diy solar

diy solar

24V AGM batteries in series at 13.40V and 13.63V. How to balance?

soylentgreen

New Member
Joined
Apr 23, 2021
Messages
96
I have two 145AH AGM 12V telecom batteries I use for emergency backup only - these batteries sit at float for 364 days per year, as they are connected to a UPS.

On charger at float stage, one is at 13.40V and the other is 13.63V, so an imbalance of 0.23V.

How out of balance is this?

What's the easiest way to do a one-time balance manually? I don't think there's any justification for adding more hardware to this system, since they are not cycled regularly.
 
The easiest way? Disconnect the battery bank from the system, then throw them in parallel for an hour or so. Any imbalance after that should be not worth worrying about. Even a quarter volt on AGM is negligible so unless you're really bored....
 
The easiest way? Disconnect the battery bank from the system, then throw them in parallel AND CHARGE for an hour or so. Any imbalance after that should be not worth worrying about. Even a quarter volt on AGM is negligible so unless you're really bored....

Revised.

Putting two 12V in parallel will transfer very little charge especially when done for only an hour or two. Any induced current raises the low voltage and lowers the high voltage further reducing the delta and the current. Also, as lead acid voltage has a tendency to settle following a charge, the voltage differential is ever-decreasing thus providing even less potential for current to flow between them.

Putting them in parallel and charging will create a reservoir from which the low battery can draw much faster than simply just being in parallel.

The only chemistry that benefits strongly from non-charging parallel wiring is 3.7V Lithium. The strong voltage to SoC correlation results in negligible "settling" and encourages charge transfer until they are very nearly equal.

For an operational system, options:
  • Charge the lower voltage battery individually while wired in 24V.
  • Drain the higher voltage battery individually while wired in 24V.
  • Invest in the aforementioned HA01 or similar balancer for ongoing maintenance.

NOTE that the most important voltage deviation is at peak 24V system voltage. If you're charging to 28.8V, imbalanced batteries could see the higher voltage battery ABOVE the specified 12V limit (14.4-14.8V typical). This can damage the high battery and undercharge the low.

If you observe your 12V are above spec, lower your system absorption voltage such that the highest voltage one is not over spec.
 
Six months later (after doing nothing), the two batteries are at 13.62 and 13.37. The spec sheet for these batteries is to float at 13.44V to 13.56V, so they are a little out of spec.

With the batteries still in service, I've hooked up a 12V automotive backup bulb to the higher voltage battery, and it's being drained at about 2 Amps.

How much should I drain?
- A 0.25 volt difference on a 12V battery at rest could be as much as 10% to 30% difference in state-of-charge. Let's say it's 20% ( I suspect it may be less than that, since they are in float).
- These are 145AH batteries, so I need to drain 20% * 145 = about 30 Amp Hours
- at 2 amps. this will take 15 hours of running the light bulb.

I have a feeling I could be off by 2x to 10x in my calculations, so I'm going to check every hour or two (turn off the light, let voltages settle, and take new voltage readings. Will report back once it's done.
 
Six months later (after doing nothing), the two batteries are at 13.62 and 13.37. The spec sheet for these batteries is to float at 13.44V to 13.56V, so they are a little out of spec.

With the batteries still in service, I've hooked up a 12V automotive backup bulb to the higher voltage battery, and it's being drained at about 2 Amps.

How much should I drain?
- A 0.25 volt difference on a 12V battery at rest could be as much as 10% to 30% difference in state-of-charge. Let's say it's 20% ( I suspect it may be less than that, since they are in float).
- These are 145AH batteries, so I need to drain 20% * 145 = about 30 Amp Hours
- at 2 amps. this will take 15 hours of running the light bulb.

I have a feeling I could be off by 2x to 10x in my calculations, so I'm going to check every hour or two (turn off the light, let voltages settle, and take new voltage readings. Will report back once it's done.

NOTE that the most important voltage deviation is at peak 24V system voltage. If you're charging to 28.8V, imbalanced batteries could see the higher voltage battery ABOVE the specified 12V limit (14.4-14.8V typical). This can damage the high battery and undercharge the low.
 
If it was me I'd probably charge them individually with a power supply to do a form of an equalize charge. If you don't already have a 30V - 10A power supply you might look into getting one. I use mine more than I thought I would.

I'd start by setting the power supply to 14.4v and it's full 10 amps and charging them in parallel. At some point the amps will start to taper but the voltage will stay at 14.4. Once you can keep both of them at 14.4v with less than 10 amps I'd split them up and charge them one at time until they both took exactly the same amps to hold at 14.4.

There's all sorts of specs out there but some of the ones I remember that on sealed telecom batteries was they talked about a finishing charge of 1 amp per 100 amp hours. That's 1.45 amps to hold your battery at 14.4v, assuming you're doing this inside at around 75F.

But I'd not focus so much on that as I would on being to able hold them at 14.4v with the same amps. EG: being able to hold them at 14.4 with 2.5 would still be good.

Not sure how far down the rabbit hole you want to go, but picking up a Victron smart shunt and doing a full discharge/charge test could be the only way to really compare thier health. Of course that has its own drawbacks.
 
Thanks for the tips - this is mostly a hobby project / UPS for emergency only and the batteries are old and should be replaced soon, so I'm mostly doing this for "fun" :)

While on a float charge of 27V: after an hour draining the higher 13.6 battery it dropped to about 12.6, suddenly much *lower* than the other battery, but I figured this was just surface charge, and concluded that trying to equalize while in float wasn't going to be simple.

So, I pulled the plug from the UPS so it began discharging, and quickly the true voltages emerged: 12.6 and 12.45, with the higher battery again higher, about 0.15V above the other. The initial 0.25V difference was probably bogus, including about 0.1V of "surface charge"

After 2 hours, with the high battery draining via light bulb an additional 2 amps, the difference has reduced from 0.15V to 0.13V (12.34 vs. 12.21) or about 0.02 V. (the entire pack is draining at 5 amps while this happens, a roughly C/30 rate).

This suggests a rough equation: 4AH drain = .02V difference, so each 0.01V difference is about 2 amp hours.

If true, to equalize the remaining 0.13V, this suggests I need to drain the high battery for about 13 hours more @ 2 amps.

This happens to be exactly what my initial prediction was (15H @ 2A) which seems almost too perfect, so I probably shouldn't trust the math.
 
Final results: after about 6 hours of drain @ 2A, the batteries are within .01V of each other. So my initial estimate of 15 hours was a bit high.

FYI, I found it much easier to balance them when discharging, or charging. Trying to balance when in float mode was difficult, because of the large "surface charge" which leads to big voltage fluctuations.
 
Final results: after about 6 hours of drain @ 2A, the batteries are within .01V of each other. So my initial estimate of 15 hours was a bit high.

FYI, I found it much easier to balance them when discharging, or charging. Trying to balance when in float mode was difficult, because of the large "surface charge" which leads to big voltage fluctuations.

NOTE that the most important voltage deviation is at peak 24V system voltage. If you're charging to 28.8V, imbalanced batteries could see the higher voltage battery ABOVE the specified 12V limit (14.4-14.8V typical). This can damage the high battery and undercharge the low.

So?
 
That's good advice in general, but doesn't apply with this setup - this UPS charges at a constant 27V (13.5V x 2) and doesn't do "bulk" or "equalize" voltages at all.
 

diy solar

diy solar
Back
Top