diy solar

diy solar

Have i drained my bank too far?

off.the.grid

New Member
Joined
Dec 26, 2021
Messages
121
I was doing a bit of a stress test on my battery bank yesterday after setting up my new install to test its limits.
I had boiled the kettle and used the toaster a few times during the day and put the dishwasher on that night.
My bank has 540Ah capacity at 24V which gives me close to 6.5kwh of usable capacity. My panels can supply up to 1600W.
Anyway, that evening when I had the dishwasher on, near the end of the cycle i noticed the battery voltage was around 23V, I know that this is OK under load.
The inverters cut off voltage is set to 21.6V under load, so I was not concerned over the system over draining going by those figures, however, it must have dropped off quite fast, as in about another 20 mins, the inverter shut off.
The battery voltage bounced back quite close to 24V after the load was removed, so that gave me some reassurance. I feel that it was probably just more current than what the battery bank could cope with as it was getting low.
I have no idea how much I had drained the batteries, but my maths goes that I would have only used about 2.5KWh of power that day that I could account for.
Im not sure how much charge the bank had that day, but it was showing what appeared to be a full charge at normal float level voltages in the morning.
What I dont know is how long does it typically take to charge a bank if its down to say 70 - 80%? I have no idea who low it even was the night prior, and measuring the capacity is not always accurate either. I was a bit concerned how my epever controller said it was at 30% capacity remaining, but things were under a small load, but nonetheless I fired up the genset for an hour before bed.
How does the boost phase work with these anyway? Its set to charge for 2 hours during the boost phase in the morning and stay at float voltage for the rest of the day, but those 2 hours might not be necessarily enough to top the battery right up to 100% all of the time?
I have no idea how to tell the capacity while its in the float range, i guess the only thing to look out for is the reading on how much current the batteries are being fed?
If i switched on the kettle during the day, you can see a good amount of current feeding into the system from the panels, so I know that it shouldnt have been using much of my batteries when the sun is up.
Am I better getting a 48V inverter to reduce current demand from the battery bank?
 
If these are LifeP04 or LTO batteries, then no problem. If they are Lead Acid batteries then yeah, you've killed them
 
If these are LifeP04 or LTO batteries, then no problem. If they are Lead Acid batteries then yeah, you've killed them
They are AGM batteries. Why do you think I've killed them? The discharge settings was set to the default of 10.8V per battery under load for lead batteries. I was more concerned if I had discharged them down below 50% or not, im starting to think this is not the case, dince the voltage popped up rather fast after shutting down. It's probably better not to drain them down too far on a regular occurrence but going by those voltages, I just got caught out by surprise, as I thought there was more capacity left in them.
Would it be better if I set the inverter to shut off at a bit higher voltage level to be on the safe side?
 
I was doing a bit of a stress test on my battery bank yesterday after setting up my new install to test its limits.
I had boiled the kettle and used the toaster a few times during the day and put the dishwasher on that night.
My bank has 540Ah capacity at 24V which gives me close to 6.5kwh of usable capacity. My panels can supply up to 1600W.
Anyway, that evening when I had the dishwasher on, near the end of the cycle i noticed the battery voltage was around 23V, I know that this is OK under load.
The inverters cut off voltage is set to 21.6V under load, so I was not concerned over the system over draining going by those figures, however, it must have dropped off quite fast, as in about another 20 mins, the inverter shut off.
The battery voltage bounced back quite close to 24V after the load was removed, so that gave me some reassurance. I feel that it was probably just more current than what the battery bank could cope with as it was getting low.
I have no idea how much I had drained the batteries, but my maths goes that I would have only used about 2.5KWh of power that day that I could account for.
Im not sure how much charge the bank had that day, but it was showing what appeared to be a full charge at normal float level voltages in the morning.
What I dont know is how long does it typically take to charge a bank if its down to say 70 - 80%? I have no idea who low it even was the night prior, and measuring the capacity is not always accurate either. I was a bit concerned how my epever controller said it was at 30% capacity remaining, but things were under a small load, but nonetheless I fired up the genset for an hour before bed.
How does the boost phase work with these anyway? Its set to charge for 2 hours during the boost phase in the morning and stay at float voltage for the rest of the day, but those 2 hours might not be necessarily enough to top the battery right up to 100% all of the time?
I have no idea how to tell the capacity while its in the float range, i guess the only thing to look out for is the reading on how much current the batteries are being fed?
If i switched on the kettle during the day, you can see a good amount of current feeding into the system from the panels, so I know that it shouldnt have been using much of my batteries when the sun is up.
Am I better getting a 48V inverter to reduce current demand from the battery bank?
Can't really answer your questions but I don't know how you expect 1600W if solar to charge 6.5kWh of batteries in 2 hours.
If the batteries aren't charging fully but the your solar produce more power when the kettle is on, then it sounds like something is up with the charge settings and your battery isn't charging properly.
 
