Issues with Dual Function Breakers on 120/208 commercial panels

[HertzGood85]

Hello, we just recently installed over 120 breakers in a 12-year-old long-term skilled nursing facility. The breakers are the Square D QOB 20-amp 120-volt DF (Dual Function) style and the panels feeding them are 120/208-volt three phase. When we initially installed them a few of the breakers would randomly trip intermittently. But the ones that did would trip with hardly any loads on the circuits. These breakers feed the power to residential rooms. There are no shared neutrals on any of the circuits, no lighting or non-linear loads and no refrigeration or heating appliances. Short circuits and ground faults have been ruled out and tested for. Over the course of 4 months, we now had to revert 30 circuits back to the original afci breakers because the nuisance tripping has become too much of a hassle and we cannot figure out the issue. The remaining 90 dual function breakers continue to work fine and have not tripped.

I have read some of the other posts on here addressing this same issue and I'm starting to wonder if the issue is either due to too high of inductance in load wire back to breaker or could it be a power quality issue? I'm no engineer and I could really use some help on this one.

Thank you.

 

[Jraef]

Square D specifically states that circuit lengths of over 250ft can cause nuisance tripping of their GFCIs.

What is the maximum total length of load wire that can be used with a QO-GFI or QO-EPD circuit breaker? | Schneider Electric USA

 

[HertzGood85]

Thank you for the reply, we did some reading on the square d instruction manual before install and did take note of the 250 foot limit on one way homeruns to breaker. So we did measurement on the building plans and ruled that out.
One thing we did notice when inspecting the home runs is that the original installers exclusively used stab in connectors in all the junction boxes to each room. Could this be a cause of issue?

 

[synchro]

In one of these other posts that you mention, I suggested that perhaps filtering and surge suppression using capacitors and/or MOVs to ground might cause a GFCI function to trip if there was a spike on the supply voltage. For example, plug-in power strips often have such functionality.
A suggestion I made was to place such a power strip on a circuit that has a GFCI breaker and is not tripping. And then see if tripping occurs in the same general time frame that the other tripping occurs. If so, I think that would give credence to a power quality issue being the root cause. It's conceivable that a large load switching on or off could be causing voltage spikes due to inductance that is present in the system. Or it could be from the power company side.

 

[HertzGood85]

That is an interesting suggestion, so you are saying to try and plug a power strip into an outlet that is powered by the same panel but not fed off a DF breaker, to see if maybe that circuit might trip around the same time period as the ones tripping? Does the power strip need to have a surge suppression feature?

As far as the large load turning off or on and inducing a heavy induction signal in neighboring wires, could a portable 120 volt xray machine induce such a load?

Also does anybody have any experience with using the ""https://www.fluke.com/en-us/product/electrical-testing/clamp-meters/fluke-368-fc"?

It is a leakage current meter and we've purchased one for the sole purpose of trying to figure out if there is too much leakage current on our problem circuits but I'm not registering anything higher than 1 mA on both ends of my homerun circuit.

Thank you

 

[synchro]

That is an interesting suggestion, so you are saying to try and plug a power strip into an outlet that is powered by the same panel but not fed off a DF breaker, to see if maybe that circuit might trip around the same time period as the ones tripping? Does the power strip need to have a surge suppression feature?

Yes, I'm suggesting to put a power strip on a circuit with a GFCI breaker but not a DF breaker.
The power strip would need to have some filtering and surge suppression, such as some from Tripp-Lite have.
You can tell if there is some filter capacitance to ground with your leakage current meter. Something on order of 0.5 mA might occur without any load attached.

 

[tortuga]

The breakers are the Square D QOB 20-amp 120-volt DF (Dual Function) style and the panels feeding them are 120/208-volt three phase.

So 1 pole breaker and a 2-wire circuit?

Also does anybody have any experience with using the ""https://www.fluke.com/en-us/product/electrical-testing/clamp-meters/fluke-368-fc"?

