Popping all CAFI breakers when charging Outback with generator

kjcain · June 2024

We have installed an outback radion on new construction that is off grid.
anytime I charge the system with the generator about 20-30 minutes after the generator starts, all the CAFI breakers pop.

i thought this was the older generator, but just replaced it with a Westinghouse WGen9500TFc and the problem persists.

Of course, there are 3 traditional breakers - they don’t pop.
The breakers never pop if on solar, only with then generator.

SumPower · June 2024

Is the ground that goes to the frame of the generator been removed? If not it will create a ground loop with the ground at the electrical panel.

There should be only one location where neutral and ground are bonded together, in the electrical panel that is fed from the output of the Radian.

BB. · June 2024

Always learning something new... CAFI breaker? Note that CAFI appears to add "series arcing" protection,

https://www.se.com/us/en/faqs/FA122435/

CAFI - Combination arc fault circuit breakers (protection against overloads, short circuit, parallel arcing, and series arcing)
Breakers with CAFI suffix provide standard thermal/magnetic protection, PLUS protection against heat damage caused by parallel AND SERIES arcing in the circuit. A series arc occurs when electricity jumps the gap between the strands within the same wire. An example of what may cause a series arc is when an extension cord is damaged because it has been smashed between furniture and the wall, cracking the insulation and exposing strands within one wire. Arcing may occur strand to strand. Breakers designed to protect against parallel and series arcs are called combination arc fault circuit breakers, have CAFI suffix, and have a white test button.

From what I kn(ow (basically about AFI breakers) is that these are looking for Radio Frequency emissions typically generated by Arcing Current. A (relatively) common complaint is that AFI breakers are known to trip when running a brushed vacuum cleaner motor (aka "universal" motor) as the brush/commutator can generate a signal that looks a lot like an arc fault.

Whether you have to use CAFI or AFI type breakers is generally up to your local AHJ (authority having jurisdiction, or "code/inspector" at the building department).

So the first question--Do you have any arc faults in the AC side of the system (typically wire connections that are loose, browned or blackend, etc.)--Can also be an insulation cut (many times caused by a sharp conduit edge as the wiring is pulled).
"
You are "sure" that the system AC wiring is good. Then the next question is "are the CAFI breakers" required for your location.

Personally, I am not a big fan of using AFI or GFI breakers in the main circuits (generator, AC mains input, etc.) as they are the most "unreliable" devices (breakers are designed to fault/open if something appears to be wrong).

I am OK with using AFI/GFI breakers/outlets on branch circuits where one breaker trip (living room outlet goes dark while vacuuming)--Leaves everything else running. And for that reason, I also like to useYou plain circuit breakers on "critical AC loads" (such as a refrigerator/freezer) where you are gone for two days and find all of your food is defrosted/warmed.

If a fridge outlet is next to a garage wash basin/then code may "force" you to use GFI protection (per inspector).

SumPower's suggestion about checking for proper grounding applies to GFI breakers (ground fault interrupters).

GFI work by measuring the current flow in the Hot and Neutral lines. The current "out" the Hot must match the current "in" on the Return line (within 0.025 Amps or less). If you have a "single" Neutral/Ground bond point, you will not have any load current flowing through the green wire ground.

Where a typical issue arrieses, is with generators that have N+G bonding inside the generator, and the house has N+G bonding inside (typically) the main AC panel. This meas that the Green wire and Neutral wires are in parallel and you can have "significant" current flowing through the green wire--And trip Generator GFI breakers.

The normal solution is to "lift" the N+G bonding inside the generator.

Note: Sometimes you can run into AC RFI/Emissions filter capacitors between Hot and Chassis ground (to reduce conducted line noise/Emissions. Not a common problem--But is possible (capacitors inject >0.010 amps of current into the green wire and "trip" GFI breaker. You could unplug loads on tripped GFI and plug in one at time and see if GFI trips resume (note that GFI breakers/outlets are prone to mechanical/electrical failures--Replacing an Old GFI device may fix the problem).

Anyway--Using CAFI breakers (not a GFI issue, that I can see)... Replacing CAFI breakers with standard breakers (if allowed by code) is the "easy" solution. Yes, no AFI protection--But that is similar to >90% (SWAG) of the North American Homes today--So not a big risk.

