Safe cutoff point for lead acid batteries

[BillGreen]

I have a small diy solar setup with 4x400w panels that for most of the day produce more than I consume. I’ve added 3 90 amp flooded lead acid batteries in series (36V) to save this excess, but I don’t seem to be getting much of a return for all the effort. The batteries charge and discharge fine, and the efficiency seems all good, but I never get much more than 400-watt hours before reaching what I believe is the safe cutoff point of 12.3 volts on each battery.

My question is; is this safe value of 12.3 volts measured on load or in a rested state? I’ve searched the various battery sites for the answer to this but can’t find an answer so any help would be appreciated.

 

[chrisski]

Really recommend a hydrometer. That will accurately tell you state of each cell. If not balanced, will see you need to equalize.
————
If these are new deep cycle, you should get more. If these are old or car batteries, that may be all they have.

I prefer tracking battery status from a shunt rather than voltage. This gives much more accurate data.

Knowing the total ah of the battery is key. Although you have 90 ah @ 36 volt, if they’re degraded the true capacity is hard to judge.

12.3 volts is around 75%.

 

[sunshine_eggo]

I have a small diy solar setup with 4x400w panels that for most of the day produce more than I consume. I’ve added 3 90 amp flooded lead acid batteries in series (36V) to save this excess, but I don’t seem to be getting much of a return for all the effort. The batteries charge and discharge fine, and the efficiency seems all good, but I never get much more than 400-watt hours before reaching what I believe is the safe cutoff point of 12.3 volts on each battery.

My question is; is this safe value of 12.3 volts measured on load or in a rested state?

The voltage charts you see are for RESTING voltage only with a wait of typically 10-24 HOURS with no charge or discharge of any kind for them to be accurate.

Based on 3 * 12V * 90Ah = 3240, you're not getting below (3240-400)/3240Wh = 88% SoC.

A battery monitor that actually counts current in and out of the battery is the best way to truly measure state of charge. The Victron Smartshunt is among the best, but there are others.

You can experiment. Even a 2 hour wait time can yield much better results.

Cut of your discharge @ 12.3V and let the battery rest for two hours. What voltage? That's closer to your actual state of charge.

 

[sunshine_eggo]

Hydrometers are good, but they aren't very accurate in an active system. Batteries need to be stable for about 2 hours before SG readings are reliable, and different manufacturers use different SG values for "full charge." Rolls is 1.265 and Trojan is 1.278 for their FLA.

 

[BillGreen]

Wow, I’m not used to such quick replies!

The batteries are all relatively new. I’m still waiting for some balancing kit to arrive, but voltage wise they all seem to track within a tenth of a volt, so I think I’m safe for now in that respect. I’ve written my own software to control things so I can measure and see things like current voltage and charge in watt hours. I’m lead to believe that I shouldn’t discharge the batteries more than 50% to extend their life but it’s this 12.3 value that’s got me wondering.

 

[sunshine_eggo]

Wow, I’m not used to such quick replies!

The batteries are all relatively new. I’m still waiting for some balancing kit to arrive, but voltage wise they all seem to track within a tenth of a volt, so I think I’m safe for now in that respect.

Tracking through the working range isn't that big of a deal. What really matter is that ALL 12V are within a safe and effective voltage range during absorption - preferably as close together as possible.

I’ve written my own software to control things so I can measure and see things like current voltage and charge in watt hours. I’m lead to believe that I shouldn’t discharge the batteries more than 50% to extend their life

That's correct. 50% discharge will increase the total # of cycles by about 4X. If they're rated for 500 100% cycles, you'll get 2000 50% cycles. With excellent FLA maintenance and equalization charges when appropriate, one can get even more.

but it’s this 12.3 value that’s got me wondering.

It means absolutely nothing unless your batteries have been completely at rest for 10-24 hours.

 

[BillGreen]

Thanks for that. Just to be clear, at this small scale, I don’t think I’m working these batteries particularly hard with the discharge rate never exceeding 40W, and charge wise my constant current phase is limited to just 4 amps.