It is a leakage current meter and we've purchased one for the sole purpose of trying to figure out if there is too much leakage current on our problem circuits

$940 ouch. well nobody can say you bought a cheap meter.

I'm not registering anything higher than 1 mA on both ends of my homerun circuit.

What happens if you bring the ECG up to 2 or 3 ma ?

 

[HertzGood85]

Synchro: Okay I see what you are saying I will try to but the tripp lite on another breaker and see what happens. We normally have hospital rated power strips in all resident rooms but they are not of the surge variety but I will find one and try that.
You are right I measured an empty room that had no loads on the circuit and I measured leakage current of 0.7 mA at the room. Does that sound like a lot for no load?

tortuga: Yeah its a single pole breaker with a terminal for line and nuetral
Yeah it was pricey meter so I'm hoping I can track something down with it.
What do you mean when you say bring up the ECG to 2 to 3 mA?

 

[junkhound]

Can the residents have anything like microwave oven, etc in their room, or any device tht uses high frequency inverters like some of the newer microwave ovens that do NOT have a big heavy 2kV 60 Hz transformer.
During covid, I was building 3 kV, 2 kW UV lamp drivers operating at 120 kHz. One of the users was powering it with an extension cord that had GFCI. The GFCI would trip often. Even the few pF of low capacitance of MOSFETs to chassis was enought to trip that style GFCI.
Solution was to add a small inductor (10 uH) in the chassis lead to limit the 120kHz capacitive conduction, with negligible effect on 60 Hz groudn currents.
No idea if the dual function SqD cbs respond to high frequency, but probably do due to AF signature algorithsm, while std GFCI do not have high frequency HF response.

Might be worth checking to see if any residents have loads with HF inverter power.

 

[HertzGood85]

Can the residents have anything like microwave oven, etc in their room, or any device tht uses high frequency inverters like some of the newer microwave ovens that do NOT have a big heavy 2kV 60 Hz transformer.
During covid, I was building 3 kV, 2 kW UV lamp drivers operating at 120 kHz. One of the users was powering it with an extension cord that had GFCI. The GFCI would trip often. Even the few pF of low capacitance of MOSFETs to chassis was enought to trip that style GFCI.
Solution was to add a small inductor (10 uH) in the chassis lead to limit the 120kHz capacitive conduction, with negligible effect on 60 Hz groudn currents.
No idea if the dual function SqD cbs respond to high frequency, but probably do due to AF signature algorithsm, while std GFCI do not have high frequency HF response.

Might be worth checking to see if any residents have loads with HF inverter power.

Thanks for the reply, these rooms are not allowed to have any cooking appliances, they mostly only have tv's, radios, clocks, cell phone chargers and maybe the occasional electric razors. I like the way you're thinking though because I'm looking for just about any way to approach this issue at this point.

 

[tortuga]

What do you mean when you say bring up the ECG to 2 to 3 mA?

That should have been EGC (Equipment Grounding Conductor)
For example (a poor mans tester) I put a 30k 3watt resistor in a cord cap wired hot to equipment ground, that pulls .004 amps @ 120V, just below the level a GFCl should trip.
If the GFCI trips at that level I say its defective and return it.
Or put two in series for a .002 Amp load (2ma).
With that nice fluke tester and standard test conditions I dont know how the manufacturer could stand behind their product and say it works.

 

[synchro]

... Also does anybody have any experience with using the ""https://www.fluke.com/en-us/product/electrical-testing/clamp-meters/fluke-368-fc"?

It is a leakage current meter and we've purchased one for the sole purpose of trying to figure out if there is too much leakage current on our problem circuits but I'm not registering anything higher than 1 mA on both ends of my homerun circuit.

Have you tried turning the "filter" function ON and OFF on the leakage current meter to see if the reading is significantly different? The filter function when engaged will narrow the response to around 60 Hz and attenuate higher frequencies.

