diy solar

diy solar

Battery Degradation and System Shut Downs

Mendoza

New Member
Joined
Aug 27, 2020
Messages
85
Location
Chile
3.25kWP Risen panels
Batteries: 4.8KWH Utracell sealed Lead acid UCG 100-12
Inverter: Axpert MKS II 5kW

Main purpose of this thread will be to discuss how I can determine how much my batteries have degraded as well as discuss why the system occasionally cuts out and needs rebooting. If anyone is able to offer advice or comments that would be appreciated.

After getting solar in June 2020, I posted a thread here https://diysolarforum.com/threads/batteries-don´t-charge-beyond-80.11448/ in Aug 2020 and got help thanks to everyone to improve the battery performance.

For a long time after that I had excellent performance. I am in the southern hemisphere and from mid December 2020 to early April 2021 I used almost no grid electricity at all. The system was still running off battery power when I went to bed, and quite often the batteries lasted through the night as well. In that period, our energy use was about 95% solar. The best month was January when I managed 98% solar, using 5.6kWH of grid electricity for the entire month or 0.18KWH per day average. In June 2021 (the darkest month) I used 70% solar and 30% grid, still a pretty decent result and at that stage I would have guessed negligible battery degradation.

I have used the batteries on about 90% of days. 10% of days I keep them in reserve as a backup source (these are days of possible power cuts, mainly rainy days). Also I have typically set the batteries to 46V minimum (i.e. inverter is programmed to switch to battery power when voltage falls to 46.0 or 45.9). This minimum is often reached. I believe this all means that the depth of discharge is probably maximum about 60%, and average about 50%.

The battery has already been through about 400+ cycles of average DOD 50%. According to the graph provided by Ultracell for this battery, the life ought to be about 1000 cycles for that sort of DOD. Given this, what sort of degradation would you expect to see after 400 cycles? It's not clear what they mean by 1000 cycles life. Does that mean after 1000 cycles the performance is at what 50%, 0%? I'm not sure how they determine end of life.

In late July, August, September and now in October I had disappointing results in terms of grid electricity spend being slightly higher than the amounts for the same months in 2020. We haven't had any periods of dark cloudy weather, so logically we are either using more electricity or the performance of the battery or the system is somehow worsened. I think the second is probably at least partly true. The peak voltages seen after sunset when the battery is under slight load seem lower than a year ago, i.e. sometimes closer to 49V than 50V, however I'll have to check a few times more to rule out that this wasn't coincidence on a few days.

The battery didn't seem to show any degradation in performance for about 12 months, then seemed to decline noticeably in a matter of weeks. Is that consistent with normal degradation, or does it suggest instead an issue to be resolved to restore the battery to prior performance levels? Then again, sudden degradation is just my perception really, not based on measurement. Maybe the decline was more gradual and other factors like weather and variation in amount of electricity used meant I missed it at first.

In September and October we've also had the system cut out - inverter showing error messages with the power out. This has happened 4 or 5 times. We've had power cuts before, but usually for a reason, such as we exceeded the maximum wattage that the inverter can handle. But these latest power cuts happened in the middle of the night for no apparent reason under low load. One at least one occasion the error message was 58 "Output voltage is too low" and on at least one other occasion the message was 8 "Bus voltage is too high". The other times I can't remember.

The shutdowns/cuts/power outages were solved by turning the system off and back on again. In one case (during the night), it immediately cut out again a few minutes after I'd rebooted it. So then I tried turning off everything and then turning back on everything except the battery. Then it worked. And I turned the battery back on the next day.

As I mentioned, I have the system set to a minimum voltage of 46V and then once that is reached it switches to grid. However, it isn't as simple as just using battery until it reaches a certain level and switching over once to grid for the rest of the night once 46V is reached. Sometimes, as the load and consequently the voltage varies, it switches back and forth between battery and grid. I have a hunch that the power cuts may happen when the inverter makes these switches but that is just a guess.

