I have read about such cases and recalls but don't remember personally ever seeing an example. Even semi-shoddy supermarket chains selling random China crap try to do minimum amount of testing to ensure they don't kill their users, which a wrong type of safety-critical Y capacitor does.

If you self-imported mains-powered random China crap from Aliexpress/Ebay then you of course are responsible for killing yourself with it. But I wouldn't say "a lot of cheaper gear". I mean, it is similar to saying "a lot of budget cars have no brakes": the situation is not that bad IMHO.

If you don't have any to explore yourself, just watch a few DiodeGoneWild or Big Clive vids. Yes, it's cheap crap from ebay or AliX, and not from a bricks and mortar store, but there's plenty of it out there

[SteveThackery]

Since these typically run in pairs (phase to earth and neutral to earth) if one holds good and the other fails leaky ... there's your imbalance.

Do we know if the OP's power supply was earthed?  It doesn't seem clear from his posts.

[jdev99]

Yes, PSU is earthed and to answer a previous comment, no mistake , it was the RCD (80A, 30mA) that tripped.

Since it is a very new 3D printer, I doubt that it is failing capacitors.

It only happened that one time, that is the strange/interesting thing.

[tom66]

It's possible some residual moisture in the power supply caused momentary leakage.  I've encountered that when equipment in my cold garage is brought into the warm house as condensation forms; I've tripped an RCD before, but after drying out the kit, it worked fine.  You only need 30mA to trip an RCD, or about 10kohm leakage. 

[AVGresponding]

Worth pointing out also that 30mA is the current limit by which the RCD must trip (for a nominally 30mA unit, there are other ratings); typically they trip quite a bit lower than that, here's the IET general specification: https://electrical.theiet.org/wiring-matters/years/2019/77-september-2019/which-rcd-type/

EDIT: Edited for clarity

[themadhippy]

Worth pointing out also that 30mA is the current limit by which the RCD must trip (for a nominally 30mA unit, there are other ratings); typically they trip quite a bit lower than that

and once you add in the other items that maybe leaking the odd ma or 2 it don't take much to get get into trip current terrotery