Strange behavior with XW6848+ today

DickyDck · April 2020

Hello all! so today I noticed that I had on the inverter the 300 second countdown, but it would only go down to like 298-ish and then go back to 300. This to me indicated a strong hard persitant ongoing fault to through the 33 code "AI Over Voltage L1L2" error. According to the description when govering over the error, it states I need to do nothing, when the fault is cleared it will countdown and be good again. That was 6 hours ago. So I decided to do some troubleshooting myself, basic things.

I reset the all parts of the system to factory and set the things like Charging amps, Battery type, AH, etc.

The system will run without throwing the error code if I leave Grid support / Grid Sell off BUT it doesn't harvest any power from the solar panels, it is in AC Passthrough

A few minutes after turning on Grid Support, it will throw the error, not like it makes much difference now since I wasn't harvesting crap from the solar array anyway.

At this point at no point has any power been generated from the solar array.

SO rinse repeat with the factory reset to default, in the PDP I hit the breaker to kill grid input, now the solar panels instantly start generating power and powering everything in the house (minus things not supposed to like A/C, Water heater, Heating, stove, etc) with no error.

This is where I have left it at since at least now my solar array isn't a backyard paperweight and I'm getting power generation and saving something on my electric bill.

Now I know Anti-Islanding can happen if AC1 Voltage goes above/below the thresholds set, and default is 106/132. Same with going over/under with the AC1 Frequency which is 55/65 default.

Neat thing is the general page on the Combox shows me that the Frequency was steady at 60Hz, and ACL1 was around 130.6V ACL2 was around the same -/+ .2 Volts so combined it is right around 260, and the AC1 High Voltage being 132 x 2 is 264 so I think that isn't enough to trigger it, at least not consistent to keep that 300 timer up.

Any ideas or suggestions are welcome as always, I also have a support request to Schneider, since this gear isn't even 2 years old installed, so I "think" it is still under warranty. I have tried to reach out to the installing company but if you guys look up PEG solar, you will see they are in financial crisis and they have not yet called me back, their info@ email address returns undeliverable, so I will probably have to find someone to troubleshoo and pay them. If any of you are up for some side business and near Winchester VA let me know!

BB. · April 2020

Take a look at the L1 to Neutral and L2 to Neutral voltage...

If you have a bad Neutral to Ground bond (typically in the main panel), you will get 120 L1 to N and L2 to N, but Lx to Ground will not be 120 VAC, and N to Ground will be more than a couple of volts AC (not sure if the XW measures L1/L2 to ground or not).

The other bad wiring can be an "open neutral" back to the pole transformer. That will mean that L1 to N or L2 to neutral is note 120 VAC... You can notice this if (for example), you turn a heavy load (such as the garbage disposal), and you notice that the lights may dim (or even go brighter),

The 300 Second timer starts counting down while L1&L2 to Neutral, and frequency (60 Hz +/- 1 Hz) for 5 minutes. If one of those falls out of range, then the timer restarts. If you have another "weak connection (like L1 or L2 to the inverter connection point), as it starts up, the Inverter Side of the connection goes out of range).

Look at the L1/L2/N connections. Check that the connections are tight and clean. I have had issues where a small water leak follows the wire (from, for example, the weather head), down to the Meter Socket or Main panel connections--And over time, the connection(s) corrode and I noticed that my house lights were "flickering".

-Bill

DickyDck · April 2020

Thanks! to be clear we are talking about checking the connections within the inverter side connections, or outside at the grid side. and Main panel ar you referring to the actual breaker box for the house or in the Schneider PDP panel.

BB. · April 2020

Basically, check everything from the Inverter AC1 input, back to the Meter Socket.

I had one house that the neutral bus connection corroded. And another house where one of the meter socket connections corroded.

Found the meter socket issue after a checked and torqued everything from the GT inverter to the main disconnect. Had to cut the seal, pull the meter, fix the socket, and call the utility to re-seal the meter (no problem).

