diy solar

diy solar

Sol-Ark TOU and Overcurrent Fault

robertk

New Member
Joined
Mar 22, 2023
Messages
13
Location
Missouri
I'm trying to figure out a problem with my Sol-Ark setup, but since this is my first post, I must first give a bit of background. We recently had a solar system installed by a professional installer. So it's not DIY, and I can't claim to be expert in anything about it. I understand the concepts of how it's done, but I didn't do it, and I didn't really watch the electrician either. So there may be questions that I'd have to hunt to find an answer to.

It's a 23+ KW system with two Sol-Ark 12K inverters and 20 kWh worth of Storz-Power batteries. This is a grid tie system. The grid feed comes in as two 200A feeds off a single meter outside, with each feed going to a separate breaker box in the house. The electricians added a third breaker box for critical loads and moved a few things over to it (fridge, freezer, some lights, etc). The system seems to work pretty good except for one thing.

Which brings me to the problem. I have the TOU set up to use batteries at night, then recharge from solar during the day, then send any excess back to the grid (solar sell). If they don't make it to full charge, I have it set to charge from the grid at 3 in the afternoon. On cloudy days, everything works just peachy. But on sunny days, it will throw an F18 AC Overcurrent Fault as soon as the batteries hit full charge. Every time. And if that weren't bad enough, it will reset itself and start back up and then immediately fault again. It will repeat this all day until the sun goes low enough, so I get basically no solar input during the day because it's constantly faulting and resetting.

BUT... if I turn TOU *off*, this doesn't happen. The batteries charge up, then it just sends the excess back to the grid just like it's supposed to and everything works fine.

The only way (so far) that I've found to actually make it work like I intend is turn TOU on in the evening, let it use batteries overnight, then in the morning, wait until the batteries are almost full (or let it fault once) then turn TOU off until evening again. Needless to say, this is not optimal.

The electricians who installed it have been out several times and have supposedly talked to Sol-Ark about it, but so far nothing they've done or suggested has helped at all. We've tried various different settings for TOU draw and SOC, but they don't matter. I even set the draw to zero during the day, figuring that would effectively turn it off, but nope -- same fault. If TOU is enabled and it's a sunny day, it will fault when it hits full charge, every time. And it does this even when we're not home, so there's very little power being used in the house at the time (maybe 1 KW or so), so I don't see how it could *really* be overcurrent, but that's the fault it throws. I have stood there and watched it do it. After a reset, I will see the solar input increasing, 1 KW, 2KW, 3 ... etc, and when it hits about 6KW, I hear a "click" like a relay or something, and it faults. I don't know if the click precedes the fault or is caused by it -- to the ear, it sounds like the click is first, then the fault, but it's almost simultaneous.

They've moved the CT sensors around, checked polarities, and so forth, and all seems good.

Since it doesn't happen if TOU is off, I can't help but think this is a software issue -- maybe a setting I have wrong?

Has anyone experienced this?

(And please don't tell me to call the electrician back out here -- I have been doing that, but the fault continues, so I'm trying to figure it out while gathering as much data as I can.)

TOU settings (on both inverters)
Screenshot of Safari (3-30-23, 9-30-52 AM).png
Screenshot of Safari (3-30-23, 9-31-15 AM).png
 
Do you have a comm cable between your batteries and the Sol-Ark?
 
I believe so. The RS485 ports have cables going from from one battery to the next and then the last battery also has another RS485 cable that heads over to the master inverter. I can't check that the other end is connected but I presume it is. I'll check it when I get back -- would one of those cables being bad (or not connected) cause this kind of issue?
 
I have a Sol-Ark without a battery, so I'm not an expert. However, I'm thinking the invertor is unaware of the true SOC of the battery. When charging from the generator the invertor should stop charging at 95% SOC. In Battery Setup make sure BMS Lithium Batt is checked and the two digit mode is correct. See your Sol-Ark install manual. The 15K manual states 01 for Storz Modbus Protocol. This RS485 cable isn't point to point. Either one came with the Storz battery for connection to Sol-Ark or your installer custom made one.
 
The easiest way to know if the batteries are communicating with the SolArk is to look at the main screen on the SolArk. Then tap on the Li-Batt Info block on lower right of screen. If it has numbers instead of a bunch of columns with zeros then it is safe to assume your battery is communication with the SolArk. The SOC number that is displayed, should be coming from the BMS. That could be a starting point in figuring out what is triggering the Overcurrent fault.
Also look at the Alerts block above the LI-Batt Info to see if the errors are showing up there.
 
