disturbance xfmr

[jday92591]

Hello all I have a sb6000US that was manufactured in 2008 and installed maybe 2009. I noticed the other day that I had a GFDI fuse open error code. I replaced the fuse now I am getting a XFMR code. Everything was working fine until about a month ago. We did do an addition to the house with a new 200 amp panel but again everything appeared to working correctly. Any thoughts?

[bcroe]

Hello all I have a sb6000US that was manufactured in 2008 and installed maybe 2009. I noticed the other day that I had a GFDI fuse open error code. I replaced the fuse now I am getting a XFMR code. Everything was working fine until about a month ago. We did do an addition to the house with a new 200 amp panel but again everything appeared to working correctly. Any thoughts?

If that is like my FRONIUS, your Ground Fault fuse (1A here, should never conduct) blew
because something in your DC circuitry was at least momentarily, or possibly just shorted
to ground. Any chance the fuse blew again?

I would disconnect BOTH DC wires from the array to the inverter. Then in some sun
connect one side of an incandescent test light (matching the system voltage) to the
system ground. Connect the other test lead to the array +, then to the array -, and see
if the lamp lights, revealing a ground fault. A DVDM won't give the answer, since it draws
no current and will read static charges. Of course the light should come on strongly if
connected to the array DC output. Array & fuse good, I'd suspect the inverter. Bruce Roe

[jday92591]

Ok it appears I was not turning the system on the right way. Got it turned back on and everything was fine for a couple of days. We had some snow in Temecula CA (very strange) and now I have the GFDI open again. I followed the directions in this video https://www.youtube.com/watch?v=GEyP5dUCvdw and from the article here https://www.mail-archive.com/re-wren.../msg00420.html I tested the positive to ground and negative to ground and the voltage started to decay. According to the article if the voltage would have stayed constant then that would indicate a ground fault in the array. Seeing as it is decaying what do I check next?

If that is like my FRONIUS, your Ground Fault fuse (1A here, should never conduct) blew
because something in your DC circuitry was at least momentarily, or possibly just shorted
to ground. Any chance the fuse blew again?

I would disconnect BOTH DC wires from the array to the inverter. Then in some sun
connect one side of an incandescent test light (matching the system voltage) to the
system ground. Connect the other test lead to the array +, then to the array -, and see
if the lamp lights, revealing a ground fault. A DVDM won't give the answer, since it draws
no current and will read static charges. Of course the light should come on strongly if
connected to the array DC output. Array & fuse good, I'd suspect the inverter. Bruce Roe

[bcroe]

Ok it appears I was not turning the system on the right way. Got it turned back on and everything was fine for a couple of days. We had some snow in Temecula CA (very strange) and now I have the GFDI open again. I followed the directions in this video https://www.youtube.com/watch?v=GEyP5dUCvdw and from the article here https://www.mail-archive.com/re-wren.../msg00420.html I tested the positive to ground and negative to ground and the voltage started to decay. According to the article if the voltage would have stayed constant then that would indicate a ground fault in the array. Seeing as it is decaying what do I check next?

As I said in my last post, measuring this with a DVM won't give a consistent result because of
static charges on such a large set of wiring. Use a light bulb, or get a moving needle meter
that will dissipate the charge more quickly, OR tie a 100,000 ohm resistor there and then measure
the voltage across it. The techs used to call me to a lab they couldn't fix, and while they stood
around with their DVMs, I fixed it in 5 minutes with a light bulb.

If you DO have a hard ground fault, any meter will find it, but not test that it is fixed.

One thing about the video; they required you to take both wire loose. That is so you won't get a
reading through the inverter, since the DC disconnect only breaks ONE conductor. Noting this
problem early on, I set up my DC disconnects to break BOTH wires, making the array easier to
work on and impossible to blow that fuse (again). These 3 pole switches have an unused section.
Bruce Roe

[inetdog]

One thing about the video; they required you to take both wire loose. That is so you won't get a
reading through the inverter, since the DC disconnect only breaks ONE conductor. Noting this
problem early on, I set up my DC disconnects to break BOTH wires, making the array easier to
work on and impossible to blow that fuse (again). These 3 pole switches have an unused section.
Bruce Roe

Under the NEC, if the array is ungrounded, any disconnect MUST open both wires. If the inverter connection grounds one lead, then its disconnect only needs to open the ungrounded conductor, but a disconnect which simultaneously opens both is just fine. You are not allowed to put in a switch that opens just the grounded conductor. And you are not allowed to put a fuse in the grounded conductor unless the fuse has a trip provision for opening the ungrounded conductor(s) too.

