Stand Alone Solar Arc-fault protection devices.

[jesse_c]

Right, the part that I bought is meant to be integrated into an existing system. I already have a PVRSS system, and wanted to add arc detection and shutdown to it.

 

[FilterGuy]

There's a tail module on every panel, thus why I have 32 tail units to go with my 32 panels. The arc fault circuitry is built into every tail module, not the head unit, which is used to help monitor the performance of the system. I haven't researched the details, but I highly doubt the tail units switch open with a pro-active signal, since they have a heartbeat signal, and will open if they don't receive it. The cost was $950 for all pieces for me, landed in the US. (They're selling at wholesale prices to even us DIYers.) Don't confuse what I bought with the pic just above ... pretty sure that's a string level detector ...

Interesting..... I did not see anything in the documentation that mentions arc fault sensing in the tail unit..... Maybe I missed it. Can you point me to any documentation about it? Thanks

 

[FilterGuy]

There's a tail module on every panel, thus why I have 32 tail units to go with my 32 panels. The arc fault circuitry is built into every tail module, not the head unit, which is used to help monitor the performance of the system. I haven't researched the details, but I highly doubt the tail units switch open with a pro-active signal, since they have a heartbeat signal, and will open if they don't receive it. The cost was $950 for all pieces for me, landed in the US. (They're selling at wholesale prices to even us DIYers.) Don't confuse what I bought with the pic just above ... pretty sure that's a string level detector ...

Interesting..... I did not see anything in the documentation that mentions arc fault sensing in the tail unit..... Maybe I missed it. Can you point me to any documentation about it? Thanks

 

[scottvanv]

There's a tail module on every panel, thus why I have 32 tail units to go with my 32 panels. The arc fault circuitry is built into every tail module, not the head unit, which is used to help monitor the performance of the system. I haven't researched the details, but I highly doubt the tail units switch open with a pro-active signal, since they have a heartbeat signal, and will open if they don't receive it. The cost was $950 for all pieces for me, landed in the US. (They're selling at wholesale prices to even us DIYers.) Don't confuse what I bought with the pic just above ... pretty sure that's a string level detector ...

I attached a doc in the post you responded to ... note the arrow from their serial Arc Fault detector into the Tail Unit on slide 2 ...

 

[johnny21]

Right, the part that I bought is meant to be integrated into an existing system. I already have a PVRSS system, and wanted to add arc detection and shutdown to it.

Hi @jesse_c I'm very curious how's your Fonrich arc detection integration with Tigo ended (if already finished)?

I have similar situation (inverter w/o afci and Tigo optimizers) and also looking to integrate stand alone arc fault detection.

 

[Hedges]

I'm considering using Tigo RSD with Sunny TriPower. TriPower has arc-fault, and it can bleed off capacitor voltage when AC goes away (half the RSD task), but I don't think it serves as RSD keep-alive transmitter.
I will need stand-alone RSD transmitter.

What I would like to do is have arc-fault, if triggered, also trigger RSD. Because, I think an arc is more likely to be extinguished if I reduce array voltage and wattage.

 

[FilterGuy]

@scottvanv
Have you progressed with the fonrich system? How is it working?

 

[jesse_c]

Hi @jesse_c I'm very curious how's your Fonrich arc detection integration with Tigo ended (if already finished)?

I have similar situation (inverter w/o afci and Tigo optimizers) and also looking to integrate stand alone arc fault detection.

I'll admit I haven't done the installation yet. I have two inverters, one the arc fault appears to be working okay, the other it is not really working at all, with the 'self test' failing every day sometimes multiple times. unfortunately even after digging into the application code and finding the password to turn off internal arc fault detection, it is not permanent... turned itself back on the next day. So... i will likely need to open the sealed side of the inverter to bypass the art detection CT and then add the fonrich system as originally planned. Other house projects have taken precedence so far though. I'll take another look at it in the new year I think, as the rest of my energy monitoring system is more complete now.