Also, were the 30 dual function breakers you replaced approximately evenly distributed among the three phases, or were they mostly on one phase?

 

[synchro]

Also, how do the leakage current measurements on problem circuits compare to those that do not trip? At least with a smaller sample of the 90 breakers that are not tripping.

 

[synchro]

Do the DF breakers have a LED for diagnosing what kind of fault is causing the tripping? And if so have you used it?

 

[HertzGood85]

That should have been EGC (Equipment Grounding Conductor)
For example (a poor mans tester) I put a 30k 3watt resistor in a cord cap wired hot to equipment ground, that pulls .004 amps @ 120V, just below the level a GFCl should trip.
If the GFCI trips at that level I say its defective and return it.
Or put two in series for a .002 Amp load (2ma).
With that nice fluke tester and standard test conditions I dont know how the manufacturer could stand behind their product and say it works.

Ahh I see what you mean now, that is definitely something we could try. I will say it has crossed my mind that maybe these breakers are bad but its hard to justify 25 percent of the 120 breakers being bad but I haven't ruled it out. Ive been trying to figure out a way to test them individually to see if they might be bad and your method seems like it may work. Through my years I have been looking for some type of product that tests low voltage breakers to see if they are bad but have never found one.

 

[HertzGood85]

Have you tried turning the "filter" function ON and OFF on the leakage current meter to see if the reading is significantly different? The filter function when engaged will narrow the response to around 60 Hz and attenuate higher frequencies.

Also, were the 30 dual function breakers you replaced approximately evenly distributed among the three phases, or were they mostly on one phase?

I had not figured out what turning on the filter on the meter meant until you just explained it, that's a good idea. What would you say a different reading would indicate through this higher frequency filter? On normal mode for the meter, I'm getting a reading of 0.7 mA leakage current on an empty (no load) room that is having issues and in a lived-in room with minimum loads I'm getting about 0.9 to 1 mA.

Also, I did do a check on that, and I would say 90 percent of the breakers I replaced were on the B and C phase in the panels. These panels are all in separate buildings and are fed off the load side of a step down 480 transformer but of course all power in the facility comes from main 2000-amp utility bus.

 

[HertzGood85]

Also, how do the leakage current measurements on problem circuits compare to those that do not trip? At least with a smaller sample of the 90 breakers that are not tripping.

Thats the thing, I originally hoped the meter would show me a discrepancy between the circuit of a room that has been tripping and one of another that has not but I'm getting the average reading in both rooms which is frustrating.

 

[HertzGood85]

Do the DF breakers have a LED for diagnosing what kind of fault is causing the tripping? And if so have you used it?

The dual function breakers do not have an indication light, but they do have some type of internal function and/or memory. There are steps in the instructions about how to hold down the test button after trip for a certain amount of time to see if the trip was due to a ground fault or an arc fault but when we do that test it comes back that neither has occurred. These tripping occurrences have really been an issue but I will say im not stranger to nuisance tripping because normally we deal with afci breakers in resident rooms installed facility wide and we have are fair share or issues with them. But we've never had this so many of them trip this often and if we did it would be one room where either a oxygen machine had to be replaced or the breaker was bad. But 30 bad dual function breakers? I can't fully believe that, but we did buy a bulk purchase order of 120 of them so I do not know.

 

[synchro]

... On normal mode for the meter, I'm getting a reading of 0.7 mA leakage current on an empty (no load) room that is having issues and in a lived-in room with minimum loads I'm getting about 0.9 to 1 mA.

Can you disconnect both the hot and neutral wires from the output of a breaker that is having issues on an empty room, and then see if that breaker will ever trip?

 

[HertzGood85]

Can you disconnect both the hot and neutral wires from the output of a breaker that is having issues on an empty room, and then see if that breaker will ever trip?

I have not tried that. If an unwired breaker doesn't trip that would indicate that breakers is bad correct?