Last edited:
So the “power” setting I set it and leave it. It specifies max power you want coming from you batteries. I have 90+ kWh of battery. I have mine set at 5000. I leave my battery SOC at 40 in all time slots except for the time period I want to top off my bank from the grid. Typically early AM at which time I pick my time to charge in slot 1 set SOC to 100% and go to bed. So it charges to 100% then we go back on battery so by morning I’m at 85-90% when the sun comes up. PV tops them back to 100% and within a couple of hours I’m grid selling the rest of my PV production. When sun goes down back on battery until the next early AM.

Since your closed communication I’m betting your BMS is reporting the error to the Sol Ark. I’m suspecting your charge settings are a little high so the BMS hits over voltage and reports it to the Sol Ark which then shuts down. There have been issues with some BMS’s and Sol Ark.

I chose to do open communication with mine and I have had no issues. I have on occasion seen one of my eight batteries flash a red overvoltage at the very end of my charge cycle when I had 1/2 the storage I now have. But it cleared itself within a couple of minutes and since I don’t have closed loop no harm no foul. Asked my battery manufacturer and he said nothing to worry about.

You might want to unhook your battery communication cable to the Sol Ark and then manually enter in your correct charge parameters and battery pack size into the Sol Ark and see how it runs. Uncheck the lithium battery box too.
 

Attachments

  • 9985CFD9-340B-4047-B2F1-9E1F5F4C6993.png
    9985CFD9-340B-4047-B2F1-9E1F5F4C6993.png
    195.1 KB · Views: 17
  • 0E9EE018-BD77-47D0-ADF5-4AC9BB1DD2CB.png
    0E9EE018-BD77-47D0-ADF5-4AC9BB1DD2CB.png
    207.6 KB · Views: 9
  • 026E194E-0EAE-4BD9-900A-79E37586E9CB.png
    026E194E-0EAE-4BD9-900A-79E37586E9CB.png
    164.1 KB · Views: 8
  • E5936E87-65B1-4F17-9F11-765AEC30EF3D.png
    E5936E87-65B1-4F17-9F11-765AEC30EF3D.png
    166.2 KB · Views: 8
  • 4E34FAC4-FC5A-4D40-9D63-467DAF43E4ED.png
    4E34FAC4-FC5A-4D40-9D63-467DAF43E4ED.png
    164.5 KB · Views: 11
  • A3539925-AAD3-4A0A-9AE1-4D28AD1783B0.png
    A3539925-AAD3-4A0A-9AE1-4D28AD1783B0.png
    211 KB · Views: 17
  • 82F07920-7749-48AC-B5E5-AB91833450AF.jpeg
    82F07920-7749-48AC-B5E5-AB91833450AF.jpeg
    1 MB · Views: 20
  • A5A11855-D45C-49ED-A7E0-30112DED414F.jpeg
    A5A11855-D45C-49ED-A7E0-30112DED414F.jpeg
    1.1 MB · Views: 21
  • 4E330DD4-F9AC-4370-AFA7-67EEA9874606.png
    4E330DD4-F9AC-4370-AFA7-67EEA9874606.png
    374.2 KB · Views: 20
So the “power” setting I set it and leave it. It specifies max power you want coming from you batteries. I have 90+ kWh of battery. I have mine set at 5000. I leave my battery SOC at 40 in all time slots except for the time period I want to top off my bank from the grid. Typically early AM at which time I pick my time to charge in slot 1 set SOC to 100% and go to bed. So it charges to 100% then we go back on battery so by morning I’m at 85-90% when the sun comes up. PV tops them back to 100% and within a couple of hours I’m grid selling the rest of my PV production. When sun goes down back on battery until the next early AM.

Since your closed communication I’m betting your BMS is reporting the error to the Sol Ark. I’m suspecting your charge settings are a little high so the BMS hits over voltage and reports it to the Sol Ark which then shuts down. There have been issues with some BMS’s and Sol Ark.

I chose to do open communication with mine and I have had no issues. I have on occasion seen one of my eight batteries flash a red overvoltage at the very end of my charge cycle when I had 1/2 the storage I now have. But it cleared itself within a couple of minutes and since I don’t have closed loop no harm no foul. Asked my battery manufacturer and he said nothing to worry about.