-Bill

DickyDck · April 2020

awesome, thanks for the clarification, I'll do some looking at. I dont want to get too in there since I am accident prone

DickyDck · April 2020

Quick run through on every connection possible except the original house fuse panel, no corrosion on anything at all, in fact they cables and mount/clamps/blocks are still shiny looking new. So one of you experts double check my logic, again I'm not a electrical engineer rather I'm an I.T. geek, I will have to test again tomorrow. If I turn off the breaker that disconnects any input from the Grid, it seemed to run just fine and runs off Solar and if it was shady out it would draw from the battery bank. This to me proves the inverter is "capable" of doing its purpose of "inverting" DC to AC and sending it to the PDP-->Critical Loads Panel. Since the issue happens when I turn on Grid input, somewhere from the Grid input breaker and upstream to the outside meter, is where something is borked. Especially since according to my readings on the combox voltage is within range, and the frequency. this all sound sane?

mcgivor · April 2020

@DickyDck What are your charging voltage setpoints for the inverter and charge controller? The inverter should be slightly lower for example 0.1V, this would allow the PV to take priority in charging as it would appear the inverter is overwhelming the controller with its unlimited capacity from the grid.

Another option would be to program a charge block during daylight, this would allow full use of the available PV with priority going to charging the bank. One possible downside is the inverter will start another charge cycle once the block period expires, this may be preventable through programming, not having grid I've not gone down that path, but have used the first method during generator charging.

These may, or may not address the the AC problem, but will allow the PV to function as intended.

DickyDck · April 2020

mcgivor said:

@DickyDck What are your charging voltage setpoints for the inverter and charge controller? The inverter should be slightly lower for example 0.1V, this would allow the PV to take priority in charging as it would appear the inverter is overwhelming the controller with its unlimited capacity from the grid.

Another option would be to program a charge block during daylight, this would allow full use of the available PV with priority going to charging the bank. One possible downside is the inverter will start another charge cycle once the block period expires, this may be preventable through programming, not having grid I've not gone down that path, but have used the first method during generator charging.

These may, or may not address the the AC problem, but will allow the PV to function as intended.

Thanks! I actually have a charge block during daylight hours already, this was programmed a while back. I dont know what happened this morning that borked everything. right around 0930 East coast time it went nutty and hasn't recovered since. that error is still blinking right now.

DickyDck · April 2020

also my CCs are set to charge 2.0 volts higher than the inverter, this was how I've had it set for over a year. Here I thought things were all good to go when I changed out the battery bank from 220ah to 440ah. I'm hunting around for someone certified to work on soalr systems out here now.

DickyDck · April 2020

Update: Seems the error cleared itself someime last night, looks like 19:49:49, and this morning so far..... everything is working normal. It was about 0900 when things went haywire yesterday so I have about 30 min to observe and see if it happens again.

DickyDck · April 2020

ok, and it is back....

DickyDck · April 2020

it is so strange, watching the 5 minute counter go from 300 down 298 then hopping back up to 300, sometimes just sitting at 300 for a few minutes, but never fully clearing

BB. · April 2020

Put a good DMM on the AC input terminals for the inverter while this is happening and see if you can find any strange behaviour in the L1/L2/N voltages.

Given that it looks to be time of day sensitive, are there any heavy loads (well pumps, large induction motors, arc welders, plasma cutters, etc.) on your place or nearby that may be creating problems on the utility line?

-Bill

Dave Angelini · April 2020

I doubt you will see much without some expensive and fast logging equipment, unless your really in the 3rd world. XW is very fast at detecting grid Hash.

DickyDck · April 2020

@BB
No nothing out of the ordinary, single family home with 1 guy and 2 pets, so no heavy gear running

@Dave Angelini
correct, I have nothing on an industrial level to perform any troubleshooting with.

Dave Angelini · April 2020

If you really think it is the grid then call the Utility. They will leave a nice logging unit for a few days.

DickyDck · April 2020

thanks! I pinged them earlier, and waiting on a call back. If the combox is anywhere near accurate then I dont think it is the utility company. I have custom logging setup to log the AC line readings every minute, and they are steady under the threshold. I also reached out to the closest solar company here and worse case I'll pay them to come out.

DickyDck · April 2020

Also, what is "strange" behavior in the line? I can only monitor voltage and frequency....