The intention is that this runs 24/7 so waiting 24 hours to check the state of the battery isn’t really an option. For sure it recovers to a slightly higher voltage as soon as I remove the load and left for several hours it creeps up a bit more. It’s almost like the batteries just have no capacity? I even tried a routine to gradually reduce the power as I approached this 12.3v point which did extend the run time slightly but it wasn’t great.

 

[Checkthisout]

I have 3 8D 240ah (rated) deep cycle batteries in my cabin system. In parallel @ 12 volts.

I cycle down to 11.5 volts under a 25-30 amp draw (with further sag when running a microwave or something) and then the generator starts automatically and brings them back up to full.

Under a slow (5 amp draw) running some security equipment when I'm not there, I cycle down to 11.9 volts.

In either case I get about 1.5 Kwh out of the batteries. Again, they are marine deep cycle, not solar batteries.

They are 2.5 years old and still deliver the same amount of energy as when new. I charge them with absorb set at 15.2 volts for 2 hours.

 

[Checkthisout]

In addition, yeah 12.3 volts isn't low enough for you to see any real out put out of the batteries. Mine drop relatively quickly from 13.2 volts resting when charge is turned off and drop fairly quickly from 13.2 to 12.4 volts.

Run them down to 11.9 see what you get.

 

[chrisgraham]

The voltage of the battery will be slightly lower than at rest. Therefore, to maintain the safety and life of the battery, set the safety cutoff point at the voltage under load.

 

[SamDeleted]

My question is; is this safe value of 12.3 volts measured on load or in a rested state?

Down to 11.6 under load is roughly 50%soc from what I understand

 

[mikefitz]

50% state of charge evaluated by voltage is a pure guess, especially as the effective capacity is reduced with increasing load.
The only useful method is to use a battery monitor that allows setting charge efficiency and Peukerts constant.
Even a correctly set up battery monitor may not give accurate readings unless the actual battery capacity is known and entered into the monitor, your 90Ah batteries may not have the specified capacity, especially after some use.
A further factor that may introduce errors is determining 100% full as a reference point, many solar charging set ups do not fully charge the battery.

Battery volts and SOC with different load currents

 

[BillGreen]

Obviously my system is on a very much smaller scale that what most here seem to have but the basic principles and battery rules must be the same. My system was designed purely as an experiment to see if it was possible to get saving from solar without it taking years to recover the costs. I only attempt to cover background demand ie, tv, freezer etc which tends to average about 300W, leaving the grid to deal with anything higher. Any excess solar is used to charge the batteries which it does easily at this time of year.

My batteries are just bottom end leisure types designed for mobile home use, but in theory, capacity is capacity and given my maximum draw is currently only about 1 amp, I’d have though such a battery type would be perfect for this. I appreciate that voltage alone doesn’t give a reliable indication as to the state of discharge depending on the current, but in my case the current is constant and minimal, but the control software still relies on this magical 12.3v number. I’m sure taking this down to below 12v will improve things albeit at the risk of a shortened battery life, but it never stays at that level for long before the next days charge cycle takes place. I’m going to give it a few days and wait for my balancing to arrive before I try this.

BTW I’m fairly confident that my charging routines are good and follow the same three stage principles that a good quality mains charger.

Anyway, thanks to all for the advice. I’ll post the results of my test when everything’s in place.

 

[Checkthisout]

Obviously my system is on a very much smaller scale that what most here seem to have but the basic principles and battery rules must be the same. My system was designed purely as an experiment to see if it was possible to get saving from solar without it taking years to recover the costs. I only attempt to cover background demand ie, tv, freezer etc which tends to average about 300W, leaving the grid to deal with anything higher. Any excess solar is used to charge the batteries which it does easily at this time of year.

My batteries are just bottom end leisure types designed for mobile home use, but in theory, capacity is capacity and given my maximum draw is currently only about 1 amp, I’d have though such a battery type would be perfect for this. I appreciate that voltage alone doesn’t give a reliable indication as to the state of discharge depending on the current, but in my case the current is constant and minimal, but the control software still relies on this magical 12.3v number. I’m sure taking this down to below 12v will improve things albeit at the risk of a shortened battery life, but it never stays at that level for long before the next days charge cycle takes place. I’m going to give it a few days and wait for my balancing to arrive before I try this.