So we have two questions of possible reduced performance/degradation and the system cutting out. And the question as to whether these two issues are linked.

My friend who installed the system suggests that noticeable battery degradation after 1 year 4 months may be normal given that I use the battery constantly rather than only have it for backup power. Regarding the power cuts, he suggested that now the battery is a bit worn, it may be a benefit to increase the minimum voltage (i.e. voltage at which the system switches from battery to grid) from 46V to 48V not just to prolong battery life but he thinks this could reduce the power cuts. That it may be harder for the now more aged battery to deal with the low voltage (compared to new). Does that seem fair?

If anyone is able to reply, keep in mind I am a novice, this was my first system, and I didn't install it myself, so I may not understand even fairly basic technical terms.

I am running the house off grid this evening so I can take an amps and voltage reading with the battery fully charged and at rest. I am not sure if these will help but I will try to post these later incase it does.
 
Last edited:
Do you have a proper amp hour meter on the batteries so you know exactly the charge level?
 
So in October 2020 I measured the amps and voltage with meters with the battery at rest - disconnected from PV and loads - and it was 0.4A and 53.05V and it was stable. This was with a high state of charge.

I did the same today and it was 51V and 1.3A. I am not sure if that tells us anything. This was probably with the battery as close to fully charged as it can.

What I did in both cases was fully charge the battery and then, in the late afternoon, set the inverter to power the house with only utility/grid power and then later come back after a couple of hours and start taking measurements. Each time the measurements were taken the battery was at rest, so not being charged or providing charge and not having done so for hours.

The voltage reading is with a Fluke Multimeter. The Amp reading is with a UNIT UT210D.

(Also, each time I checked the voltage today it was lower. At 8.45pm it was 51.1V. At 9pm it was 51.0V. At 9.30pm it was 50.9V. At 10pm it was 50.8V. The Amp reading was always in a similar range of 1.29A and 1.33A across several hours.)

EDIT: Just changed the settings and now the battery is providing load. It stabilized at 50.0V at 150W load. A year ago, it would have showed 51.1-51.5V at this time of year for that level of load at the end of a sunny day. So the voltage at full charge seems to be lower. Is it fair to say that this could be indicative of capacity loss / degradation of battery over time?
 
Last edited:
So I have not read through all your other post, but have you installed an amp hour meter on the battery?
 
Yes.

I measured the amps on the battery with a meter, that is UNIT UT210D. If you do an image search with a search engine you can see a picture of it.

I think that is an amp hour meter, so I think the answer to your question is yes.
 
Yes.

I measured the amps on the battery with a meter, that is UNIT UT210D. If you do an image search with a search engine you can see a picture of it.

I think that is an amp hour meter, so I think the answer to your question is yes.
That is not an amp *hour* meter. That is am amps-right-now meter. It does not measure over time, so it can't tell you how may amps went into the battery, and how many came out over a 24 (or longer) hour period.

This is an amp hour meter.

If you check your other thread, several other people also said to get an amp hour meter.
 
Last edited:
The AiLi monitor is a good option. Get the 350 amp shunt version. Price is around $40.

 
Thanks for your help.

Thanks for your messages. I can´t seem to find that (or similar) product here in Chile yet. I supposed I could get it shipped internationally, but that adds to the cost and hassle, especially when you get a call from the airport saying you have to go to the airport to get the product through customs.

The system shut downs have stopped happening (none this month), but I expect they´ll happen again, but as long as they are very occasional and can be fixed with a reboot not sure if it´s worth the hassle of fixing it. I googled it and there are some possible solutions out there and it might be something to do with a circuit board in the inverter, but it looks difficult and one person at least reported still getting the error after fixing the circuit board.

EDIT January 2022: The battery is now doing really badly and barely works at all, at the moment I just have it disconnected which seems to stop the system shut downs. At the moment I just have direct use from solar panels, and grid.
 
Last edited:
Back
Top