And to be fair, my late sister's home had an Arc Fault in her main / sub panel (humidity, ocean salt fog, >70 year old home). Was a very messy fault (melted copper in the bottom of the panel).

The failure knocked out/destroyed (?) the pole transformer to the neighborhood. Now an AFI main breaker would have stopped the damage/neighood blackout--So would have a properly designed "Federal Brand" main circuit breaker.

https://homeinspectioninsider.com/federal-pacific-breaker-panel-recall/

Obviously, I am just a guy on the Internet... And I am not your code guy or insurance company (they look for any possible reason to deny a claim). I can offer observations--But these decisions must be yours and based on local regulations (I have not needed to keep up NEC code for a couple decades).

Your thoughts?
-Bill

kjcain · June 2024

Thank you! Nice response with some meat to chew.

Given that this is new construction - low possibility of arc faults in house. If they are present, I would expect them to be consistent on inverter only as well as on generator. Also you would expect them to be on one+ circuit vs flipping 16 circuits (sounds like popcorn!)

unfortunately CAFI is part of the 2019 national code that this was built under

grounding looks like the logical source. I’ll dig into that aspect.

BB. · June 2024

Dave A... Any help here?

Some background reading on Outback hardware (and a second person with Schneider HW?) + Genset tripping GFI outlets after 0-30 minutes of running:

https://forum.outbackpower.com/viewtopic.php?t=17314

A quick read through the 4 page thread--It appears that they have reduced (not 100% solved?) the nuance GFI fault by adding 15-20 microFarad filter capacitors to reduce possible EMI interference (caused by genset?).

Not to say that the above link will help you--But indicate these types of problems can be very difficult to solve (or not) even with factory support. (and Outback is currently mostly "radio silent" at this point in time due to sale of company? Last I read)

Adding motor run capacitors would be a possible "solution" to filter out the Arc Fault "noise" in your case?

Anyway... Back to Arc Fault. What is an CAFI? (try to find as much as possible about how CAFI devices work to better debug issue):

https://www.electricaltechnology.org/2021/09/afci-arc-fault-circuit-interrupter.html

Note that it is reported that about 1/2 of electrical fires are caused by Arc Faults....

https://en.wikipedia.org/wiki/Arc-fault_circuit_interrupter

The electronics inside an AFCI breaker detect electrical current alternating at characteristic frequencies, usually around 100 kHz, known to be associated with wire arcing, which are sustained for more than a few milliseconds. A combination AFCI breaker provides protection against parallel arcing (line to neutral), series arcing (a loose, broken, or otherwise high resistance segment in a single line), ground arcing (from line or neutral to ground), overload, and short circuit. The AFCI will open the circuit if dangerous arcing is detected.
...

AFCIs are designed to protect against fires caused by electrical arc faults. While the sensitivity of the AFCIs helps in the detection of arc faults, these breakers can also indicate false positives by identifying normal circuit behaviors as arc faults. For instance, lightning strikes provide voltage and current profiles that resemble arc faults, and vacuum cleaners and some laser printers trip AFCIs. This nuisance tripping reduces the overall effectiveness of AFCIs. Research into advancements in this area is being pursued.[11]

AFCIs are also known to be sensitive (false tripping) to the presence of radio frequency energy, especially within the so-called high frequency (HF) spectrum (3–30 MHz), which includes legitimate shortwave broadcasting, over-the-horizon aircraft and marine communications, amateur radio, and citizens band radio operations. Sensitivities and mitigation have been known since 2013.[12]

So using an Oscilloscope or Spectrum Analyzer may be helpful to track down the source of the Arc Fault signal.

A typical fault isolation procedure would be to be:

All breakers "on" and wait for the Arc Fault trip (say 30 minutes)
Turn off 1/2 the breakers and wait for the Arc Fault trip (wait 30-60 minutes)
If fault still occurs, turn off 1/2 off remaining breakers and wait 30-60 minutes
If no fault occurs (step 3 or 4), switch breakers (turn on the "off breakers" and turn off the "on breakers")
Continue until you have found the 1 circuit causing Arc Fault trip

Then run the tests again with "bad circuit off"--Just in case there are 2 (or more) sources of Arc Fault signals.

You have swapped out the generator--So it would be unusual for two completely different gensets to be the source of an Arc Fault signal (never say never).
The fact that all (?) CAFI breakers trip at same time--It does sound strange--Like an "upstream of breaker" AF signal source (genset or genset wiring would have been my first guess).

