Kohler 12RES Generator tripping GFCI outlets - solved at last!

[Pacifico]

(See update on page 3 dated 11-3-23)

Friends,

I recently added a Kohler 12RES propane generator to my off-grid system but am experiencing strange behavior when the generator runs and charges the batteries. Some, but not all of my GFCI protected house electrical outlets eventually trip, one by one, after about 10-20 minutes. It's always the same outlets that eventually trip and others never trip at all.

Some background:

My house was built 4 years ago according to modern codes by competent tradesman. It was built as an off-grid home. An experienced electrician and his apprentice did the house wiring and a solar contractor experienced in off-grid systems assembled and connected the off-grid power system.

When our off-grid system was installed, we used a Honda EU6500i as our backup generator since we already owned it and wanted to spread out construction costs over time wherever possible. For that reason, we only got around to having the Kohler 12RES installed a couple months ago. When we were charging batteries then and now with the Honda, we never experience this problem. Only when we use the newly installed Kohler generator does this happen so that's why I'm zeroing in as the Kohler being the source of this trouble.

The Kohler is set up to charge at maximum input for my battery bank but even if we dial back the amount of power, the specific GFCI protected circuits in the house still begin to trip one by one. Sometimes, one or more of the GFCI's will accept a reset while the generator is running but mostly, they stay tripped and won't reset until I turn the generator off. My solar contractor (who did the Kohler install) is at a loss to what's causing this behavior even after consulting with tech support.

I tried replacing one of the GFCI outlets with a new one just in case I had an unusually sensitive GFCI and it tripped just as fast as the original one. Again, only some GFCI outlets in the house trip. Others never trip at all.

Perhaps, one of you have seen this problem before and may have a direction to point me to in solving this mystery. If I can clarify anything, please ask.

Thanks in advance.

J

[jbakker]

Check to see if the Generator Neutral is bonded to the Generator frame.
I am assuming that the Generator is Grounded.
Only supposed to be one spot in the system where the Ground and Neutral are connected and that is the main panel in your house.
This may be the issue. There will be smarter people I am sure to help out soon as well.

[raysun]

My first reaction would be to believe the generator is creating a ground fault.

Is neutral tied to ground at the generator?

If the ground between the generator and the inverter is removed temporarily, does the same tripping occur?

[jbakker]

RAYSUN we must have been typing a reply at the same time.

I suspect the generator neutral is bonded to the generator and the generator grounded to the main panel.

Let's hope he finds a solution

[fcwlp]

On the 12-RES the N-G comes bonded from the factory and is located in the junction box on the generator.

[raysun]

On the 12-RES the N-G comes bonded from the factory and is located in the junction box on the generator.

So the solution is to remove the N-G bond (jumper) in the junction box. Hot L1, L2; N; and G; all connected to the GSLC (or whatever AC wiring panel is used) as per the normal scheme.

[Pacifico]

Thanks everyone for your responses. I'll run your comments by my solar guy and see if some of these ideas aren't ones we've already considered or could try next. He'll be back soon to continue searching out solutions to this conundrum and I'll report back once this happens.

[Pacifico]

September 21st update:

I am cautiously optimistic that the mystery of the tripping GFCI outlets when on generator power has been solved. It seems that the Leveton brand of GFCI's were overly sensitive and would all eventually trip within about 10 minutes of the Kohler generator coming on. all of the Leveton brand GFCI's were removed and replaced with a different brand made by Legrand. After running the generator for 40 minutes today, none of the Legrand GFCI's tripped!

Since my first post on this issue back in April, Kohler had been kind enough to send out an experienced Kohler generator repairman as well as a licensed electrician to look over my entire off-grid system since no one could make sense of why this was happening. Together with the solar contractor that installed my system and many calls to tech-support, they all spent quite a bit of time trouble shooting and testing. It was the electrician who suggested switching to a different brand GFCI after exhausting all other possible issues. He had a different brand GFCI outlet (made by Legrand) in his truck, popped it in and it seemed to work without eventually tripping like all the ones made by Leviton would do when the generator was running.

