diy solar

diy solar

Stand Alone Solar Arc-fault protection devices.

FilterGuy

Solar Engineering Consultant - EG4 and Consumers
Joined
Nov 26, 2019
Messages
7,870
Location
Los Gatos CA
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?
 
Dirty Pictures!

 
Here ya go


But I only see commercial size units there - 250A
Haven't found any small ones.



"Integrating AFCI functionality within the PV system inverter eliminates the cost and effort of installing additional arc-fault circuit protection components to meet 2011 NEC section 690.11 requirements."

Sounds like they do exist, or did back then.
 
Sorry, I didnt see this before. After a truely stupid amount of searching I've found one and only one solution: A Sensata PVAF-T or PVAF-R, you thread your Array wire through it and it has an internal NC switch that opens if it detects an arc fault. You can use this signal to drive a disconnect. Module requires 12-24v power applied. It says its not for use with optimizers but tigo says they will not interfere with arc fault detection circuits. Guess I'll find out. Right now there are PVAF-Ts on ebay for cheap.

What I'm using it for is to shut down my panel optimizers, I'm injecting the signal break it into the rapid shutdown loop.

And if the inspector doesnt like it, he can kiss my foot 'cause it's the only thing out there that doesnt require me to repurchase very expensive charge controllers.
 
Last edited:
That is an interesting device. I had seen it before but did not pay much attention to it because it is only a partial solution.

I like the idea of tying it into a rapid disconnect.

As I was looking at it just now it occured to me you could repurpose a PV GFP breaker and set it up like this:
1610268768137.png
When the signal from the detector goes high it would trigger the SSR wich would dump enough current through the GFP breaker to trip it.

I don't know if an inspector would accept it but it would probably work.
 
Take a look at the inside of Outbacks AFCI combiner box, you'll see it's got what looks like a PVAF in it. And a gigavac contactor too. They sell their combiner and initiator for $1400 and it has about $200 in parts in it.
PVAF-T_Series.jpgicsbox.JPG
 

Attachments

  • ics_plus_flyer.pdf
    699.7 KB · Views: 27
Good Catch. I figured it must be used in other products someplace.

The contactor is interesting. Notice it has 4 smaller wires going to it. That tells me it is either a latching relay or has secondary contacts for a feedback loop... It is hard to tell from the picture

1610302362075.png


What did you have to do to interface it to your rapid disconnect?
 
All the panels will have Tigo TS4-O optimizers that are also capable of rapid shutdown. There is a wireless 'keep alive' signal sent by a gadget called a CCA along with a TAP. As soon as the power is cut to the CCA/TAP the optimizers shut down.

I'm running the 24V <1A power for the CCA/TAP through the NC switch in the PFAP. The 120v AC for the 24V wall wart DC power is fed direct from the Radian so as soon as the inverter is shut off the panels go into rapid shutdown. The same DPST switch used to cut the inverter also routes power from the chargery to the main gigavac contactor on the battery bank.

This way I get both: rapid total system shutdown, or just AFCI array panel rapid shutdown. Total cost, about $300 plus the optimizers. I have serious winter shading issues so I was always going to have optimizers.

Its dead simple, but you wouldnt believe how many brain bytes and google time I had to burn to come up with it. I've literally spent several weeks on and off trying how to figure how to meet the AFCI and rapid shutdown code requirements and not break the bank.

If the PFAP internal relay can't handle the power for the CCA I'll use the signal to drive a relay control board instead. I got 3 of them: one to screw up, one to use, and one as a spare.
 
Last edited:
This is my kind of geek discussion. I've been considering how to do both AFCI and whole system rapid shutdown, on the long to-do list. Have you gotten yours set up yet, and how's it working?

I got two of those PVAF-T off ebay last year, I see none left now. They were only $20 I think. Planning to use on a garage roof mounted DC array w/ a Midnite Classic 250-CP (DC Solar parts, without AFCI protection). Haven't put them in yet, but a famous youtuber's latest video has me remembering I should have AFCI.