Anyway--CAFI are a "problem"... For instance, hospitals and ICUs appear to be "exempt" from the CAFI requiremnt were false trips can cause life threatening conditions (hard to have any home circuit defined as "critical").

Also, different brands (and even production changes of CAFI breakers) are known to be "more sensitive" to external interferance (such as HAM/Short Wave transmitters in home and near by homes):

https://www.eham.net/article/31262

....

Gruber said that as AFCIs became more common in new construction in the US, reports started coming in that AFCIs in the vicinity -- not just in the radio amateur's home -- would trip in the presence of RFfrom an Amateur Radio transmitter. While each manufacturer's design is proprietary, most AFCIs detect arcs by monitoring the shape of the alternating current waveform, changes in current levels, voltage irregularities, and the presence of high frequency emissions or"noise." The ARRL Lab dug into the problem.

"Last summer we built a test fixture in which we could test any type o fcircuit breaker," Gruber said. It involved using W1AW as an RF source. Gruber said he bought one of "every AFCI that I could get my hands on,"but when the Lab began testing them during W1AW transmissions, none of the devices tripped.

A ham in New Mexico who had reported AFCI problems sent some of his breakers to the ARRL Lab, "and those tripped when we tested them,"Gruber said. The problematic breakers were certain models made by Eaton Corporation. "We already had an Eaton breaker, an older model, but it did not trip," he noted, adding that the breaker had a yellow button.The newer model, which had a white button, did trip in the presence of RF, however, even at power levels down to about 50 W on 17 meters.

Gruber contacted Eaton, and two of the manufacturer's engineers visited ARRL Headquarters in August. "Eaton was extremely cooperative and eager to resolve this," Gruber recounted. "They spent the day with us, going over our test methods and took some of the problematic breakers back with them, eventually developing a modified version.

"We have just finished testing the new version of the breaker, and it did not trip during W1AW transmissions and in other tests," Gruber reported. He said the new breaker is still in the queue for UL approval.

Eaton Engineering Director Andy Foerster said arc fault detection is challenging, in part because so many common household devices -- such as vacuum cleaners and power tools that use motors with brushes --create arcing. In information provided to ARRL Eaton engineer Lanson Relyea said that because AFCIs rely on HF emission detection to verify arcing, "any signal that conducts or radiates a signal within the detection band of the AFCI can cause interference and cause the device to trip without the presence of a true arcing condition."

Eaton and ARRL agreed that when the manufacturer comes out with any new models of breakers, it will ask the League to test them at W1AW. "It'sa win-win situation," Gruber said. Eaton also has agreed to work with anyone having a problem with RF tripping its AFCIs.

Hams experiencing unwanted tripping problems with their or their neighbors' AFCIs should first contact the manufacturer. In the case of Eaton breakers, contact Bob Handick RobertEHandick@Eaton.com (412-893-3746) or Joe Fello JosephPFello@eaton.com (412-893-3745).Read more http://www.arrl.org/news/arrl-helps-manufacturer-to-resolve-arc-fault-circuit-interrupter-rfi-problems.

The above is dated from 2013--So may be of limited help to you--Hopefully current production units of CAFI are "better" at rejecting false AF signals. But contacting the CAFI breaker manufacturer may help.

Electrically speaking, higher power AC loads--Such as the Outback inverter/chargers(?)--Would be a prime suspect of generating AF noise (as laser printers which can draw higher power when printing).

If you turn off the Outback Radian breaker(s), I presume that you don't have any faults?

Can you put the Outback in "by-pass" mode so you can power the other AC "protected" AC loads (I.e, confirm that it is the Outback Charging that is causing AF noise, and not down stream home loads)?

OR if the batteries are charged (or Charging Function disabled for testing), trip or no trip CAFI breakers?

Again, suspecting the Outback Charging circuit is the source of the AF noise--I have seen "H" type switching currents (used in inverters in DC brushless fans) and synchronous rectifiers (more efficient than Diode Rectification) can be "electrically noisy". Calling Outback (if their support people are available) would be a good start.

Using motor run (or motor "start"?) capacitors on the CAFI to Radian wiring to "filter noise"?