You might want to unhook your battery communication cable to the Sol Ark and then manually enter in your correct charge parameters and battery pack size into the Sol Ark and see how it runs. Uncheck the lithium battery box too.
If it runs fine ditch the closed loop comms and buy yourself a Solar Assistant and/or a Victron smart shunt.
 
Since your closed communication I’m betting your BMS is reporting the error to the Sol Ark. I’m suspecting your charge settings are a little high so the BMS hits over voltage and reports it to the SolArk.
That is where i would start by lowering charge settings, however as I found out with my SolArk closed communications, those settings may be coming from the BMS. If that is the case, then next step would be to do as jfpetesn suggests.
You might want to unhook your battery communication cable to the Sol Ark and then manually enter in your correct charge parameters and battery pack size into the Sol Ark and see how it runs. Uncheck the lithium battery box too.
 
That is where i would start by lowering charge settings, however as I found out with my SolArk closed communications, those settings may be coming from the BMS. If that is the case, then next step would be to do as jfpetesn suggests.
Yes with closed comms the BMS sets the charge settings. I have been working with the Trophy battery people getting them to properly communicate. They Jack up my charge setting when I try closed comms. Also the more I play around with it I have decided I prefer open comms with the Solar Assistant. I get more detailed info and better battery management which consequently is better for troubleshooting. Honestly if I could figure out how to get my Victron Shunt SOC to the Sol Ark I’d be 100% happy. As it stands now I’m about 98% happy.
 
Honestly if I could figure out how to get my Victron Shunt SOC to the Sol Ark I’d be 100% happy.
My Orion BMS outputs a Victron protocol so the Victoron Shunt should work if it communicates. You may need another piece of Victron equipment for communications.
 
My Orion BMS outputs a Victron protocol so the Victoron Shunt should work if it communicates. You may need another piece of Victron equipment for communications.
I know I could use it with the Solar Assistant not sure about the Sol Ark but if I was given a choice of closed comms I don’t think I’d use it. I like my Solar Assistant and I’d hate to loose it. With closed comms I think I’d have to choose either or.
 
.I did not give up either. I have both through the use of a splitter that duplicates the pinouts on the SolArk so that both my BMS and SolarAssistant can comunicate on those two pins. I got it on Amazon and it cost less than $2

RJ45 Splitter

I’m going to give that a try. If I can feed the Sol Ark my SOC everything would be perfect. Too bad the solar assistant doesn’t feed data to the Sol Ark now that would be the perfect solution.
 
Since your closed communication I’m betting your BMS is reporting the error to the Sol Ark. I’m suspecting your charge settings are a little high so the BMS hits over voltage and reports it to the Sol Ark which then shuts down. There have been issues with some BMS’s and Sol Ark.
That sounds reasonable, given that it always happens when the battery hits full charge. It seems weird that it's an AC overcurrent fault rather than DC, though, if it's on the battery side.

I checked the cabling this evening, and it does indeed have a data cable from the battery to the master inverter, and then another data cable from the master inverter to the slave inverter. The battery icon on the inverter gives me a page that certainly appears to be populated, so I'm guessing the cables are good. (These were taken just about sundown, BTW.)

master_battery.jpg slave_battery.jpg

It is set to "BMS Lith Batt" and type "01". It has this for the charge settings:

battery_settings.jpg

I saw that and thought maybe the voltage should be lower, since the batteries are presently at 95% and showing 53V. But the spec sheet says the max is 56.8 so I guess it's OK? (Spec sheet here)

So I will turn the battery SOC down on the TOU settings during the day to 40% and see what happens tomorrow. It's supposed to be sunny all day.
 
Well I am hopeful, but I know better than to call anything definitive after just one day. But today I had the SOC at 40% for the daytime slot. It was bright and sunny all day, and the inverters did not fault! It’s supposed to be sunny again tomorrow.
Let’s hope it sets a trend. I haven’t changed the float voltage yet (trying to only change one variable at a time…)
 
Another bright sunny day and … no faults! So I guess turning SOC down low was the trick. I don’t understand why — I mean I tried it at various levels from 100% down to 80% over the course of several days and none of those mattered, it would always fault. But at 40%, at least so far, it’s working exactly like I want it to.

HUGE thanks for the help!

I have a feeling I will be learning a lot more about all this as time goes on. I can already see adding more storage, and a standby generator wouldn’t be bad either. But for now I am hoping to see a solid month of data with it running like it’s supposed to. The electric company I use does their billing cycle on the 6th, so if it continues to run without faulting, it’s good timing.