I was thinking similar to your picture FilterGuy. Additionally, maybe interrupt both PV pos and neg? I'm not sure if that would matter or if needed? The most common AF is a series fault, which opening the PV pos should stop. It would not stop a parallel fault. The PVAF-T detects both series and parallel faults. With some other controller, trip the breaker first to open PV pos for a series fault. If the PVAF-T is still faulting a few seconds later, next close a contactor (or better, a mechanical latching relay) across PV pos & neg to kill the parallel fault. If it's still somehow detecting arcing, I don't know, maybe it was a series fault, try again? Or sound an alarm.
 
This is my kind of geek discussion. I've been considering how to do both AFCI and whole system rapid shutdown, on the long to-do list. Have you gotten yours set up yet, and how's it working?

I got two of those PVAF-T off ebay last year, I see none left now. They were only $20 I think. Planning to use on a garage roof mounted DC array w/ a Midnite Classic 250-CP (DC Solar parts, without AFCI protection). Haven't put them in yet, but a famous youtuber's latest video has me remembering I should have AFCI.

I was thinking similar to your picture FilterGuy. Additionally, maybe interrupt both PV pos and neg? I'm not sure if that would matter or if needed? The most common AF is a series fault, which opening the PV pos should stop. It would not stop a parallel fault. The PVAF-T detects both series and parallel faults. With some other controller, trip the breaker first to open PV pos for a series fault. If the PVAF-T is still faulting a few seconds later, next close a contactor (or better, a mechanical latching relay) across PV pos & neg to kill the parallel fault. If it's still somehow detecting arcing, I don't know, maybe it was a series fault, try again? Or sound an alarm.

Me too. If I ever move panels from ground mounts to roof top, I'll add AFCI and RSD.

I've seen 4-panel RSD units. For my "12V" 120W Astropower panels, three in series would be within the voltage/wattage limits, so 12 panels on one RSD unit, two of those for a 24 panel series string.

I use AC coupled PV inverters, so all strings feeding one or more inverters would go through a single PVAF-T AFCI, and it would open a relay to shut off AC. That is similar to the ground-fault function of the inverters, which have a 1A fuse from negative (or positive, for positive grounded PV array) to ground. If fuse blows due to a ground fault in the PV array, inverter shuts off.

What's more difficult in your case is interrupting DC, rather high voltage or current. (Unless you have a remotely controllable breaker.) But if you also have rapid shutdown units, just trip those. That will isolate each panel, should stop any arcing. (I should consider same, do away with the AC relays I was planning.)

Unlike my AC coupled, your panels would have a path to ground through negative side. But only if battery negative grounded? Thought for a moment that if ungrounded, no arc would occur. But come to think of it, in that case would drive battery to a high voltage relative to ground which would be hazardous.

Don't think you want to short PV positive to negative, as you note that would carry a series fault.
 
Another thought I'm having, is to pull a permit and have my system inspected. (shudder...) I would need rapid shutdown for the DC array, and the shunt trip breaker for AFCI would do both. I think my state currently uses the 2017 NEC code, so the array has to be de-energized _outside_ of 3ft from it. My DC panels are on garage roof right above the balance of the off-grid system, so I could move the combiner box up on the wall a bit to be within 3ft, and put a two pole shunt trip breaker there to disconnect the combined pos, and the combined neg PV. At the charge controller, another single pole PV shutoff breaker and the battery breaker.

For the AFCI / RapidShutdown disconnect means, there's a Midnite breaker (MNDC-GFP100) that would work. Any other good ideas or alternatives?
 
Last edited:
Midnite has a few shunt trip breakers, for example the MNEDC250RT, the only caveat is that they require 1 amp to trip, which is more than the PVAF can handle directly but a interposing SRR would handle that.
 
