Solar edge 11.4 on high leg shutting down

wwhitney said:
How quickly do the inverters trip out? If they don't trip until they try to put out some current, and then quickly trip on overvoltage, that would point to a source impedance issue. That might happen fast enough your meter wouldn't catch it?

You could try turning on the inverter in low light conditions (or otherwise limit how much current it can put out), that would let you see if it will hold without tripping, and checking what the voltage rise is compared to the current it is putting out.

Cheers, Wayne
Click to expand...
I do not know the specifics of the 5 minute grid qualification procedure. I dont think they are 'pinging" the grid with any output, but that is a total guess. I dont think it would be an impedance issue anyway. I was there around 5:30 and the other inverters were only making about 1500 watts, so it was far from peak output time. Also the other inverters were pushing juice into B and C with other phasing combinations and I did not see any strange voltage behavior.
 
In post #5 you said "none of the inverters would work" on the BC phase and referred to "trying to get them to come on". Implying they never did? Did you see any indication that they turned on even for a second and tripped off? Or did they just indicate 'waiting' the whole time?

My impression was they never came on at all and therefore never thought they had a 'clean grid'. If so, I don't think source impedance has anything do with it. Somehow they can tell that they are on the open phase, and it's not by the RMS voltage.
 
jaggedben said:
In post #5 you said "none of the inverters would work" on the BC phase and referred to "trying to get them to come on". Implying they never did? Did you see any indication that they turned on even for a second and tripped off? Or did they just indicate 'waiting' the whole time?

My impression was they never came on at all and therefore never thought they had a 'clean grid'. If so, I don't think source impedance has anything do with it. Somehow they can tell that they are on the open phase, and it's not by the RMS voltage.
Click to expand...
Right I don't think they ever came on. I was just going off the silly LEDs, as you know solar edge does not have any sort of display. The red "fault" LED would come on after I don't know maybe 20 seconds, and then it would try again in a bit and same thing. Never saw any current with my clamp meter, and the green LED was never solid, just blinking. Solid means producing blinking green means has AC power but not producing.

BTW this sort of is not my problem. I just did the AC tie in for this, and then said I would look at it. I don't have access to the online/wifi information as to the specifics of the fault, I am just going off what the solar contractor told me. The solar contractor is talking to solaredge, but they don't seem to be particularly helpful from what I hear.
 
electrofelon said:
I don't think so. I was playing around in the combiner panel which is right next to the inverters. I was looking at voltage the whole time, specifically when the inverters were first energized and qualifying the grid. I never saw any high voltage, or any change whatsoever from approx 246.
Click to expand...
It's hard to troubleshoot intermittent problems. It can look great 99% of the time but that 1% is tripping the inverters off. You would have to have a recording voltmeter on the AC combiner and cross check the voltage reading at the time the inverter recorded the error and shutdown. It sounds like the owner needs to bring in a troubleshooter with the right equipment and access to the error codes. Like you said, this is not your problem and it's not something that is going to be a quick fix.
 
electrofelon said:
BTW this sort of is not my problem. I just did the AC tie in for this, and then said I would look at it. I don't have access to the online/wifi information as to the specifics of the fault, I am just going off what the solar contractor told me. The solar contractor is talking to solaredge, but they don't seem to be particularly helpful from what I hear.
Click to expand...
This is the solar contractor's problem, not yours,, but they might want to reconsider their interconnection strategy. If the center tapped utility transformer is up to the task, I would think about ditching the connections to the B phase and just connecting to the A and C phases as if it were a standard single phase service. Did anyone ask the POCO which type of connection they want? For all of these I have dealt with for the last few years the POCO has told me not to connect to the B phase.
 
electrofelon said:
Update:

Had a phone meeting with the solar contractor and the utility field rep. Kind of went the may i expected, with the utility guy skeptical there was anything wrong on his end :rolleyes: . One interesting thing I found out was the utility bank is an open delta.
Click to expand...
I have worked with the local POCO's here on allowing PV to be on Open Delta's and the only thing they allow is for the inverters to be matched to the transformers, and a most will only allow it on the lighting transformer. Keep in mind a open delta is often unequal size transformers also. I have seen combinations like a 167kVA with as small as a 15kVA

My extremely limited understanding of inverters is they need to be matched with actual transformers.
I would imagine if you put a inverter across the open or ghost phase (B-C) you could have several problems
the B-C inverter will push power towards the other inverters raising their voltage unequally as well as the other two transformers
I would suggest Solaredge to update their diagram to add this;

solar_open_delta.png
 
ggunn said:
This is the solar contractor's problem, not yours,, but they might want to reconsider their interconnection strategy. If the center tapped utility transformer is up to the task, I would think about ditching the connections to the B phase and just connecting to the A and C phases as if it were a standard single phase service. Did anyone ask the POCO which type of connection they want? For all of these I have dealt with for the last few years the POCO has told me not to connect to the B phase.
Click to expand...

One thing occurs to me. If all the load shuts down isn't there a transient voltage spike because of the inductance of the transformer, and if the load is the only load on the B phase, mightn't that spike be more evident on the B phase than on the other two that are more loaded? Just an idea...
 
