enermax liberty tripping gfci.

[Mad_Professor]

My recently repaired ENERMAX Liberty ELT500AWT was working fine on normal unprotected circuits, but then I move the computer to my recently renovated shop (detached garage) Where code has required I use these gfci receptacles.

Anyways when I try to fire up my rebuilt workstation, it trips the gfci. I thought bad cord or something wrong. I switched out cord with a new cord, tripped again. I tried a different circuit with a matching GFCI recep, tripped that one too. But then I went outside which has a weather-resistant gfci which is a different brand, didn't trip. Yes I did push the test button to make sure it works.

I don't think there's anything wrong with the power supply, I've google it and it seems like it's a common problem for enermax PSU because of it's active PFC.

Solutions?

EDIT: also I am in the market for an UPS Battery Backup, so that could be a solution but I don't know what to buy that could stop this gfci tripping.

[domas]

Re: enermax liberty tripping gfci.

did you check Y and X caps?

[Mad_Professor]

Re: enermax liberty tripping gfci.

I'll be honest, when I say I don't know what you mean.

I'll be an ass and assume that it's those blue thingyies near plug that bridge the neutral and ground.

Also how do test them?

[Mad_Professor]

Re: enermax liberty tripping gfci.

Well I somewhat solved it.
I replaced the gfci receptacle which was a leviton brand with a pass & seymour gfci, ($16 at Lowes). It works, just as long as I don't frequently restart/reset the computer, like overclocking which requires a lot of the "reset button" then it will trip. Hopefully when I get a battery backup it will somewhat help with the problem. But for right now It allows me get the computer up and running from a cold start.

[Behemot]

Re: enermax liberty tripping gfci.

There is always some loss in SMPS as part of the power is radiated away from the circuitry and catched by the grounded cover. I've read something about this, for that reason there should be these current protectors with built-in tolerance for several tens of mA (depends on exact type) or with slower reaction time intended for SMPS.

Or just don't use it at all, this usually is usefull in bathroom or so for situations like using hair-dryer in bath.

[kaboom]

Re: enermax liberty tripping gfci.

Failed/insufficient line filter allowing RFI/EMI to backfeed into GFCI. "Donut" current transformer in GFCI picks up these currents, unbalanced to a point. This unbalance looks just like a ground fault to the opamp/comparator, and the device trips.

Such RFI/EMI varies with load, and as you've found, "rapid resetting" trips the GFCI.

There is always some loss in SMPS as part of the power is radiated away from the circuitry and catched by the grounded cover. I've read something about this, for that reason there should be these current protectors with built-in tolerance for several tens of mA (depends on exact type) or with slower reaction time intended for SMPS.

Partially correct. A 4700 pF line-to-ground cap (of the two) has an XC of about 560k ohms at 60 Hz. This allows about 214 microamps to flow. This is the so-called leakage current when dealing with EMI filters. This current occurs because of line voltage and capacitive reactance of the device.

The noise current is not trying to "get to ground." It tries, with all parallel paths, to go back to its source, namely the common point of the DC bus.

This can be the "film cap" after the rectifier in a PFC booster, the electro(s) in a flyback or forward converter, or the center tap of two electros in the case of a half bridge.

That return path is not perfect nor can such noise be easily controlled. Hence common mode chokes, forcing these currents back to the supply common by presenting a higher common mode impedance up the line cord than back to the common return.

But in order to present a high impedance on the node which faces the power supply, the node facing the AC line must have a very low impedance (at high frequency). Hence the X cap across the "line side" of any common mode choke(s).

If the X cap opens or loses capacity, the filtering action is greatly reduced.

-There must be both X caps and common mode chokes.

[Behemot]

Re: enermax liberty tripping gfci.

There always are some losses, you cannot supress them to 0. Than it just depends on how sensitive GFCI is. And as I said, there should be some especially for SMPS, designed to tolerate some losses.

[Nitro468]

Re: enermax liberty tripping gfci.

GFCI receptacles have about a 10% failure rate... OUT of the box... Test them monthly..

Any issues, change the outlet 1st..