Depending on the size of your Radian--A pair of 30 Amp RFI Line Filters between Radian and CAFI breakers here:

https://www.curtisind.com/wp-content/uploads/F1700-F1799-RFI-Filters.pdf

If your Radian is >30 amps, need to look for higher rated RFI Line Filters... If the cheaper Motor Run/Star capacitors don't work. (I would be aiming towards motor run capacitors--Not sure motor start capacitors are the right choice here--Motor Run capacitors are "exposed" to AC voltage/current 100% of the time, motor start capacitors are only in the AC circuit when the motor is starting). A bit more on Run vs Start Capacitors (motor run capacitor appears to be the correct choice here).

https://capacitorindustries.com/wp-content/uploads/Run-and-Start-Capacitors.pdf

Starting with smaller value of Motor Run capacitor and working larger is the better choice... Just placing a MR capacitor across your Hot/Neutral for 120 or 120/240 split phase, or Line to Line (240 VAC circuit) does become an "electrical" load (get into VA vs Power discussion if needed).

Note that these suggestions are getting pretty technical--Not sure if you are "comfortable" with trying any of these solutions (at least without talking with Outback or having an engineer on-site for this work).

I can only do so much guessing reading & typing on my computer and not being on-site/with test gear that I don't own. Lots of guesswork on my side (and a bit out of my comfort/experience zone).

-Bill

BB. · June 2024

Hmmm... My apologies to SumPower.

As I do some more reading, it does appear that CAFI breakers do include the Ground Fault Interruption function too.

A GFI fault is when your two power wires (hot/neutral or Hot/Hot) current do not "add up to zero" current (I.e,. 10 amps out and -10 amps back = zero current, vs 10 amps out and -9 amps back on return, with 1 amp flowing through green wire or conduit ground). Note that the "difference in current balance" reuired to trip a GFI breaker is on the order of 10-30 milliamps or so (0.010 to 0.030 amps)--So it does not take much current "leakage" to trip (I.e., 10.030 amps out, and 10.000 amps return).

So, going back and checking for multiple N+G bonding (such as in both the Genset and the main panel at the same time) is still important.

Do your breakers "tell you" if you have an Arc Fault or Ground Fault trip?

Intermittent failures are a real pain to debug.

I would not assume that new construction would "rule out" installation errors or damaged wiring (nail through ROMEX, insulation cut on sharp conduit burr, loose wire nut/electrical connection/binding screws, crossed Line/Neutral/ground connection somewhere, etc.).

Unless your CAFI breakers have a fault status LED or similar telling you which fault you have (I.e,. over current, CAFI, or GFI fault)--Things get difficult as you have to assume that any fault is possible (probably not an over current trip on all 16 breakers at the same time).

A GFI fault would not (usually) be "fixed" by installing Motor Run or Filter Modules... And MR/FM module could make a GFI fault by itself (if cause unbalanced Hot/Return current flow, or leak some current through green wire ground).

Why does the fault only occur after XX Minutes of charging? Checking binding screws on breakers are tight, double check wiring (especially) to Radan (high current flow during charging) and any wire nuts. Check AC terminal connections in Radian. (poor connections get hot, metal expands, and electrical connection could become "intermittent" or possible arc).

Tripping all CAFI breakers--Hmmm... Possible G+N short (I.e., G+N bond in main panel--per code. And a G+N short down stream (in any branch circuit) could cause (?) multiple GFI breaker trips (allowing current flow in green wire instead of proper/all current in "return"/white wire).

A down stream N+G "short" will not not trip a non CAFI/GFI breaker.

Sorry for my confusion... I have not worked with Arc Fault breakers in my past jobs/homes--So debugging a CAFI arc/gfi fault is new to me.

Can you tell us which CAFI breaker (brand/models) you have? Can look at their data sheets for more information.

-Bill

SumPower · July 2024

Specifications I reccomended as a guideline when a customer wanted a generator for solar backup.

THD (Total Harmonic Distortion) must be 5% or less.

Voltage regulation must be plus or minus 10%.

This typically avoids the problem of the new electronic breakers from tripping.

An added comment is to buy a generator with a 4 pole head.

Or choose any Honda or Yamaha 120/240 inverter generator.

Many of the off the shelf cheaper generators have a THD of 20% or in some cases more. These gen sets will have a higher probability of tripping the newer the Arc Fault/GFI combo breakers.

Just my opinion from personal experience as an electrical contractor.

Popping all CAFI breakers when charging Outback with generator

Comments

Categories