At that point, it was decided that we should try replacing all the Levtion GFCI's with Legrand GFCI's and see what happens. After another week, my order of a dozen Legrand GFCI outlets arrived and I replaced all the original Levitons with Legrand and anxiously awaited results after switching on the generator. After 40 minutes of the generator running, I didn't get any tripped GFCI's. I'll run the generator again in the morning and check all the GFCI's but I'm at this point, cautiously optimistic that this strange and frustrating conundrum has been solved.

I still don't know why none of this ever happened in the four years I was running a Honda EU6500i generator. It was only when I upgraded to the Kohler 12RES that the issue of the tripping GRCI's started. Some of the other electrical outlets in my house are served by Leviton brand combination GFCI/AFCI outlets and none of those ever tripped, only the GFCI's.

[fcwlp]

He had a different brand GFCI outlet (made by Lebrand) in his truck,

Thanks for the update. For those of you looking for Lebrand GFCIs it is Legrand.

[Pacifico]

Thanks for the update. For those of you looking for Lebrand GFCIs it is Legrand.

Thanks, I corrected my post. Don't know how I missed the spelling, the box they came in is sitting right in front of me!

[raysun]

still don't know why none of this ever happened in the four years I was running a Honda EU6500i generator. It was only when I upgraded to the Kohler 12RES that the issue of the tripping GRCI's started.

Im not sure it is causal, but the EU6500i is an inverter generator with very clean AC, while the Kohler is an alternator generator with (a bit less) clean AC.

[Pacifico]

Well, it's back to square one. The Legrand GFCI outlets that I installed in place of the original Leviton ones also trip intermittently when the Kohler generator runs. It's an enigma. Kohler sent a top technician to look at my generator and he's say's the generator is functioning normally. He and the licensed electrician that came with him spent the day trouble shooting the generator and to some extent, the rest of my electrical system and could not point to a source for the intermittent tripping of some (but not all) of my GFCI protected wall outlets.

[pss]

https://www.engineersedge.com/instrumen ... tioner.htm

Power conditioners are often overlooked and may provide a solution for you. In places like hospitals where equipment is very sensitive, they make certain that ground is zero volts and hot voltages are constant. Many conditioned outlets are identified by a small green dot on the outlet.

So while I don't know if this direction can help you, it may be a place to explore and maybe identify some solutions for you. Maybe even something as simply as coiling up your wiring.

[Keithvcollins]

I have a Kohler 20RCA in new home with lots of automation. On generator power I experienced nuisance tripping of random GFCI breakers and outlets, trouble with a shark vacuum cleaner not operating correctly, and the elevator causing the generator to hunt. The problem presented itself under load loads where the %THD was higher. We tried isolating circuits/loads but could never find a smoking gun. I also replace all the GFI outlets with a Legrand. After 8 months of Kohler pointing to the non-linear loads in the house the installer wired in two 20uF run capacitors in parallel from each leg to common. It appears that this successfully filtered out the harmonics that are causing the poor behavior of the Kohler voltage regulator and now everything works correctly. Sharing and hoping this helps the next person chasing nuisance tripping with a Kohler generator.

[Jesse Lankhaar]

Please contact Technical Support M-F 6am-5pm PST at 360 618 4363 for the best troubleshooting and support.

[Pacifico]

I have a Kohler 20RCA in new home with lots of automation. On generator power I experienced nuisance tripping of random GFCI breakers and outlets, trouble with a shark vacuum cleaner not operating correctly, and the elevator causing the generator to hunt. The problem presented itself under load loads where the %THD was higher. We tried isolating circuits/loads but could never find a smoking gun. I also replace all the GFI outlets with a Legrand. After 8 months of Kohler pointing to the non-linear loads in the house the installer wired in two 20uF run capacitors in parallel from each leg to common. It appears that this successfully filtered out the harmonics that are causing the poor behavior of the Kohler voltage regulator and now everything works correctly. Sharing and hoping this helps the next person chasing nuisance tripping with a Kohler generator.