Should I go ahead and lower the float voltage to 54? Assuming it’s working as is, what’s the advantage in doing that? Does that affect battery longevity?
 
Float at 3.45V per cell. Check if you have 15 or 16 cells in your batteries.
Then multiply that voltage above with the number of cells.
 
I can’t seem to find any documentation of how many cells are in each battery. But either way, 15 or 16, that still comes out less than the 56v it’s currently set to. What happens if it’s set too low? Or too high (as it apparently is now)?
 
I can’t seem to find any documentation of how many cells are in each battery. But either way, 15 or 16, that still comes out less than the 56v it’s currently set to. What happens if it’s set too low? Or too high (as it apparently is now)?
When I set mine up I called my manufacturer for their recommendation on charge settings. So with my Sol Ark15 I always leave them at 40 for all time slots with grid charge boxes all checked for all time slots. When I want to charge from grid typically in the middle of the night, I set a start time in the first TOU slot (say 0100) and set my battery SOC to 100%. Then I pick an end time (say 0300) in the second TOU slot and leave the battery SOC at 40%. It starts charging at 0100 to 100% and ends charging at 0300. Works like a charm. If I want no charging then I leave them all at 40%. You don’t have to charge to 100 but I find my SOC measurements on the Sol Ark and my Victron shunt stay way more accurate if I get a “knee” in my charge cycle a few time a week and setting to 100% charge does this very well (see photo) and it also helps keep all my cells balanced.
 

Attachments

  • 34602B83-22C2-4D9A-B79F-DD10C0EA052F.png
    34602B83-22C2-4D9A-B79F-DD10C0EA052F.png
    1,020.8 KB · Views: 8
  • 5E2FD52B-70B2-477B-92EC-85FE987669F2.png
    5E2FD52B-70B2-477B-92EC-85FE987669F2.png
    211 KB · Views: 8
Last edited by a moderator:
Looks like I may have declared victory prematurely. I got the overcurrent fault again this morning. It wasn’t even full sun at the time, and we weren’t home. Power draw to the house couldn’t have been much and the PV was only about 13 kw when the batteries got to full charge. (The last couple of days they were closer to 20 kw and did not fault). So I guess I’ll try float voltage next.
 
LifePO4 should NEVER be charged/absorb with more than MAX 3.65V per cell.
You can damage the batteries.
Absorb should be max 3.65V per cell or lower, I prefer 3.55V to enlong the lifetime of the cells.
Float should be lower as I wrote above. 3.45V float is a good number per cell.
 
Well I called the battery manufacturer (Storz-Power) to talk about these issues, and the person I spoke to said that if the battery is in closed loop communication, the battery itself will tell the inverter what voltages to use, so it won't do my any good to set it. He said they could log into the inverter remotely and check things if I shared my plant in PV Pro, but apparently I can't do that because the electrician set it up and while I can change settings all day long, I don't seem to have rights to share the plant. Sigh. So, the electrician will be back out here next week and I'll see if we can make any more headway.

At this point, the settings I've got (40% SOC except in the charging slots) seem to work most of the time. Prior to that, it would throw AC overcurrent once the battery was at full charge on a sunny day, every time. But it didn't fault on cloudy days. This past week, running on the 40% setting, it did not fault on sunny days, but did fault on most of the cloudy days, but not all. The only part of the pattern that remains consistent throughout is that the fault happens when the battery reaches full charge.

Hopefully the electrician will have better luck this time....
 
Id unplug that battery communication cable and enter in lower charge values and see if that helps
 
I tried changing the float value without unplugging the cable and even though it let me change it and saved the changes, they reverted shortly thereafter. So I guess that at least verifies that the closed loop part is working. :O

The electricians will be back tomorrow to take another look.

In the meantime, I had a thought. Since the fault only happens when the batteries hit full charge (so therefore the PV is switched from DC side to AC side to sell back to grid), and it seems to mostly (but not always) happen when generating near capacity... could this be a capacity problem? I mean, there are two 12K inverters. They are rated for 9 kw continuous and 12 KW max, so a combined 18 kw continuous. If the panels are putting out more than that continuously, would that make it fault? (The panels are rated at max of 23.8 kw, but the most I've seen so far is about 19 kw on a sunny day).

Though I guess if it were a capacity problem, it would happen whether or not TOU was on, and it just doesn't happen if TOU is off.
 

diy solar

diy solar
Back
Top