[bcroe]

Under the NEC, if the array is ungrounded, any disconnect MUST open both wires. If the inverter connection grounds one lead, then its disconnect only needs to open the ungrounded conductor, but a disconnect which simultaneously opens both is just fine. You are not allowed to put in a switch that opens just the grounded conductor. And you are not allowed to put a fuse in the grounded conductor unless the fuse has a trip provision for opening the ungrounded conductor(s) too.

Not an authority on the rules, but I believe the DC disconnect switch in my inverters only
breaks the +, others do the same which is the reason for disconnecting the wires. My
Fronius inverters detect a ground fault by detecting a blown 1A fuse between - and ground.
If the - were DIRECTLY connected to ground, the fuse would be shorted and never blow.
Bruce Roe

[Sunking]

Bruce or anyone else reading along, there are two ways to provide Over Current Protection.

One is use a Grounded System where one of the circuit conductors is referenced to earth or something in place of each. The other Hot conductor then uses then uses a fuse or breaker. If there is a fault, the hot conductor is opened deenrgizing it and the grounded circuit conductor remains at earth potential keep it safe.

Th eother way is a Floating system where both polarities are HOT, and both circuit conductors are required to have over current protection and Ground Fault Protection. If a fault is detected nothing happens except an alarm is indicated to alert maintenance personnel for an orderly shut down to correct the fault. The only way to denergize a floating system is to disconnect both circuit conductors because both are HOT. You cannot disconnect one without the other to make it safe.

[jday92591]

Forgive my ignorance but how do I build the test light? I have a 4.5 kV systems. the DVM reads 500 volt. what size bulb or bulbs? Thanks for your help!!

As I said in my last post, measuring this with a DVM won't give a consistent result because of
static charges on such a large set of wiring. Use a light bulb, or get a moving needle meter
that will dissipate the charge more quickly, OR tie a 100,000 ohm resistor there and then measure
the voltage across it. The techs used to call me to a lab they couldn't fix, and while they stood
around with their DVMs, I fixed it in 5 minutes with a light bulb.

If you DO have a hard ground fault, any meter will find it, but not test that it is fixed.

One thing about the video; they required you to take both wire loose. That is so you won't get a
reading through the inverter, since the DC disconnect only breaks ONE conductor. Noting this
problem early on, I set up my DC disconnects to break BOTH wires, making the array easier to
work on and impossible to blow that fuse (again). These 3 pole switches have an unused section.
Bruce Roe

[bcroe]

Forgive my ignorance but how do I build the test light? I have a 4.5 kV systems. the DVM reads 500 volt. what size bulb or bulbs? Thanks for your help!!

4 @ 70 watt bulbs IN SERIES should do fine. Smaller will work, cost more. More voltage,
add a bulb. Do be careful with these dangerous levels. Bruce Roe

[bcroe]

Bruce or anyone else reading along, there are two ways to provide Over Current Protection.

One is use a Grounded System where one of the circuit conductors is referenced to earth or something in place of each. The other Hot conductor then uses then uses a fuse or breaker. If there is a fault, the hot conductor is opened deenrgizing it and the grounded circuit conductor remains at earth potential keep it safe.

Th eother way is a Floating system where both polarities are HOT, and both circuit conductors are required to have over current protection and Ground Fault Protection. If a fault is detected nothing happens except an alarm is indicated to alert maintenance personnel for an orderly shut down to correct the fault. The only way to denergize a floating system is to disconnect both circuit conductors because both are HOT. You cannot disconnect one without the other to make it safe.

I agree with that. But apparently the Fronius fails the requirement of either grounding a conductor or
having an internal dual disconnect, and probably other inverters with built in disconnect (check it out).
That is why I installed my own dual disconnect. Bruce Roe

[inetdog]

I agree with that. But apparently the Fronius fails the requirement of either grounding a conductor or
having an internal dual disconnect, and probably other inverters with built in disconnect (check it out).
That is why I installed my own dual disconnect. Bruce Roe

Since the disconnect in the Fronius is a part of the system rather than part of the building wiring, it does not have to follow the NEC, just UL requirements.
But that means that where the NEC requires a disconnect you cannot count the one inside the inverter.

[bcroe]

Since the disconnect in the Fronius is a part of the system rather than part of the building wiring, it does not have to follow the NEC, just UL requirements.
But that means that where the NEC requires a disconnect you cannot count the one inside the inverter.

Agree again. I wonder how many string systems out there do not have dual conductor
DC disconnects or a hard grounded DC input? Bruce Roe