Here's an odd one. I have a GFCI-protected circuit with a Uninterruptible Power Supply on it. I've found that when I test the outlets on the circuit with a GFCI outlet tester the tester does not trip the GFCI. However, when I unplug the UPS from the circuit the tester trips the GFCI as expected. Why would this be? The UPS is a Minuteman Entrust 500VA in case it matters.

Note: I'm am not plugging the tester into the UPS, I am plugging it into either of the 2 "regular outlets" or the actual GFCI outlet.

enter image description here

  • 1
    \$\begingroup\$ This is the tester, a standard household electrical tester, you press the button and it should trip the GFCI outlet on the circuit amazon.com/Klein-Tools-RT210-Receptacle-Tester/dp/B01AKX8L0M \$\endgroup\$ – Todd1561 Mar 26 '18 at 22:55
  • 1
    \$\begingroup\$ I believe the way most of these testers work is to create a small current leakage to ground, but not so much as to create a full short and thus trip the circuit breaker \$\endgroup\$ – Todd1561 Mar 26 '18 at 23:44
  • 1
    \$\begingroup\$ I should also add that while the academic discussion of why this is happening is of interest I'm also interested to know if I'm effectively losing GFCI protection when I use this UPS, or if these simple testers are not adequately testing real-world situations. It's worth noting that when I press the test button on the actual GFCI outlet feeding this circuit it trips as normal, UPS or not. But that wasn't good enough for the city building inspector that's approving my finished basement, which is how all this started. \$\endgroup\$ – Todd1561 Mar 26 '18 at 23:48
  • 1
    \$\begingroup\$ It does, please read my comment above. \$\endgroup\$ – Todd1561 Mar 27 '18 at 0:04
  • 1
    \$\begingroup\$ Not an answer, because I'm not sure it's the case, but make sure that you don't have neutral and hot reversed either at the GFCI or between the GFCI and UPS. \$\endgroup\$ – Bob Jacobsen Mar 27 '18 at 5:28

If the UPS is drawing significantly high frequency or noisy current the GFCI might refuse to trip.

This is why they sell various classes of GFCI.

types of GFCI

Source: ABB Protection against earth faults with Residual Current Devices (Technical guide | 2013)

The wrong type might refuse to trip because the coil is saturated with DC current.

  • \$\begingroup\$ Interesting, this may be the most plausible explanation so far. I have a query out to the manufacturer of the UPS, I'll report back with any info they provide. But I'm sure it'll be along the lines of "we don't support our UPS on GFCI circuits". Do you think I'm effectively disabling GFCI protection by using this UPS? The test button on the outlet does work, however. \$\endgroup\$ – Todd1561 Mar 27 '18 at 12:13
  • \$\begingroup\$ @Todd1561 yes, try the test button on the GFCI. Should be similar to your cheap tester. Or rent an installation tester. It's also strange to purposely use a GFCI on critical systems, but you have a small home/office ups. Which could mean it's not that critical. \$\endgroup\$ – Jeroen3 Mar 27 '18 at 13:12
  • \$\begingroup\$ Yes, using the test button on the outlet trips the circuit with or without the UPS in play. So there must be something different with how these plug in testers work. And no, nothing critical about this equipment, just some home networking stuff in my house. I head back from Minuteman, that was worthless. Verbatim response: "UPS's have leakage current. GFCI are for leakage current. The UPS's pass UL." I don't think he understood what I was talking about and assumed the UPS was causing nuisance trips. \$\endgroup\$ – Todd1561 Mar 27 '18 at 13:23
  • \$\begingroup\$ Without really tearing into the minute details of the components involved I think Jeroen3 has the most plausible explanation so I'm going to mark this as the answer. \$\endgroup\$ – Todd1561 Mar 27 '18 at 13:25

take 2.

Class C, D, and E GFCIs trip at 20 mA rather than the 6 mA trip current mandated for Class A GFCIs

Assuming this is a 6mA GFCI then this can be simulated on a 120V line by drawing >6mA between Line and Earth or > 720mW or 20 Kohm.

The UPS is supposedly Simulated Sine Wave The GFCI ought to its own self test button which needs no external tester. This is the test hat should work

If the tester and GFCI tolerances overlap such that the GFCI is less sensitive than the tester expects, and the UPS has a slightly lower voltage it possible to not induce enough leakage current.

But this is a guess since we have no measurements for V and leakage currents or tolerances fo each device.

  • \$\begingroup\$ To clarify, my problem is not that the GFCI is tripping, it's that it is NOT tripping. When the UPS is plugged in the tester shows that the circuit has no GFCI protection and does not trip. When I unplug the UPS I can test any outlet in the circuit and they all trip as normal. Although, I'm not an EE, so I may be misunderstanding what you're saying. I have found many references to UPS's tripping GFCIs when they shouldn't, ironically I don't have that problem. \$\endgroup\$ – Todd1561 Mar 26 '18 at 23:01
  • \$\begingroup\$ I read too quick and assumed that common issue. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 26 '18 at 23:38
  • \$\begingroup\$ I still think we're not on the same page. I'm not plugging the GFCI tester into the UPS. I'm plugging the tester into other outlets on the load side of the GFCI outlet. When the UPS is plugged into an outlet on the GFCI load side it seemingly disables the GFCI protection. See the drawing I added to my original post, the tester is plugged into either of the "regular" outlets or the GFCI outlet, not the actual UPS \$\endgroup\$ – Todd1561 Mar 27 '18 at 0:19
  • \$\begingroup\$ It's almost as if the UPS is "abosorping" the current leakage to the point where the GFCI outlet is not detecting it, if that makes any sense. \$\endgroup\$ – Todd1561 Mar 27 '18 at 0:26
  • \$\begingroup\$ No. The GCFI is supposed to trip using a relay which acts upon the difference current between Line and Neutral and the tester is supposed to inject that load to ground to create the difference current. Even if the outlet current is zero it uses a Common mode transformer but with reverse winding to make it a common mode difference , which wont make any sense to you. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 27 '18 at 1:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.