BTW I’m fairly confident that my charging routines are good and follow the same three stage principles that a good quality mains charger.

Anyway, thanks to all for the advice. I’ll post the results of my test when everything’s in place.

What is your absorb time and voltage?

Edit: The 90 ah rating is what you'll get if you draw the battery down to 10.5 volts, probably 9.6 volts under a 10 amp load.

You didn't give group size, brand type.

 

[RCinFLA]

Lead acid batteries have high overpotential voltage slump under load so best way to decide when to stop discharge is via a Columb counter. At 0.25 C(A) discharge rate a 12v lead acid will drop from full charge open circuit voltage of 12.7vdc down to 12.0v in about 15-25 minutes of load current. It will recover to near full charge 12.7vdc if load removed in about 20 minutes.

For inverter low voltage cutoff, you have to take battery voltage slump under load and cabling voltage drop into account or you will cut off on heavy inverter loading. 12.0v is highest voltage I would use for inverter low voltage cutoff. A lot of cheaper inverters have a 10.5 vdc shutdown to account for heavy inverter loads and a lot of cable voltage drop.

 

[BillGreen]

I don’t understand what you mean by group size? The batteries though are Powerline Leisure XV190MF and I have three of them in series.

The absorb voltage is set for 42V (14V) and I normally hold them at that for a several hours depending on the amount of solar I’m getting. I declare them fully charged when the current eventually drops below 1 amp. After that I maintain at 39V (13V) at a maximum of 200 mA

It generally takes a total of about 8 hours under reasonable sun, 5 hrs of boost and 3 absorb. I don’t allow any discharge until the boost phase is complete. My charge indicator seems to consistently clock about 450-500 Watt hrs of charge

edit to add

My inverter for battery to mains is a cheap 600W Chinese micro inverter with an input range of 22 – 60V. With a nominal 36V I’ve measured its efficiency at about 90% which I’m happy with. I tried a 1200W version, but it wasn’t very stable at the low power levels I need and kept tripping out. I use an identical 600W inverter for solar to mains. Now that’ll probably sound a bit odd to most here I bet. ?

 

[sunshine_eggo]

Based on typical values, both absorption and float are low. Typical FLA absorption is 14.4-14.8V depending on brand. Float is generally around 13.5V. A 1A cut off is about right, but at 14V, it may be a little low.

 

[Checkthisout]

I don’t understand what you mean by group size? The batteries though are Powerline Leisure XV190MF and I have three of them in series.

The absorb voltage is set for 42V (14V) and I normally hold them at that for a several hours depending on the amount of solar I’m getting. I declare them fully charged when the current eventually drops below 1 amp. After that I maintain at 39V (13V) at a maximum of 200 mA

It generally takes a total of about 8 hours under reasonable sun, 5 hrs of boost and 3 absorb. I don’t allow any discharge until the boost phase is complete. My charge indicator seems to consistently clock about 450-500 Watt hrs of charge

edit to add

My inverter for battery to mains is a cheap 600W Chinese micro inverter with an input range of 22 – 60V. With a nominal 36V I’ve measured its efficiency at about 90% which I’m happy with. I tried a 1200W version, but it wasn’t very stable at the low power levels I need and kept tripping out. I use an identical 600W inverter for solar to mains. Now that’ll probably sound a bit odd to most here I bet. ?

You're under-charging and your batteries amp hour rating is extremely optimistic and you're grossly under-cycling the batteries.

Charge to 15.2 for around 2 hours and cycle down to at least 11.9 volts. Lower if under moderate to heavy load.

 

[Vigo]

Yeah, i dont have much to add in actual detail but my big picture read on the situation is that you're being overly cautious and that's the primary reason for the disappointing results.

Despite that, you have MANY other things right. You just need to get a little more aggressive with your settings to get your money's worth out of those batteries in a reasonable time span.

 

[BillGreen]

Here’s my current setup that needs a bit of a tidy up. Its feed by four panels on the garage roof, one east, one west and two south.

I’ve set all the levels based on what I’ve read online and not wanting to overly shorten the battery life. But I think the advice here is correct, I just need to work them a bit harder and hope for the best. And at least doing this will mean I’m sending more power to the battery rather than back to the grid for which I receive nothing.