TIME	(0)XW Grid AC Current [A]	(0)XW Grid AC Frequency [Hz]	(0)XW Grid AC Input Current [A]	(0)XW Grid AC L1 Current [A]	(0)XW Grid AC L1 Voltage [V]	(0)XW Grid AC L2 Current [A]	(0)XW Grid AC L2 Voltage [V]
4/17/2020 12:00	-3.3	60	3.3	2.7	129.3	-6	128.4
4/17/2020 12:01	-12.6	60	12.6	-4.2	129.3	-8.5	128.2
4/17/2020 12:02	0	60	0	0	129.2	0	128.9
4/17/2020 12:03	-10.5	60	10.5	-1.6	129.6	-9	128.8
4/17/2020 12:04	-9.8	60	9.8	-1.5	129.3	-8.3	128.6
4/17/2020 12:05	-11.7	60	11.7	-2.7	129.4	-9	128.4
4/17/2020 12:06	-10.9	60	10.9	-2.6	128.8	-8.2	128.2
4/17/2020 12:07	-11.5	60	11.5	-2.5	129.3	-9	128.4
4/17/2020 12:08	-14.8	60	14.8	-3	129.6	-11.8	128.4
4/17/2020 12:09	-11.9	60	11.9	-3	129.6	-8.9	129
4/17/2020 12:10	-12.1	60	12.1	-3	129.5	-9.1	129
4/17/2020 12:11	-11.2	60	11.2	-2.9	129.7	-8.3	129.3
4/17/2020 12:12	-11.3	60	11.3	-2.9	129.9	-8.3	129.1
4/17/2020 12:13	-11.3	59.9	11.3	-2.9	128.4	-8.4	127.9
4/17/2020 12:14	-11.9	60	11.8	-2.9	128	-8.9	127.4
4/17/2020 12:15	0	60	0	0	128.3	0	128.2
4/17/2020 12:16	-11.1	60	11.1	-2.1	128.2	-9	127.4
4/17/2020 12:17	-11	60	11	-2.1	127.9	-8.9	127.2
4/17/2020 12:18	-11.2	60	11.2	-3	128	-8.2	127.3
4/17/2020 12:19	-10.1	60	10.1	-2.2	129.8	-8	128.8
4/17/2020 12:20	-10.9	60	10.9	-2.1	129.6	-8.8	128.9
4/17/2020 12:21	-10.6	60	10.6	-2.4	129.7	-8.3	128.9
4/17/2020 12:22	0	60	0	0	130.6	0	129.8
4/17/2020 12:23	-10.1	60	10.1	-2.1	129.8	-8.1	129.4
4/17/2020 12:24	-10	60	10	-2	129.9	-8	129
4/17/2020 12:25	-10.2	60	10.2	-2.3	130.1	-7.9	129.2
4/17/2020 12:26	-10.8	60	10.8	-2.1	130.1	-8.7	129.2
4/17/2020 12:27	-11.2	59.9	11.2	-2.4	130.1	-8.8	129.4
4/17/2020 12:28	-10.5	60	10.5	-2.4	130	-8.2	129.3
4/17/2020 12:29	-10.5	60	10.5	-2.3	129.7	-8.1	128.9
4/17/2020 12:30	-11	60	11	-2.3	129.7	-8.7	128.9
4/17/2020 12:31	-11	60	11	-2.3	129.4	-8.7	128.8
4/17/2020 12:32	-11	60	11	-2.3	129.6	-8.7	128.9
4/17/2020 12:33	-11	60	11	-2.3	130	-8.7	129.2
4/17/2020 12:34	-10.2	60	10.2	-2.3	130.2	-7.9	129.1
4/17/2020 12:35	-11.8	60	11.8	-3.1	129.8	-8.7	129.1
4/17/2020 12:36	-11.2	60	11.2	-3.1	129.6	-8.1	128.9
4/17/2020 12:37	-10.9	60	10.9	-2.3	130.8	-8.6	129.1
4/17/2020 12:38	-11	60	11	-2.3	130.2	-8.7	129.8
4/17/2020 12:39	-11	60	11	-2.3	129.9	-8.7	129.3
4/17/2020 12:40	-11.1	60	11.1	-2.3	130.1	-8.8	129.3
4/17/2020 12:41	-10.4	60	10.4	-1.6	130.2	-8.8	129.4
4/17/2020 12:42	-9.8	60	9.8	-1.6	130	-8.2	128.8
4/17/2020 12:43	-10.2	60	10.2	-1.5	130.6	-8.7	129
4/17/2020 12:44	-10	60	10	-1.8	129.7	-8.2	129.3
4/17/2020 12:45	-10	60	10	-1.8	129.6	-8.3	129.3
4/17/2020 12:46	-9.4	60	9.4	-1.8	129.9	-7.6	129.1
4/17/2020 12:47	-11.1	60	11.1	-2.6	129.6	-8.6	128.9
4/17/2020 12:48	-10.2	60	10.2	-2.5	130	-7.7	128.8
4/17/2020 12:49	-9.4	60	9.4	-1.7	130	-7.7	129.6
4/17/2020 12:50	-9.7	60	9.7	-1.7	129.3	-8	128.7
4/17/2020 12:51	-10.1	60	10.1	-2.2	129.4	-7.9	128.6
4/17/2020 12:52	-9.9	60	9.9	-2.1	129.5	-7.8	128.6
4/17/2020 12:53	-10.6	60	10.6	-2	129.3	-8.6	128.7
4/17/2020 12:54	-10.1	60	10.1	-1.5	129.5	-8.6	128.3
4/17/2020 12:55	-9.3	60	9.3	-1.5	129.5	-7.8	129
4/17/2020 12:56	-4.3	60	4.3	2.6	130.1	-6.9	129.5
4/17/2020 12:57	-10.1	60	10.1	-1.5	129.6	-8.6	128.7
4/17/2020 12:58	-9.1	60	9.1	-1.5	129.9	-7.6	129.2
4/17/2020 12:59	-9.1	60	9.1	-1.5	130.1	-7.6	129.2