If these are LifeP04 or LTO batteries, then no problem. If they are Lead Acid batteries then yeah, you've killed them
Where did you get the idea that he killed the batteries?

lead-acid bouncing back to 12 volts (no-load voltage) still has 30-40% charge left. (Maybe even bit more as it takes some time for batteries to settle to no-load voltage)
 
Can't really answer your questions but I don't know how you expect 1600W if solar to charge 6.5kWh of batteries in 2 hours.
If the batteries aren't charging fully but the your solar produce more power when the kettle is on, then it sounds like something is up with the charge settings and your battery isn't charging properly.
The controller does a boost phase for 2 hours each morning. I know that it doesn't feed typically anymore than 10 amps max into the bank that I've seen at 29.2V. I'm assuming this is because the batteries were largely fully charged.
I can change the setting to do a boost charge for longer if I want, but 2 hours appears typical on most installs. I need to check again first thing in the morning to see if it pumps in more juice than this, but the panels are able to supply enough power.
I thought this was normal to draw more power from the panels when under load? Wouldn't this indicate that the batteries were fully charged? It was around an amp or two prior at float levels.
I dont know how many hours at float charge would be needed to fully charge.
This is what confuses me about these controllers. The profiles charge the batteries for 2 hours once the sun comes up at boost voltage and then sits at float and can see the current drop away slowly over the day. But if the battery is more empty on any given day, would 2 hours be enough? Or is this just a standard thing regardless?
 
The controller does a boost phase for 2 hours each morning. I know that it doesn't feed typically anymore than 10 amps max into the bank that I've seen at 29.2V. I'm assuming this is because the batteries were largely fully charged.
I can change the setting to do a boost charge for longer if I want, but 2 hours appears typical on most installs. I need to check again first thing in the morning to see if it pumps in more juice than this, but the panels are able to supply enough power.
I thought this was normal to draw more power from the panels when under load? Wouldn't this indicate that the batteries were fully charged? It was around an amp or two prior at float levels.
I dont know how many hours at float charge would be needed to fully charge.
This is what confuses me about these controllers. The profiles charge the batteries for 2 hours once the sun comes up at boost voltage and then sits at float and can see the current drop away slowly over the day. But if the battery is more empty on any given day, would 2 hours be enough? Or is this just a standard thing regardless?
If done/set up properly the controller wont enter the 2hour boost/absorbtion stage before bulk charge stage is finished and batteries are nearly full.
 
Your problem is that you never fully charge your bank. That will kill your agm battery fast.

I just met a guy who managed to kill his 480 Ah agm in one and half years time. (Killed=15% capacity left)
Same problem: not charging to 100% regularly. He charged to 70-80% only during this 1.5 years period.

Point is that after the boost phase it takes 2-4 hours to do the absorption from 80 to 100%, and his charger switched to float way to soon. It went from absorption to float in 30 minutes.
And in float mode you do not charge!

Two hours boost/bulk with 10A is almost nothing for a 540Ah bank. 2 hours * 10 A = 20 Ah.

I would strongly recommend a battery monitor like the Victron SmartShunt or SmartBMV.
Then with your new batteries, monitor that you fully charge (that is 100%) at least once a week.
 
Last edited:
Where did you get the idea that he killed the batteries?

lead-acid bouncing back to 12 volts (no-load voltage) still has 30-40% charge left. (Maybe even bit more as it takes some time for batteries to settle to no-load voltage)
Well this is what I was worried about mostly. I didn't leave them sit for long, but it appeared to be climbing before switching on the generator.
Either way, is the rule that 10.8V under load is safe? This is supposed to be the 50% limit voltage for pretty much any lead batteries as far as I know when under load. Or with no load it's 23.2V and I was already hitting 24V.
 
If done/set up properly the controller wont enter the 2hour boost/absorbtion stage before bulk charge stage is finished and batteries are nearly full.
That's the thing with epever controllers they only do a boost phase at the beginning and then switch to float charge. There is no bulk charge.
 
Your problem is that you never fully charge your bank. That will kill your agm battery fast.

I just met a guy who managed to kill his 480 Ah agm in one and half years time. (Killed=15% capacity left)
Same problem: not charging to 100% regularly. He charged to 70-80% only during this 1.5 years period.

Point is that after the boost phase it takes 2-4 hours to do the absorption from 80 to 100%, and his charger switched to float way to soon. It went from absorption to float in 30 minutes.
And in float mode you do not charge!

Two hours boost with 10A is almost nothing for a 540Ah bank. 2 hours * 10 A = 20 Ah.