Thanks for that tip. I'll run this by my installer and get his thoughts as I'm still plagued by the GFCI's tripping when the generator is charging the battery bank. Neither Kohler (who sent their own expert to our site to troubleshoot) nor Outback Tech Support have been able to offer us any ideas to resolve this issue.

[clivusmul]

New forum member here... Very interested in this topic as I experience the EXACT set of nuisance GFCi tripping events...

My setup is as follows... Generator is a older Kohler 10RY, 10kw, 1800 RPM, propane fueled generator... Never experienced a nuisance tripping until the old family cabin was razed and a new cabin constructed conforming to modern US building codes as the old cabin did NOT have GFCi breakers installed anywhere (duh!)... I now have 9 GFCi breakers plus several ArcFault breakers also installed.. some are combo GFCi/Arcfault models... ONLY the GFCI breakers demonstrate nuisance tripping and only when powering the house power system from the Kohler during battery charging... Never trip using the Inverter as a AC power source...

Some interesting details...
1) The new cabin on the generator was initially powered by the Kohler via a Magnum PAE4448 (for 2 years) prior to installing the new Schneider XW-Pro. It was limited by design to about 45 amps of DC charging current. Nuisance tripping when using the single Magnum was never observed... The new XW can easily hit 110 amps. The nuisance tripping seems worse if charging at higher battery charge amps and is a consistent problem observation when using the new XW-Pro inverter/charger as it is typically 2X the charge amps compared to using the prior Maghum inverter/charger...
1A) I added a 2nd Magnum PAE4448 and ran this dual inverter combo for about 2 year always having issues attemting to perform 2X battery charging... firmware issues and the use of Lithium batteries made these attempts difficuilt and the solution was the new XW-Pro inverter...
2) My tripping (over time) breakers all/each have ZERO loads connected to their output receptacles - and, yet they randomly trip in different physical order and at different times... Some just trip and others will begin to flash their visible RED LED before tripping. Some can be reset while flashing RED and others will not respond to a reset until the generator has stopped. I have 5 GFCi breakers on my outside decks and they will all trip off, if exposed to sufficient generator run time, when charging using the Kohler/XW-Pro sytem. My view is it would be difficult if not impossible to create a differential L1 or L2 to White differential load current if there is no load connected... This observation points to generator created AC Noise being the culprit... Stray capacitance on the various internal Romex power cabling can explain the variety of tripping if the problem is AC Noise/Hash...
3) My generator has accumulated about 4000 running hours... I'm sure it is a brushed machine using slip rings and they are likely producing more arc-ing now than when new back in the mid-90s... So, I conclude the THD being produced is worse now than when new...
4) Other observations... Our main room heater has a internal 4 speed blower that goes "wonky" if running when performing battery charging via the Kohler. Also, the light fixture on the ceiling fan will flicker at generator activation; blinking several times ON and OFF untill the power appears to stabilize... Given the XW-Pro battery charger will not begin to produce charge amps until the AC is fully qualified (about a minute), and yet the ceiling lights flash and the heater fan sill spin-up points to AC noise being the cause... The XW-Pro's xfer switch is immediate when it detects "good AC Power" but the battery charger will delay for a TBD time until it is activated...
5) 6 weeks ago, I was doing a high amperage battery charge and audibly heard multiple GFCi breakers trip off in succession (I have 9 in the power system) and I also experienced a smoke detector alarm event AND blew-out 4 very cheap LED 45-watt equivalent bulbs in a upstairs bathroom fixture... I was running the generator at about 6500 watts of output power - approx 6000 watts of XW-Pro power to the battery bank (Discover 4248-6650's). REducing the AC generator load seems to solve this issue - blowing LED lights that are ON when the generator is producing 6500+ watts... The LEDs may have blown when the charger shutdown due to reaching a shut-down prescribed SOC... I'm not sure if it is a graceful shutoff of battery amps or sudden. If sudden, a generator Over-spin could likely result with a big positive spike in AC voltage output...