BB. · April 2020

129.x volts is very close to the "good/bad" voltage cutoff for the grid... Typically around 132 / 264 volts, and minus a couple volts (margin of error).

Some GT inverters can have the voltage/frequency range "tweaked" a bit... Trying that, or telling the utility that you are very near "high line" voltage limit will get some action.

-Bill "in my humble opinion" B.

DickyDck · April 2020

thanks again BB! yeah the threshold in the inverter is set at 132, and I dont have historical since I just set up the custom logging, I "could" increase the high voltage cutout but I feel it was set at 132 for a reason. It could also be spiking above that so fast that my logging isn't catching it since Dave mentioned the XW are pretty quick at reading the measurements, possible sub-second processing my logging may or may not catch it right at that time of failure.

BB. · April 2020

Setting it 2 volts higher for a day is not going to hurt anything. See if the problems goes away, or at least, not as often.

Turns out there are no official high or low line voltage specifications for the USA.

Personally, I have used 212 to 264 (106 to 132) volts for USA. If you include Japan and 208 volt three phase power, then 90/180 vac would be low line.

Bill

DickyDck · April 2020

I shall give it a go, thanks!

DickyDck · April 2020

so this morning, everything was fine again until about 1200 EST, and this makes me believe more that it is something in utility line coming in. I heard one of my UPS fan going on, but I did not suffer a black/brown out and it wasn't running on battery, but when I looked on the lcd it was showing input of 130-133v which normally it is much lower in the 110v range. I jacked up the AC1 L1 voltage in my combox, but I suspect it alternates from L1 to L2, since I'm looking at my UPS now and it is within the 110s but the error is not clearing on my inverter, doing to 300 countdown, I watched it get down to a low of 273 but then triggered again and went back to 300. Since I know not of a way to adjust the L2 high limit on the inverter, I suspect it requires one of those super secret squirrel schneider black boxes to do it. This is just my theory of course, really no way of knowing how things are going to go, looks like I'll have to hire someone to come out and see, hopefully they have one of those programming boxes that can see the other hidden settings

BB. · April 2020

Sounds like it is time to call the utilty and complain about high line voltage.

If nothing else, you can probably get them to put a logger on your meter and get some more data.

Remember too, that when your XW enters "GT Mode", that the "wiring voltage drop", is actually a voltage increase from the solar energy being pumped back to the grid.

-Bill

DickyDck · April 2020

yeah I called and left them a message asking if they have done anything recent, I was excited this morning when I saw I was generating 6.5k on the lcd from about 8ish to noonish then I heard the inverter stop and the countdown timers showed up, was like mother &^%$#@$

BB. · April 2020

It can be a bad/weak neutral connection too... Check both L1 and L2 to neutral. Ideally, the voltage should always be 1/2 of L1-L2. If you get L1-N as 133 volts and L2-N as 110 volts (as an example) and L1-L2=244 VAC, then either a Neutral wiring problem, or possibly an "imbalance" between L1 and L2 120 VAC loads (such as big 120 VAC loads on L1 to Neutral, and little L2 to neutral loads) (Neutral current is always L1-L2=N current... If L1=L2 current, then Neutral current should be around zero amps).