The built in arc detection won't trigger the RSD so it wouldn't extinguish a parallel arc fire, which was the original reason for getting the fonrich in the first place, so it is still high on my list of things to tackle.

 

[FilterGuy]

I'll admit I haven't done the installation yet. I have two inverters, one the arc fault appears to be working okay, the other it is not really working at all, with the 'self test' failing every day sometimes multiple times. unfortunately even after digging into the application code and finding the password to turn off internal arc fault detection, it is not permanent... turned itself back on the next day. So... i will likely need to open the sealed side of the inverter to bypass the art detection CT and then add the fonrich system as originally planned. Other house projects have taken precedence so far though. I'll take another look at it in the new year I think, as the rest of my energy monitoring system is more complete now.

The built in arc detection won't trigger the RSD so it wouldn't extinguish a parallel arc fire, which was the original reason for getting the fonrich in the first place, so it is still high on my list of things to tackle.

What inverter do you have that detects arc-fault?

 

[jesse_c]

The problematic one is a delta m10-tl-us on-grid inverter with buggy firmware and no vendor support.

 

[Hedges]

My less-old Sunny Boy 10000TL-US and newer Sunny TriPower have arc-fault detection.
TriPower also has part of the RSD function: discharge of input caps when AC removed. May be part of a firmware update I received but haven't installed yet (I requested in order to have Rule-21). But no keep-alive so I've bought that separately (Tigo, who also makes the RSD boxes).

I would also like to trigger RSD in event of arc-fault, haven't figured out yet if there is a status output which could do that.
If I get a Sensata arc-fault detector working that could trip RSD.

 

[jesse_c]

in general all grid interactive inverters from the last few years have arc fault detection, since it's a requirement in the NEC. This thread was originally about off-grid systems, which for some reason the inverters often do not have it.

 

[JL_sponge]

My less-old Sunny Boy 10000TL-US and newer Sunny TriPower have arc-fault detection.
TriPower also has part of the RSD function: discharge of input caps when AC removed. May be part of a firmware update I received but haven't installed yet (I requested in order to have Rule-21). But no keep-alive so I've bought that separately (Tigo, who also makes the RSD boxes).

I would also like to trigger RSD in event of arc-fault, haven't figured out yet if there is a status output which could do that.
If I get a Sensata arc-fault detector working that could trip RSD.

Looked into this option (AFD triggers RSD) as a stand-alone AFCI solution and the only challenge with properly satisfying NEC is the discrepancy in clearing times. RSD code allows 30s for PV voltage to get to 80V, while the required AFCI clearing time is a function of arcing power, with max of 2 seconds (https://www.nrel.gov/docs/fy14osti/60660.pdf, pg 7). If you look into SunSpec rapid shutdown requirements, (pg 15) the transmitter is supposed to respond within 2s, but there is an additional 26s allowance for de-energizing and capacitor bleed off.

So while I agree it's a much more comprehensive and safe solution, it seems a fully compliant system would also require a circuit interrupter that can hit the specified AFCI clearing times (built into the inverter or otherwise). But i suppose it depends on the inspector...

 

[daklein]

Looked into this option (AFD triggers RSD) as a stand-alone AFCI solution and the only challenge with properly satisfying NEC is the discrepancy in clearing times. RSD code allows 30s for PV voltage to get to 80V, while the required AFCI clearing time is a function of arcing power, with max of 2 seconds (https://www.nrel.gov/docs/fy14osti/60660.pdf, pg 7). If you look into SunSpec rapid shutdown requirements, (pg 15) the transmitter is supposed to respond within 2s, but there is an additional 26s allowance for de-energizing and capacitor bleed off.

Would the arc bleed down any capacitor within that time on the pv side ( now disconnected ), that'd be a function of the whole system and inverters.
And besides that, the table is the minimum time to detect (or interrupt) an arc of the specified current and voltages. The table any doesn't say what the remaining arc fault voltages need to be after initially detecting and starting the interrupting process.

I still have not installed my Sensata PV-AFTs. I'll be sure to do it after any arc fault issues....

