Crippled Midnite Classic 250s no AFCI workaround

[Maast]

Turns out NEC 2014 and on code requires PV installations to have arc fault interrupts on the incoming PV DC and unlike the normal Classic 250s these dont have AFCI in them.

After a bit of a panic and a few intimate hours with google I found that there is a 3rd party DC AFCI solution, the Sensata PVAF-T, it mounts on a din rail and is UL listed. It can monitor two separate lines theough the holes in it and is good up to 1000v DC and 40A. It also requires a external power supply and a contactor but that part is easy. Best of all I found them on ebay for 20 bucks. The only other one I found was $175 by dcsunvolt.

 

[Maast]

Well, crap. Still screwed. 300V DC contactors w/ a 24V coil are expensive as hell. Best so far is direct from Gigavac at about $160, and I need 3.

 

[Maast]

Less screwed - Outback has 75A 300V shunt trip breakers for about $70. So for $90 to overcome the lack of AFCI on a charge controller that was $400 less. I call that a win.

 

[-Geoff-]

Most if not all of the inspectors where I use to work in Chicago and surrounding areas would never approve of someone making an AFIC, or any equipment for that matter. One can not simply take UL listed parts, assemble them and expect it to be approved for use. I have seen this happen with something as simple as someone making a light fixture from UL listed parts, and an inspector not allowing it to be installed. There is a explanation in the following link, in document form UL, about recognized components. E351896-20130717 Certificate of Compliance. There are requirements for an AFCI of which the Sensata PVAF-T is just one part of, it is only an Arc Fault Detector. In the following from Sensata they states there are signals for different functions. So since ther are many different types of signals, I aks which types of signals are needed for the PVAF-T to operate? It also notes a latching reset is needed.... System designers need to fully understand, ect. As i see it ,taking this product is not as simple as connecting a contactor and power supply and meeting all the requirments for an AFCI. It is not at all for a DIY.


From Sensata about their PVAF-T
"If an arc is found, a signal is sent via an open collector transistor to the string control circuitry of the inverter or combiner box."



System designers using the PVAF-T should
incorporate appropriate system responses for both series and parallel
faults in their designs. System designers should also read the UL 1699B
standard to fully understand the differences and capabilities of Type
1 and Type 2 AFD’s before designing an arc fault interruption system
including any UL 1699B recognized AFD

Self Test Feature
Connector Terminations (Pins)
As required by UL, a self test feature is provided on every unit. The self
test interface (PIN 4) provides the user the ability to send an electrical
signal to the PVAF-T to simulate an arc fault condition in the system.
The PVAF-T will perform a self test when the signal is low (short) for
a minimum of 10ms. Note: a latching reset is required to restore the
PVAF-T and the system to proper working function.

 

[Maast]

You may very well be right about inspectors not allowing an assembly from UL parts, however the function should work. In their documentation it states that the output is high at the power supply voltage (24V). The latching reset you refer to is to remove power from the PVAF-T and then repower it.

I've received all 3 units so I'll have to test it on my bench.

I would argue that the detector is performing its one and only function: output of arc fault signal upon detection. Its clearly designed to be mounted on a DIN rail like a breaker and has two holes you thread a PV power cable through and a 4 pin output to a wiring harness. That signal is then input to another piece of equipment with one and only one function, opening a circuit upon a signal. The two are not an assembly but two pieces performing a function.

The DCS sunvolt does the same thing: Outputs a arc fault signal, only in their case it's not a shunt trip its a contactor.

 

[Joelectron]

Hi Maas, any update about your AFCI build ? I'm in the same situation and looking for a secure if not code approved way of retrofitting AFCI in my system.

 

[Maast]

Not yet, since spring I've been focusing on a massive landscaping project: 8 foot high by 98 foot long concrete retaining wall and a colossal amount of fill dirt. I'll pick up the solar/battery bank project in the fall when the weather goes to crap.

 

[summoner12]

Not yet, since spring I've been focusing on a massive landscaping project: 8 foot high by 98 foot long concrete retaining wall and a colossal amount of fill dirt. I'll pick up the solar/battery bank project in the fall when the weather goes to crap.

How is your progress?