I would strongly recommend a battery monitor like the Victron SmartShunt or SmartBMV.
Then with your new batteries, monitor that you fully charge (that is 100%) at least once a week.
I was told that putting too much juice into these batteries at once isn't a good idea either. I think 30 to 40 amps or so will extend their life instead of going for 50 amps or more.
Anyway, isn't 26V the normal voltage for a fully charged bank? Thats what it was always reading at the end of a days charging.
 
For a 540 Ah bank you can charge with 60 amps no problem.

After the boost/bulk phase ends your batteries should go to absorption (for agm that is 28.8-30 Volts, see your specsheet)

Then after a few hours absorption the charger goes to float (26.4-27.6 Volts, see your specs)

Looks like your charger isn’t following the specs for agm batteries. Change the settings or buy a better charger.
 
Last edited:
That's the thing with epever controllers they only do a boost phase at the beginning and then switch to float charge. There is no bulk charge.
I’m not familiar with epevers, based on manuals some of them implement proper 3-stage charging and on some other models the manual is pretty vaque.

Boost phase timer SHOULD start when battery voltage reaches boost voltage setting-value. You can verify this by yourself, discharge your battery to 50% or so and keep log of your battery voltages and current every 30 minutes in the next morning. It should start from somewhere 12.2 volts or so, charge at maximum current until boost voltage limit is reached, start the boost timer and stay at boost voltage stage for 2 hours.

If your battery bank is badly sulphidized it is possible that the charge controller goes straight to boost stage as the battery doesnt charge. But if I understood correctly this was new batteries and new installation?
 
For a 540 Ah bank you can charge with 60 amps no problem.

After the boost/bulk phase ends your batteries should go to absorption (for agm that is 28.8-30 Volts, see your specsheet)

Then after a few hours absorption the charger goes to float (26.4-27.6 Volts, see your specs)

Looks like your charger isn’t following the specs for agm batteries. Change the settings or buy a better charger.
I need to take a better look when they are charging next time, ive been away at work and have not been around to monitor voltages. I haven't been using them long and suspect that they are likely taking more current now that I'm using the bank and draining them. The batteries were new when I was making the measurements initially and the whole bank was only taking about 10A in total.

I was told here not to use equalize charge with AGM batteries and either set equalize charge time to zero minutes or just set its voltage the same as the boost volts.
I just turned it off completely by setting to zero minutes.
That advice seemed to make sense as my inverter charger also said to disable equalize if using AGM batteries.

My datasheet only has data for boost volts and float, which is 29.2v boost and 27.2 float respectively.

Does this look right?
 

Attachments

  • Screenshot_20220619-135215_EPEVER Pair.jpg
    Screenshot_20220619-135215_EPEVER Pair.jpg
    147.2 KB · Views: 7
Last edited:
I’m not familiar with epevers, based on manuals some of them implement proper 3-stage charging and on some other models the manual is pretty vaque.

Boost phase timer SHOULD start when battery voltage reaches boost voltage setting-value. You can verify this by yourself, discharge your battery to 50% or so and keep log of your battery voltages and current every 30 minutes in the next morning. It should start from somewhere 12.2 volts or so, charge at maximum current until boost voltage limit is reached, start the boost timer and stay at boost voltage stage for 2 hours.

If your battery bank is badly sulphidized it is possible that the charge controller goes straight to boost stage as the battery doesnt charge. But if I understood correctly this was new batteries and new installation?
Ok, perhaps this is where I'm going wrong. (Ser my above post) my epever definitely does 3 stage charging, but I'm only using boost and float.

The retailer was hopeless when I spoke to them and they said just leave it on the default setting from factory for sealed batteries, but the voltages were a bit on the low side for these batteries.
Yes they are new batteries.
This is the data here.
Is absorption the same thing as equalize?




Edit:
It's come back to me and I got it wrong.
This controller is just 2 stage from what I can tell.
Equalize is only for flooded cells.
Not sure what I might need to tweak here.
 
Last edited:
Your boost duration, absorbtion period is too short at 2 hours, with your charge currents 4 to 5 hours are needed before float.

Consider the batteries fully charged when the charge current into the batteries falls to less than 1% of capacity, 5 amps, at 29.2 volt. Or when the current stops falling.

Chargers using a timer based absorbtion period often do not complete the charge process, relying on a high float voltage to 'top up' the battery. This is OK for AC shore chargers with unlimited time, with solar there is not enough hours in the day.

The result with AGM batteries of not completly charging is a gradual run down of capacity.

Unless AGM are charged in the bulk stage with a current approaching 2C, 100 amps fot your battery, service life may be reduced.

Consider adding a mid point balance circuit.