 

[Hedges]

I see what you mean.

That the separately triggered RSD takes time is OK, but need to stop series current flow sooner.
The fireman can wait 30 seconds before climbing on the roof, but 30 seconds of arc could start a fire.

Inverters that to do arc-fault stop converting DC to AC, so capacitor voltage same as PV Vmp. PV would then charge capacitor up to Voc (through the arc) before current stopped. I've measured about 1% ripple in operation, so 1% of Vmp in 1/120th second. Charging up 10% or so from Vmp to Voc could be 1/12th of a second, no problem there.

Internal AFCI, if it has a signal to also trigger RSD, should be OK. Current from operation is stopped immediately. When keep-alive signal is turned off, per-module electronics turn off so if any parallel arcs were occurring, they also stop or are reduced to a single PV panel. That doesn't have a (large) capacitor to discharge, so is probably quick, but possibly waits a few seconds to confirm still no keep-alive. Anyway, even if slow 30 seconds, that would be better than not doing it at all.

If external AFCI, your point would be that simply turning off RSD keep-alive so module level electronics disconnect might or might not happen in 2 seconds. So arc might continue longer. I think those will be quick, but maybe not if they wait so more immune to interference with signal.

I was thinking of using external Sensata AFCI with Tigo RSD module electronics and transmitter. My thinking was AFCI turns off keep-alive transmitter. No need (for AFCI not RSD purposes) to discharge inverter input capacitors. Your point is it might be too slow, although I'm guessing not. If Sensata also disconnected AC from inverter, it would stop immediately.

I have a few inverters. Sunny Boy 10000TL-US-12 has AFCI but not RSD. 6000US has neither. TriPower 30000TL-US has AFCI and capacitor-discharge function of RSD (when AC removed) but not the keep-alive signal. In the case of TriPower, which needs 480V transformer, I would have a relay to disconnect at night so could disconnect that. The others, will need circuit to discharge capacitor for RSD. Spec of 30 seconds vs. 1% ripple in 1/120th second, load about 0.0003 of full inverter rating, about 2W drain. Maybe a resistor always connected, not switched. Or an AC rated relay; DC current low enough to not arc.

 

[scottvanv]

I have been studying the NEC requirements for wiring solar panels and one of the requirements is arc-fault protection if the PV array is at 80V or more.
However, when I search for PV arc-fault protection product.... I find no stand-alone products.

• Some SSRs have it built in.
• Most grid-tie inverters have it built in.
• Some Solar optimizers have it built in.
• Some combiners have it integrated.
• Some Rapid disconnect systems have it built in.

I have not found a stand alone product.... Does anyone know of one?

https://www.fonrich.com/

Only one in the world as far as I know. I bought a full system for 32 panels, have not yet installed it. Had to wire cash in advance. They delivered, and seem like honorable folks.

 

[JL_sponge]

Good points. I guess with the RSD method any series arc should be interrupted as soon as the panel shutdown devices are actuated, regardless of capacitor voltage. In that case we don't care how long it takes to get to 80V. If it were a parallel arc, could sustain longer, but as @daklein mentioned, arc may bleed capacitor rather quickly, and code is only written for series arc anyways. Really comes down to the heartbeat signal frequency, and logic on the receiver end of the RSD system in place, to see how fast it would respond to the AFD cutting the heartbeat signal.

 

[xjan]

@scottvanv any update about your installation of the Fonrich devices?

 

[sircambridge]

I think these are now finally starting to be available at a consumer level as of mid-2023- Im starting to see these arc fault detection protection inturrupt devices on aliexpress, which means there will be an american branded version of the same thing probably soon. here is an example of what i found, I might try to order a few to try out:

I read the description in chinese and it says it protects against arc fault, ground leakage, spark and short dection.

think its worth a try?

https://www.aliexpress.us/item/3256804784629344.html

 

[Mattb4]

Unless I am mistaken those are for AC voltage (230v single phase) and not for DC.