Last edited:
ggunn said:
One thing occurs to me. If all the load shuts down isn't there a transient voltage spike because of the inductance of the transformer, and if the load is the only load on the B phase, mightn't that spike be more evident on the B phase than on the other two that are more loaded? Just an idea...
Click to expand...
We had a case where a worker accidentally shorted the L1 and L2 buses of a residential panel, which caused a large amount of current to flow through the transformer that supplied both the residence and the next door neighbor. When the MCB of the faulted panel opened there was a voltage spike on the shared service conductors that zapped some sensitive electronics in the neighbor's house. As we learned in EE101, you cannot instantaneously change the current through an inductor, and Ohm's Law did the rest.
 
I wonder what % of utility 240 delta services are open delta? Here in Oregon you often find they only have 2 primary hots and neutral, so there is not even a option to close the delta.
If I were the utility engineer and the lighting (A-C) transformer is indeed larger than the power transformer (A-B) I would have the PV match the ratio of power pot to lighting pot kVA say you have a 15kVA power pot (A-B) and a 75kVA lighting (A-C) I'd say follow that ratio for connected inverter nameplate:
solar_open_delta_lighting_larger.png
 
tortuga said:
I wonder what % of utility 240 delta services are open delta? Here in Oregon you often find they only have 2 primary hots and neutral, so there is not even a option to close the delta.
If I were the utility engineer and the lighting (A-C) transformer is indeed larger than the power transformer (A-B) I would have the PV match the ratio of power pot to lighting pot kVA say you have a 15kVA power pot (A-B) and a 75kVA lighting (A-C) I'd say follow that ratio for connected inverter nameplate:
View attachment 2576924
Click to expand...
I don't know about "matching" the inverter Pmax to the kVA of the transformer, but it definitely should be less than or equal to.
 
tortuga said:
I have worked with the local POCO's here on allowing PV to be on Open Delta's and the only thing they allow is for the inverters to be matched to the transformers, and a most will only allow it on the lighting transformer. Keep in mind a open delta is often unequal size transformers also. I have seen combinations like a 167kVA with as small as a 15kVA

My extremely limited understanding of inverters is they need to be matched with actual transformers.
I would imagine if you put a inverter across the open or ghost phase (B-C) you could have several problems
the B-C inverter will push power towards the other inverters raising their voltage unequally as well as the other two transformers
I would suggest Solaredge to update their diagram to add this;

View attachment 2576922
Click to expand...
A while back we had a thread where (ircc) Wayne calculated the voltage shifts for us from backfeeding a hypothetical-but-realistic open delta and it was a relatively minor effect. Of course it depends on the actual transformers' impedances but I think the point was there is no general theoretical reason you shouldn't be able to backfeed the open phase of an open delta if it behaves pretty ideally with respect to impdeance. According to electrofelon the utility decided the open delta was okay to backfeed, and I'm assuming they decided that after studying the impedances. So I think something else is going on here. I understand that the product of the AB and AC voltage sine waves should ideally result in a pure sine wave for the BC voltage, but I suspect for some reason on this site it actually isn't. (Because of capacitive or inductive stuff on the network or site? Something causing extra zero-crossings? Above my head to figure out, I'm just guessing.)

To rephrase what I've already said, from what electrofelon said it seems like the inverter never tried to output power, which means that if the measured RMS voltage was in range, and the frequency is okay on the other phases, there is something about the BC voltage *waveform* that the inverter doesn't like. I suspect some non-ideal behavior from the interaction of the two other phases.
 
ggunn said:
I don't know about "matching" the inverter Pmax to the kVA of the transformer, but it definitely should be less than or equal to.
Click to expand...
Not Pmax I was suggesting matching the kVA ratio of two unequal sized transformers so if I had a 25kva and a 75kva open delta bank and 4 equal size inverters I would put one inverter on the 25kVA and three on the 75kVA.
 
tortuga said:
Not Pmax I was suggesting matching the kVA ratio of two unequal sized transformers so if I had a 25kva and a 75kva open delta bank and 4 equal size inverters I would put one inverter on the 25kVA and three on the 75kVA.
Click to expand...
Of course, but if it were me I would put all the inverters on A to C at 240V as long as the center tapped transformer could handle it. The POCO won't care about the power imbalance; in fact, they are counting on it.
 
Here's a stupid question: Are the inverters that are connected to the B phase trying to use the neutral?
 
jaggedben said:
A while back we had a thread where (ircc) Wayne calculated the voltage shifts for us from backfeeding a hypothetical-but-realistic open delta and it was a relatively minor effect. Of course it depends on the actual transformers' impedances but I think the point was there is no general theoretical reason you shouldn't be able to backfeed the open phase of an open delta if it behaves pretty ideally with respect to impdeance. According to electrofelon the utility decided the open delta was okay to backfeed, and I'm assuming they decided that after studying the impedances.
Click to expand...
We had a phone call with the POCO DG guy and they definitely are aware of and requested the inverter arrangement. I specifically told the POCO guy that I was surprised they allowed such a connection (I wanted to just double check it wasn't a mistake somewhere that somebody missed) and he said something like, "well we've never really had any problems, and as long as the transformers are sized correctly there shouldn't be an issue."
 
tortuga said:
I
I would suggest Solaredge to update their diagram to add this;

View attachment 2576922
Click to expand...

Yeah I cannot find any information in any solaredge literature differentiating an open and a closed Delta, not allowing an open Delta etc. I also looked at the fronious SYMO and the same thing, they just say Delta but no mention of if open is acceptable or not. I think these inverter manufacturers have a little naivety on Delta power systems in the US
 
You must log in or register to reply here.
Top