diy solar

diy solar

Struggling with battery capacity

They do sell 12v balancers for AGM batteries. Hopefully they’ll set you up for the long haul. Sounds like a nice system you have there.

 
Well maybe we're not done.

I swapped out the bad battery of one string and I'm just running off that one string of four for now.

I measured the voltages of the four batteries while charging (lots of sun) and the first in the string is at 16.3V! The others at about 13.7V. What the heck?

I disconnected the panels and they all are around 13V.

PXL_20210903_184041814.jpg

I'm afraid to hook up the panels again - isn't 16V way too high? what could cause this?
 
Can you share a picture of the battery terminal connections and wiring?
 
You really should label your batteries so we can track them from the pics. It may be too late for you to take notes on all those previous readings but it will help you going forward so you can see if any start going south and proactively do what you can to keep your battery bank happy.
 
You really should label your batteries so we can track them from the pics. It may be too late for you to take notes on all those previous readings but it will help you going forward so you can see if any start going south and proactively do what you can to keep your battery bank happy.
Oh they are labled, A thru H. I forgot to write the letters on the above drawing of four, but my earlier posted data and graphs have the letters.
 
PXL_20210903_164218132.jpg
Well the batteries were just swapped out. Two new ones replaced A and F. Pretty good customer service. The owner didn't come by, just two of his employees and since I had turned off the panels I couldn't really show them the 16V charging measurement. They didn't seem too interested, they just said everything looks good now and left.

I will continue to monitor individual battery voltage from now on.
 
Here's a photo of my setup. Two strings in parallel just like my sketch, one above the other. The negative from the controller goes to the top left battery negative, and the positive to the bottom right battery positive. The breaker box to the left of the Growatt is 240V in from my generator (when needed), 240V comes out of the inverter to the breaker panel on the right.

The batteries on top of the plywood are not part of the system, it just seemed like a good place to charge things like tools, motorcycle batteries, car battery, etc. They are plugged into a smart outlet that only comes on 6 hours during mid-day when we have lots of sun to keep them topped up. One of the motorcycle batteries actually runs our 12V water pump that keeps the blue pressure tank full.

PXL_20210903_201949629.jpg
 
1 ton mini split AC keeps the room cool during the roasting summer months. The exterior of the block walls are covered in 2" foam insulation, which will eventually be covered in stucco.

PXL_20210903_202100485.jpg
 
Hi. So I am returning to this thread because we are still having battery problems.

Summary:
I started this thread trying to understand why I wasn't getting even close to the stated capacity of my batteries after just a few months of installing them and they had been working fine. After much help from this forum, we found that two of the eight batteries were very low in voltage in the morning and the others were up over 12V. My battery vendor immediately replaced the two bad batteries and everything was good, batteries staying up over 50 volts all night long...for a couple months...

Then, a couple months later, all of a sudden we lost power overnight. I check the batteries and again there are another two at 6 or 7 volts (not the two he had replaced), and the other six batteries are over 12.5 volts. I contacted the battery vendor again, and he replaced those two batteries, free of charge. He tells me there must be something wrong with my system, because in the 4 years he has been installing systems with these batteries he as never had one fail, and he also says the "bad" batteries he replaced are now installed in his system and doing just fine.

I don't know how my system can be causing two batteries to fail - doesn't my controller just see one large 48V battery? Each time there has been one bad battery in each string of four.

Some here had recommended installing an equalizer so the batteries charge and discharge equally.

I went ahead and installed two of these from Watt247, one in each string of four batteries:
https://watts247.com/product/eq-48-...-batteries-or-any-x-3-2v-batteries-in-series/

PXL_20220131_165717686.jpg

I like the fact they display the voltage of each battery and that they do keep them equally charged.

Everything has been good for the last few months, even with some overcast skies the batteries have easily made it through the night.

Then just last week, we started losing power again! Now that I have the equalizers I can see that all of the batteries are equally down to 11.5V, causing the inverter to turn off at the <46V set point.