The other voltage to check is Neutral to Ground (safety ground/ground rod/cold water pipe). Ideally the voltage should be very close to Zero VAC or at least less than a few volts.

If you see 10 volts or higher, then there many be some other wiring issues (have read of folks that L1 was bonded to ground rod/cold water pipe instead of Neutral--Home does work but obvious safety and leakage current issues).

At some large computer installations, have seen even a non-zero ground to ground voltage (would burn up Computer and Terminal grounding internal and data ground connections). One place, I even got a good 60 VAC or so shock from green wire safety ground to an irrigation water pipe (was in a facility with large salt water pumps for aquariums).

Probably not your issues, but I check everything at least once before I assume "it is OK".

-Bill

DickyDck · April 2020

Good copy, I'll have to find someone a little more comfortable poking and proding around 240v 200amp circuits, I had all the manels off and did a visual but I also doubt my harry homeowner electrical equipment is proper or adequate to be getting readings on ay of this stuff. I really do thank you for all the suggestions, wish I could go out there and test myself but I live alone and if I get fried to a connection they wont know I'm dead for days upon weeks

mcgivor · April 2020

Some random questions

Do you live in an area where there is industry, as Covid related shutdowns if applicable, could affect the voltage, less load on transformers. .

Was everything working prior the the first fault event, did you change anything hardware or software before the fault.?

Have you attempted to reduce the amount of solar input to see if that perhaps is causing the problem.?

Do you have an IR thermometer to check the temperature of all connections including the crimp connection at utility drop to the meter mast (this is a common failure point with aluminum ) ?

DickyDck · April 2020

Thanks!

I live in a very rural area so I don’t think there is much in the arena of businesses being closed that would cause enough of a surge to create a large enough increase to see what I’m seeing.

everything was working as expected for a very long time, close to a year if not longer. I was on here and you guys were helping me a lot with tweaks and such and once I got a proper sized battery bank it was smooth sailing. Nothing changed since that hardware or configuration.

yep, tried cutting down the PV input in various chunks all the way down to turning them all off at the combiner box. Would still trip the error, which at this time the only power of any sort is the grid.

i do have an IR laser thermometer! I have not however measured temperatures. It was in the 40 yesterday and rather cold but I’ll measure today if it happens again which I’m sure it will.

since this has started I see the error goes away at night. Right now at 745 the system is error free and working albeit not a great deal of sunlight yet.

mcgivor · April 2020

The questions were what I would ask if I was called time diagnose the system, the answer to reducing the PV to zero is interesting, in that it would indicate that the utility may be having issues, sometimes the branch transformer can have internal dielectric breakdown which causes intermittent spikes. Some utility companies gaurentee a voltage threshold, any deviations, especially those logged with accurate instrumentation, should be investigated by the utility on their side of the demarcation, your main disconnect.

Having the utility prove that the errors are not caused by them would be less expensive than paying for someone to investigate your system as a whole but may involve having someone with a quality power analyzer to monitor the grid input for a few days to record any sags or swells. Request a harmonic report especially in the lower range as these can create eddy current in the transformer resulting in manifestation of heat in the transformer, as harmonics cannot pass through a transformer. Armed with a report should result in a rapid response if there are any irregular results. These tests cannot be performed with equipment commonly used by the average electrican as the sampling rate is too slow and the parameters too limited.

This may result in discovering the problem is on your side of the demarcation, but at least you will be aware of the facts and know where to look. There may be one other possibility, because you are rural the voltage may have been increased to compensate for voltage drop, your feeding into it could push it over the top, but that wouldn't /shouldn't occur if the PV is disconnected.so, in summary it points towards the utility.

These comments are from an outside point of view, please understand.

DickyDck · April 2020

your assessment seems to match up with my logic based on marking off possible causes. With no PV input and it still tripping the error, it is the strong deciding factor that it is indeed something with the utility. I also pulled my old kill-a-watt plug thing, and keeping an eye on it. as of this moment it is hovering at 128.5v in my kitchen. Not sure what line that part of the house is running on, but at least it can tell me what the voltage is. And my UPS is also reading the same, where as last night it was hovering at the 130.x voltage area. PV is producing a good 5.5k now and feeding about 4k back into the grid, at 1045 EST. lazy sunday for me so I am keeping a window open for the combox page while watching netflix

Strange behavior with XW6848+ today

Comments

Categories