This is how your controler works,
You dont need equalise, set the volts equal to boost and duration to zero.

When it wakes up it gets maximum power from the panels and converts to a charge volts and current. It continues to do this until the boost volts ' target' is reached. This is the bulk charging stage.
The controler then enters a constant voltage stage where the controller maintains the battery voltage at boost volts. This stage is the boost duration, absorbtion period.
Once the boost duration is completed the controller drops to float voltage.
If battery load or solar conditions drop the battery volts below 'boost reconnect' the controller renters the bulk stage.
During the solar day the total boost duration is limited to the value set

Mike
 
Last edited:
Your boost duration, absorbtion period is too short at 2 hours, with your charge currents 4 to 5 hours are needed before float.

Consider the batteries fully charged when the charge current into the batteries falls to less than 1% of capacity, 5 amps, at 29.2 volt. Or when the current stops falling.

Chargers using a timer based absorbtion period often do not complete the charge process, relying on a high float voltage to 'top up' the battery. This is OK for AC shore chargers with unlimited time, with solar there is not enough hours in the day.

The result with AGM batteries of not completly charging is a gradual run down of capacity.

Unless AGM are charged in the bulk stage with a current approaching 2C, 100 amps fot your battery, service life may be reduced.

Consider adding a mid point balance circuit.

This is how your controler works,
You dont need equalise, set the volts equal to boost and duration to zero.

When it wakes up it gets maximum power from the panels and converts to a charge volts and current. It continues to do this until the boost volts ' target' is reached. This is the bulk charging stage.
The controler then enters a constant voltage stage where the controller maintains the battery voltage at boost volts. This stage is the boost duration, absorbtion period.
Once the boost duration is completed the controller drops to float voltage.
If battery load or solar conditions drop the battery volts below 'boost reconnect' the controller renters the bulk stage.
During the solar day the total boost duration is limited to the value set

Mike
Ok thats helpful, so really I should set this to say 5 hours and perhaps slightly lower the boost voltage to 29V?
If I knew the limitations on these controllers, I would have gone for something different, but I didn't know enough at the time about how the stages of charging worked, but I was recommended this and seemed to be a popular brand among our retailers.
Speaking of a balancer/equaliser, I've been looking at this.

My batteries have largely leveled out, but would still like a balancer, they are out by about 0.2 of a volt each. im wondering if I should shop around for a different charge controller, but I've already spent on this one now.
 
Epever solar controllers are popular due to a resionable performance and cost. Lots of endorsements on you tube. Good profit margin for retailers.
The default boost duration ( absorbtion period) of two hours is a compromise, 2 3,4 or 5 hours may actually be needed to fully charge. This will depend on battery type, SOC when charging commenced, available charge current, type of application, standby or daily cycle.
You may need to experiment a little to determine the optimum boost duration.

Having a battery monitor that can be programmed for lead acid batteries may help, recommended are Victron smart shunt or Victron BMV712.

Mike
 
First off is to gain an understanding of what a battery is. Batteries are a chemical device and not a can of gasoline. A good write up is at: https://climatebiz.com/battery-capacity/

All the well meaning advice on proper charging techniques does not change the basics. Voltage as a indication of capacity is a real loose measurement. Frankly most inverter setups have a ridiculously low voltage cutoff as standard. By the time they cutoff you have already passed where it is good for the batteries chemistry to be at. As a illustration about voltage and capacity is solar panel Voc rating. You can see a high voltage on you panels in dim daylight with no load. However try loading them and the voltage drops like a rock if they are not in full sunlight. Your batteries can show normal voltage also without load. However they lack the chemistry inside to deliver anything.

Rather than waste too much time in the weeds in micromanaging your charging you probably want to look at your loading and solar supply. Suffice to say, if you can avoid putting much use on a battery you are better off.
 
Epever solar controllers are popular due to a resionable performance and cost. Lots of endorsements on you tube. Good profit margin for retailers.
The default boost duration ( absorbtion period) of two hours is a compromise, 2 3,4 or 5 hours may actually be needed to fully charge. This will depend on battery type, SOC when charging commenced, available charge current, type of application, standby or daily cycle.
You may need to experiment a little to determine the optimum boost duration.

Having a battery monitor that can be programmed for lead acid batteries may help, recommended are Victron smart shunt or Victron BMV712.

Mike
Yes I'm looking at those battery monitors.
I dont know if it's going to help me much if I'm away and can't switch on the genset to boost the battery.
Here I was thinking that the charge controller was able to monitor the batteries and just connect it to the cloud via the app.
These victron ones seem to use Bluetooth, I don't know if they have a wifi module, but I would have to keep taking my phone out to the shed to check on it where it would be in range of Bluetooth.
 
Back
Top