A couple of things I will try next and report -

a) isolate the cabin from the Inverter charger using the Inverter ByPass switch thus allowing the generator to power the house without the XW-Pro being part of the series power thruput... Run the system for 30 minutes and note any nuisance tripping... If the tripping is associated with generator noise production, I would expect the GFCi breakers to continue to trip off..
b) remove the generator's N to Gnd bond. I'm sure it remains bonded internally inside the generator's control box... I doubt this alteration will affect the issue being reported...
c) I'm intruiged with the suggestion to install a pair of 20 uFD caps between L1 and N and between L2 and N at the generator output lugs... I'm not sure if the suggestion in the thread was to use a pair of 20's for each Leg or to use just ea. 20 Ufd between each power phase and Neutral...

Kindly reply ideas and suggestions... This is a very interesting problem...
Bob Jarrett; aka ClivusM

[clivusmul]

Hello... In thinking - One problem using a pair of 20uFd shunt capacitors (L1/L2 to N) to filter generator produced hash/noise is the shunt capacitor load/reactance placed across each generator leg being about 1 ampere more/leg... It might be better to install a single pole inductive an/or pi-type filter in series with the generator feed legs to the inverter/charger. Several commercial EMI powerline filters are available with series ampere handling capacity of up to 60 amps which is ideal for a inverter/charger similar to the Outback Radian's or Schneider XW series units... If the problem is indeed generator produced hi-freq noise, a inline filter could be very effective... I intend to try this possible solution and will advise... ClivusM/BobJ

[Keithvcollins]

The 20uF run capacitor is only one on each leg. Your description of the GFCI experience is very close to mine. My installation has no inverter. I have two 200 amp panels served by two Kohler automated transfer switches connected to the Kohler 20kw generator. Setting up a Fluke quality meter next week trying to compare with and/without the capacitors. I’m a novice so open to any suggestions on what and how to measure to understand why it appears to be working.

[clivusmul]

Thanks!
To "see" the generator created "noise" (if any..) you need a high-speed digital oscilloscope or a spectrum analyzer.. The later can be expensive but the former isn't cheap either..
My guess is the noise produced has a very wide frequency range, say a few hundred hZ to several megaHz.. I doubt the commonly installed GFci breakers can respond to the very higher end of the noise spectrum so a digital O-scope should easily capture the problematic generator noise on the AC output.
I no longer have access to a quality hi-freq oscilloscope. The least costly "test" IMHO is to install 2ea, passive lowpass inline power EMI filters @ 1/leg on the 230V feeds.. I just received two units made by Schaffner rated for 250V AC/60 amps & will install ASAP & do the b4 & after evaluatuon.
I'll report my results when known..
Lastly, your results using the 2ea, 20uFd shunt capacitors strongly support the AC powerline noise as the root cause for GFI nuisance tripping... I just don't like the high current reactive "load" they place across the mains... The passive filters I bought were about $110 each... More $$ for sure but if they work, the cost diff is worth the extra$ to me...
ClivusM/Bobj

[Pacifico]

Thank you clivusmul and Keithvcollins for your comments. Glad I'm not the only one having issues. I've shared this thread with my installer and hope we can get this issue resolved soon.

[Keithvcollins]

WRT what to monitor. I will have a Fluke 43b quality meter for the weekend so I want to maximize my test runs. Where will be the best point(s) to take readings? At the generator or in one of the transfer panels on the load side? Measure current from one of the legs or common? Focus on voltage harmonics, current harmonics, power, voltage and current wave form? Focus on specific harmonics (triplen)? Sags, swells, transients? Should I try to isolate and measure what’s happening on the common of the capacitors? Any an all pointers/education are welcomed.

[fcwlp]

I would test at load centers and as close to problem CFCI outlets as possible. Do tests with and without capacitors in place.