Midnite has a few shunt trip breakers, for example the MNEDC250RT, the only caveat is that they require 1 amp to trip, which is more than the PVAF can handle directly but a interposing SRR would handle that.
The midnite MNDC-GFPxx series have a 1/2A trip breaker. However, you are probably right. You would want to drive at least an amp through it to get it to quickly trip.

Did you see this image from my previous post?

1622870299770.png

However.... the NEC talks about 'approved' equipment. I am not sure this would pass muster even though it would work.
 
Yes, your picture would work, probably most any shutoff device would need a relay or SSR between the PVAF-T and the shutoff. For the same shutoff to also meet rapid shutdown, a 2 pole breaker could be used. I was wondering about any other ideas about a 2 pole dc breaker with shunt trip or remote trip. Maybe two single pole remote trip breakers like MNEDC125RT or 250RT. I'll keep poking on ebay, or I have a huge carling 4pole remote trip switch, 150v, 156A per pole, that would work! Kind of saving that for a battery disconnect though, using another one 4poles to disconnect batt pos, >600A!

MNEPV-600RT, 16 or 20a, removing the jumpers between poles, that could disconnect two <150v strings, pos & negative (4 poles, minding polarity). The disconnect could be on the array side of the existing combiner. I have three strings (and potentially a 4th), so 2 of those RT breakers would do it. _And_, it wouldn't require combining the strings, so I'd be free to mix and match the strings to different battery bank charge controllers. That might be the best.

Approved equipment would be the breaker and the PVAF-T, at least should meet NEC2017. 2019 requires the rapid shutdown equipment to be listed so I could see someone saying the whole system with the little bits like SSR included has to be listed as a system for that purpose. My state is currently on NEC2017.
 
Last edited:
For the same shutoff to also meet rapid shutdown, a 2 pole breaker could be used. I was wondering about any other ideas about a 2 pole dc breaker with shunt trip or remote trip. Maybe two single pole remote trip breakers like MNEDC125RT or 250RT.

A 2-pole breaker would disconnect positive and negative of PV, but you'd still have Voc between the wires and full wattage.
At one point it was to be < 80V between any two points and < 300W (which wouldn't work even for microinverters with higher wattage panels) but not sure that any watt limit was kept.

RSD would have a box per panel (or some support 4 panels), and a device that transmits "keep alive" signal. If you have a switch to enable keep alive (not integrated into inverter and controlled by AC), just route the signal through AFCI contacts.


"Remote Trip Input: Normally Open (N.O.) disconnect switch closure to GND Internally pulled to +3.3 VDC Not recommended for DC levels"

If ACFI is controlled by a GT inverter, use a relay to disconnect AC.
 
A 2-pole breaker would disconnect positive and negative of PV, but you'd still have Voc between the wires and full wattage.
At one point it was to be < 80V between any two points and < 300W (which wouldn't work even for microinverters with higher wattage panels) but not sure that any watt limit was kept.
Doesn't that only apply to outside of the 'danger zone' around the panels? As I recall, the zone was something like 2 feet away from any of the panels.
 
Earlier revision was only outside the array.
Later revision included all wires of the array. Exception for building-integrated, like solar shingles with wires inside the attic.
For a time it was 30V and 300W limit. Not sure if that was enacted before being changed, but now 80V and possibly no wattage limit for RSD. I think AFCI has a wattage limit.
 
I was wondering about any other ideas about a 2 pole dc breaker with shunt trip or remote trip.
Midnight and outback both make multi-pole ground fault breakers:
 
Earlier revision was only outside the array.
Later revision included all wires of the array. Exception for building-integrated, like solar shingles with wires inside the attic.
For a time it was 30V and 300W limit. Not sure if that was enacted before being changed, but now 80V and possibly no wattage limit for RSD. I think AFCI has a wattage limit.
Wow.... I just looked it up and your right. That is tough requirement to meet without a lot of equipment if you are using a long string of panels.... It kinda drives toward micro-inverters and 'just' shutting off the AC so the Micro-inverter shuts off.
 
Back
Top