I haven't done it yet because it is a pain, but should I remove the equalizers and see if two of the batteries are down lower than the rest again...?

The guy who sells the batteries is surely going to blame my Growatt unit again. But how could my controller keep killing batteries?

Thanks!!
 
After some discussions with Watts24/7, they thought the 58V bulk and 56V float (that the battery vendor recommended) is too high for our AGM batteries and they are not actually charging to capacity with that high of an input.

I reset the settings on my Growatt 6000 to default AGM settings, and now we are getting through the night just fine, barely dropping to 80% on some mornings. So the moral here is don't overcharge AGMs - they like a slower, lower voltage charge.

EDIT: After about a month later, we are again not making it through the night. Only using 30 to 90W (depending on refrigerator state) all night long. I am getting nowhere near the 24 kWh I should have.
 
Last edited:
the grow watt only puts out 80a max no matter how big the solar panels. this is only 4500 watts going into the battery. this is supposed to be going into a 24kw battery system, so maybe a total of 9kw for 2 hours. plus some 10amp float charging for 4.5 hours (540w x 4.5 hours = 2430w), which isn't that much. total charge going in is 11430w. the efficiently of charging batteries is taken away from that...i.e. charge 10 watts, but after charging you only managed to get 8 w into the battery. 24kw -11.4kw = 13kw wasn't being charged into the batteries, minus efficiency of the battery accepting charge. I think you need to increase the number of hours you charge before going to float to 4 hours.
 
It would be nice if you had a way to accurately measure the current in and out of the pack.

How old are the batteries? Kinda sounds like they are failing .... and if they keep being discharged to the point where the inverter shuts down they are being damaged even more.
I'm thinking the same, if he was dropping down to 75% discharge, that's torture on batteries & won't last long. I never go past 40% drain on mine.
 
I have 5 banks of 48v strings. 1 thing I discovered is if 1 or 2 batts have lower charging volts than the others, it will force overcharge the rest in that string causing them to fail faster so I frequently monitor them during charging & move them around to make strings of the batts with the closest charging voltage. Your chart shows a trend towards that's going on. I haven't tried this but I've seen on amazon a battery balancer that's suppose to keep them equal, sorta like a bms but for sla's. That might be a good investment worth trying out.
 
the grow watt only puts out 80a max no matter how big the solar panels. this is only 4500 watts going into the battery. this is supposed to be going into a 24kw battery system, so maybe a total of 9kw for 2 hours. plus some 10amp float charging for 4.5 hours (540w x 4.5 hours = 2430w), which isn't that much. total charge going in is 11430w. the efficiently of charging batteries is taken away from that...i.e. charge 10 watts, but after charging you only managed to get 8 w into the battery. 24kw -11.4kw = 13kw wasn't being charged into the batteries, minus efficiency of the battery accepting charge. I think you need to increase the number of hours you charge before going to float to 4 hours.
Thanks, I like the approach. I am a little confused by some of your numbers. I see 9kW x 2 hrs = 18 kWh. Then 540W x 4.5 hrs = 2.43 kWh. So a total charge of 20.43 kWh. And then a little less for efficiency. Which wouldn't be so bad, but still could use a little longer before float I agree. I don't know if the Growatt allows me to change the point at which it goes to float.
 
I have 5 banks of 48v strings. 1 thing I discovered is if 1 or 2 batts have lower charging volts than the others, it will force overcharge the rest in that string causing them to fail faster so I frequently monitor them during charging & move them around to make strings of the batts with the closest charging voltage. Your chart shows a trend towards that's going on. I haven't tried this but I've seen on amazon a battery balancer that's suppose to keep them equal, sorta like a bms but for sla's. That might be a good investment worth trying out.
I do have a battery balancer - there are photos of it above. I thought that would solve all my problems, but no :(
 
So here is a chart of our charge and use from yesterday. From this data I am not getting near 24 kWh of charge. The spikes in the afternoon are when using a vacuum - solar power increases to supply the load.