[clivusmul]

I have no experience with the Fluke 43B but it looks to be a very useful tool based on a brief look at its specs...I believe I read it can measure/capture up-to the 50+ harmonic from the fundemental @ 60 hZs to report power line tHd (which is one indication of "noise" on your internal home grid)... I would think this frequency range would be sufficient to capture the THD produced by the generator well inside the range of sensitivity of our GFCi breakers...

Agree 100% with the prior post just above suggesting you measure at the generator source and at one of the more sensitive GFCi locations. If you have the meter for the weekend, why not "map" the noise signature of the generator at multiple places/outlets across your home... Suggest you also "load" the generator if possible to see how the noise/distortion changes with increasing load power/amps... I think the tHD measured will get worse as you load the generator...

We have zero idea of the magnitude of tHd @ the power feed to the GFCIs required to produce nuisance tripping. I don't think Leviton publishes these types of specs...

Last suggestion - I suggest you connect your meter to a known "clean power" source like the AC power coming to your house from the utility. This is the "no tripping" scenario... See what the tHd reads on the Fluke 43B and note... Measure this characteristic at several known points in your house power distribution system... Next, activate the generator/Xfer switch and move your house to the generator and measure the tHd at the same locations while running on the generator and compare readings... This should provide some indication of how much worse the generator is compared to known "clean power"... If the breakers remain stable while on clean utility power, that is 1 good data point... You could also connect and disconnect your shunt 20ufD capacitors and observe the tHd changes with and without the added shunt filters while running on the Kohler...

Lastly... I did install two in-line/powerline, 60 amp, EMI filters this afternoon at my OffGrid location and the jury remains out on results given I was nearly fully charged and my inverter charger only ran for about 15 minutes from Absorb to Float... I can report that there was no nuisance trippin during this brief test... I have seen this before and will test a few more times before claiming any victory or failure...

bob

[clivusmul]

Not great news to report... I ran my Kohler 10RY for about 30 minutes this AM.. My starting SOC was 85% and after supplying 80 DC amperes to the battery bank for over 20 minutes and bringing it up to 93%, I began to hear various GFCi breakers begin to "trip"... It did run for 25 minutes without tripping which is a new record! I then stopped the generator and began to survey the breaker status individuially...

There's a audible sound heard inside the house as they trip off... It is also clear that my just installed EMI series pass powerline filters are indeed helping as 2 of 9 actually tripped off and 5 remained ON and the two remaining recovered on their own from a "blinking red" LED seen on the bottom corner of the physical breaker when the generator was stopped... All the breakers that tripped off OR began to flash their respective blinking RED LEDs had zero loads connected... I do have GFI breakers installed in my main lug panel and they appear to be OK at this time - I believe these panel mounted breakers are combo ARC-fault/GFI units. I have 3 breakers in my basement near the main lug panel and 2 of 3 were OK after today's test and prior these specific breakers would all trip OFF during generator charging - if allowed to be exposed to the generator running for sufficient time... So, my installation/system seems better now (more robust for the breaker trip-off event) but it is clearly not "bullet proof" at this time...

I have a great room ceiling fan/LED light combo that will flash the LED light as the generator is powered-up. If the fan/light are ON together, there is no OFF flash - only a ON flash if the light is OFF at the time the generator is started. The fan is running however in all instances... This seems to be a indication of a brief power-change transient of some sort. I'm not running the mainroom heater fan so don't know if it would be similarly affected as prior.. I plan to totally bypass the inverter/charger and start the generator to see the affect of the generator on this specific ceiling LED. I believe it will still flash as it has some type of Bluetooth or close proximity RF link to remote control the light ON/OFF from the hand remote... It is pretty "low tech" in terms of robust control of the LED... I think this observation is unrelated to the popping GFCi breakers however...

More study and interactions with the community will be needed to fix this annoying issue... I will also try to chat with a engineer at Schneider if I can.. It's very difficult to chat personally with these experts due to the firewall of CS support agents they tend to surround themselves with that are sited in the Phillipines or India... Who would have thought!

Bob J/AKA Clivusmul