Arc fault nuisance tripping

[Austint]

I have a deye 8 kw with built in arc fault but it nuisance trips with ac loads switching on and off , Why is an ac load causing the inverter to think an fault exists on the dc side ? Thank you

 

[John Frum]

Could it be induced current from the ac circuit to the dc circuit?
To expand... Is it possible that you have an arc signature on the ac circuit that is being induced into the dc circuit?
Or possibly some noise that just looks like an arc signature.

 

[Austint]

It could but I’m not sure how it’s getting there , I initially thought it was going back up the pv ground wire so I temporarily disconnected it but it did not help

 

[John Frum]

It could but I’m not sure how it’s getting there , I initially thought it was going back up the pv ground wire so I temporarily disconnected it but it did not help

Can you do some differential analysis?
When you disconnect the the ac loads the problem disappears?

 

[Austint]

One particular time is when I turn on the ac to the inverter , It trips every time

 

[Austint]

If I leave the incoming ac off to the inverter it will not trip

 

[John Frum]

One particular time is when I turn on the ac to the inverter , It trips every time

Are the ac-in wires in proximity to the pv wires?

 

[Austint]

They are in the same gutter box under the inverters. They run parallel for about 1 foot with 4” of separation

 

[codfish]

you might try twisting the PV leads around each other for as much distance as you can manage. Otherwise they may need a tinfoil hat.

 

[John Frum]

They are in the same gutter box under the inverters. They run parallel for about 1 foot with 4” of separation

you could try to temporarily isolate the cables to see if that helps

or try to turn off the ac charger in the deye.

or try chokes

or try a shield(tinfoil hat)

 

[Austint]

Thank you both I will try this , I thought I had enough separation but maybe not . I was concerned it was just a flaw with the deye

 

[John Frum]

Thank you both I will try this , I thought I had enough separation but maybe not . I was concerned it was just a flaw with the deye

It still could be the ac charger.
Or some other problem with the deye.

 

[Supervstech]

They are in the same gutter box under the inverters. They run parallel for about 1 foot with 4” of separation

I would route the pv elsewhere.

 

[John Frum]

Just had a thought.
If the relay in the transfer switch was marginal it would be pretty similar to a telegraph key.
Also is there a configuration option to disable arc fault protection?
If yes it would be a good baseline to make sure it actually works as expected.

 

[Hedges]

Is that PV arc-fault protection that is tripping? Depending on AC loads and AC input?
EMI filters on the AC side, both input and output, might help.
Filters on PV side aren't recommended because could mask actual arcs. But you said it tripped even with PV disconnected? If so, likely wouldn't even succeed in masking the issue you're having.
AC AFCI sometimes give people trouble due to brush-type motors. But it sounds like just closing AC switch triggers yours.
What kind of AC loads do you have?

The arc noise from AC wires could pass through the unit as conducted interference (which is why I suggest EMI filters.)
Or it could couple as radiated interference. When you had PV disconnected, was that disconnected at the array (open circuit wires still act as antennas to pick up and carry interference into inverter), or was that disconnected at the inverter?

Is battery negative terminal grounded? Is it supposed to be according to manual? That is one more section of wire which could affect radiated and conducted interference.

 

[Austint]

Just had a thought.
If the relay in the transfer switch was marginal it would be pretty similar to a telegraph key.
Also is there a configuration option to disable arc fault protection?
If yes it would be a good baseline to make sure it actually works as expected.

I have it off temporarily right now and it does work as expected.

I would route the pv elsewhere.

How about the dc from the battery in the gutter box ?

 

[Hedges]

I have it off temporarily right now and it does work as expected.

How about the dc from the battery in the gutter box ?

Good thought, try moving battery cables away from PV (and any other) cables. Keep battery + and - cables tightly together.

Battery current draw is DC with rectified AC riding on top, so that is a potential source for interference.
PV AFCI has to ignore SCC switching frequency and harmonics, while looking for much lower noise from an arc. Possibly it sees the 60 Hz harmonics (and inverter switcher harmonics) coupled in from battery cable.

 

[Austint]

But you said it tripped even with PV disconnected? If so, likely wouldn't even succeed in masking the issue you're having.

I disconnected the ground wire that runs parallel to the pv wire to the roof as a test .

Is that PV arc-fault protection that is tripping? Depending on AC loads and AC input?

It is the pv ark fault but I can tell it’s from ac loads or switching . If I have the ac off coming into the inverter and then flip it on , it triggers an arc fault every time .

Is battery negative terminal grounded? Is it supposed to be according to manual? That is one more section of wire which could affect radiated and conducted interference.

It’s not bonded to ground

 

[Austint]

Good thought, try moving battery cables away from PV (and any other) cables. Keep battery + and - cables tightly together.

I had my 8 “ gutter box nice and clean but when the electrician got gone it was just lacking a little spaghetti sauce

 

[Hedges]

It is possible you have a bad connection in the AC circuit, and when current is drawn (inverter charges battery or loads draw power), the bad connection arcs and that is picked up by PV AFCI.
Or that noise present on the AC line, even without you drawing current, is the source.
Are you able to reset AFCI an operate successfully with AC input still connected? (or with AFCI off, connect AC, then reenable AFCI?)

Haven't determined if the noise pickup is conducted or radiated. Routing through alternate path (e.g. extension cord) would be a way to reduce radiated coupling. EMI filter wired in series is a way to reduce conducted coupling (although it's main purpose is usually to prevent noise generated by system from conducting into wire and radiating from there.)

Parallel (and close) wires couple more noise. Orthogonal and widely separated couple less. Closely bundling a wire, its return path, and ground, serves to reduce what it radiates or picks up.