Screenshot 2022-03-14 081724.png
 
The batteries ramp up to 14.4 V or so in just a couple hours in the morning. I don't understand how I have 24 kWh of batteries drained to 42V, and they charge up to full capacity in just 3 hours, and using max 1500W of solar? (I have 4500W of panels available) Something is not right. I tend to believe the batteries are not actually getting fully charged to capacity, but they are reading 14.4V which should be full charge, right?
 
Last edited:
24kWh of AGM, discharged to 42V, recharged to 57.6V in 3 hours, 4.5kWh?

14.4V per 12V pack charged, 10.5V per pack discharged?

For my AGM, 14.4V is absorption voltage, then hold about 2 hours.
Preferred charge rate is at least 0.2C, which would be 4800W; your 1500W charge rate is 1/3 of that.
The technical manual suggests an alternate absorption profile if low charge current. (Not necessarily a programmable option in commercial chargers.)

I have a 12 year old 100 Ah 48V pack. I load tested it with 600W to 70% DoD and estimated 40% of original capacity.
My system has a 2 year old 400 Ah pack, have only cycled a couple times.



Confirm charge voltage at battery terminals, and voltage after sitting no-load for a couple hours. See what SoC is indicated.
Confirm discharged voltage at battery terminals, and voltage after sitting no-load for a couple hours.
High resistance path would cause voltage sag.
During discharge, internal resistance causes further drop in voltage (Peukert effect, which may include other factors.)
It could be your capacity is degraded. It could be the batteries just aren't achieving full charge; make sure battery terminal is reaching 14.4V or whatever and remaining their for intended absorption time.

Some charge controllers will switch from absorption to float too soon because loads or source have changes that confuse them and cause change of state.
 
I just spoke to Ian from Watts247, where I bought the Growatt and balancers. I have to say his support and response has been excellent.

He looked at my data and now thinks that my charge current is too high. I had it set at the default 80A. He said I have a common problem of having way to much solar power available for not a lot of batteries, and I'm "cooking" the batteries by charging them too quickly. His analogy was wetting a sponge with a fire hose.

So I've dropped the charge current to 40A and will see how that goes.
 
Last edited:
I have read through this entire scenario and I believe the only way to get to the bottom of the situation is to monitor the current into the batteries and out of the batteries. They have a number of devices on the market that do just that. With my system the Overkill BMS does a good job of keeping track of the cell voltages (Lifepo4) and also how many amps that the Cell packs have taken in during the day. It also keeps track of the current draw during the night and how many amps are left in the cell pack in the morning. In my opinion, if you are not keeping track of the current going into your batteries and the current that is coming out of your batteries then you will never be able to solve this problem. With all batteries of the different chemistries it is very important to keep the batteries happy. I am not sure of the specs on the AGM batteries that you have but you can look them up and make sure that they are happy. Charge current is important, Discharge current. I believe that you need to occasionally take the batteries to 14.4 volts for a period of time and then drop them down to an absorb voltage. I would become very familiar with the actual parameters if it were my system. When I purchased my Lifepo4 cells, I read everything about the voltages and currents that the cells were to receive. If you exceed the voltage or current then damage will occur. I am not a fan of Growatt but I will not criticize those who have elected to go that route. The All in one concept sounds good on paper, but I suspect that it has some issues which I do not want to deal with. I like the concept of if something goes wrong then replace one component rather than the system. Like if the Charge controller quits working then replace the Charge controller. If the Inverter quits working then replace the Inverter. With the Growatt system that you are dealing with who knows how much power that thing is consuming during the night. For all you know it could be sapping all the power out of your batteries and you would not know. Again enough of the philosophy, in my opinion the only way to determine what the problem is would be to measure the current into the batteries and the current out of the batteries. If the batteries are for example 100 amp batteries, then you should be able to put 100 amps into them and then take 100 amps out of them in a perfect world.
 
Back
Top