Parallel series batteries

Thread Starter

hazim

Joined Jan 3, 2008
430
Hi everybody.

We have a 24V to 220V UPS system including 4 x 12V 250Ah Gel batteties.
I am wondering if it's good to put a connection like the red one in the attached diagram, because I think it balances the voltages of each of the two parallel batteries.

I'm saying that because I noticed one time in a 48V UPS system that the voltage of one battery of a four series bank was less than the others by about 2V, maybe because of a bad cell.

Theoretically and practically what doea this red connection changes?
 

Attachments

dl324

Joined Mar 30, 2015
12,871
Theoretically and practically what doea this red connection changes?
Series and/or parallel connected batteries are always a compromise. I think you're proposed additional connection just makes things worse.

The red connection would cause problems whenever the parallel connected batteries had different voltages. Without it, it would be possible for the batteries to all have mismatched voltages, but the combination of top series batteries might be the same as the bottom two.
 

BobaMosfet

Joined Jul 1, 2009
1,768
I would not recommend this, and this is why- when you connect batteries head to head, the battery with the higher potential will try to charge the battery with the lower potential. A variety of negative effects can result if the system is not actually designed for this.
 

KeithWalker

Joined Jul 10, 2017
1,779
Batteries can be safely connected in parallel as long as you make sure that the terminal voltages of batteries to be connected are the same. Adding the link will make very little difference to the performance of the battery.
Once the battery is fully charged, its capacity will be limited by the one that has the least capacity. With or without the link added, the weakest battery will maintain its voltage until it becomes discharged. Then the batteries in parallel with it will maintain terminal the voltage until one in that parallel path becomes also discharged. Then the battery terminal voltage will drop rapidly.
 

Ian0

Joined Aug 7, 2020
2,190
The most important thing about wiring large batteries in parallel is to keep the lead lengths the same for each string.
 

ApacheKid

Joined Jan 12, 2015
385
I would not recommend this, and this is why- when you connect batteries head to head, the battery with the higher potential will try to charge the battery with the lower potential. A variety of negative effects can result if the system is not actually designed for this.
For parallel batteries is there any gain to be had by sticking say a diode on each battery? that way current can only ever flow out of the battery's +ve terminal, not in, it could not get a "charging" current...
 

andrewmm

Joined Feb 25, 2011
1,465
For parallel batteries is there any gain to be had by sticking say a diode on each battery? that way current can only ever flow out of the battery's +ve terminal, not in, it could not get a "charging" current...
If you put diodes in
two things

a) How do you charge the batteries
b) the diodes are going to dissipate power you'd rather use yourself

The advantage of the diodes,
if you connect a lower voltage battery in parallel with ta higher voltage battery
you wont get a short circuit charge ( bang )
 

Ian0

Joined Aug 7, 2020
2,190
Measuring the voltage between the two halfway points is a very useful way of identifying a potentially faulty battery.
 

ApacheKid

Joined Jan 12, 2015
385
If you put diodes in
two things

a) How do you charge the batteries
b) the diodes are going to dissipate power you'd rather use yourself

The advantage of the diodes,
if you connect a lower voltage battery in parallel with ta higher voltage battery
you wont get a short circuit charge ( bang )
Well the diode would not be part of the battery, just present when the battery is connected to the inverter.

Yes it will dissipate power, something like 0.7 * I where I is the total DC draw, which on an inverter could be 10 Amps per 100 W AC load - give or take. So if drawing 1 KW that diode could dissipate about 70 watts.

But 70/1000 = 0.07 or 7% wasted energy due to diode heat, not insignificant.
 

andrewmm

Joined Feb 25, 2011
1,465
Well the diode would not be part of the battery, just present when the battery is connected to the inverter.

Yes it will dissipate power, something like 0.7 * I where I is the total DC draw, which on an inverter could be 10 Amps per 100 W AC load - give or take. So if drawing 1 KW that diode could dissipate about 70 watts.

But 70/1000 = 0.07 or 7% wasted energy due to diode heat, not insignificant.
You still have to heat sink the diodes to dissipate the 70 watts ,

Your proposing taking the diodes out of circuit to charge ?
 

Ian0

Joined Aug 7, 2020
2,190
So, if the batteries are on load, is there really going to be enough voltage difference between the two string for current to flow into the lower voltage string? Not unless there's a short-circuit cell. All that will happen is that the higher voltage string will supply more of the load current, no different to what would happen if each string has a diode in series with it.
 

PaulVdB

Joined Jan 7, 2016
5
Before putting two (equal !) batteries in parallel, make sure both voltages are the same. You can do this by connecting the two minus and put a resistor (lamp!) between the two plus poles. Check the voltage over the resistor (lamp). When this voltage is as close as possible to 0.000V, it is safe to connect the two + poles together. Now you can charge/discharge the two paralleled batteries.
 
Your battery configuration is similar to the modules in the early 24kWh Nissan LEAF battery packs (96S2P). For each of the 48 modules they employed a 2S2P arrangement as you've shown (with your red bar across the middle). For their design I believe they've added the bar across the middle to allow for the BMS to better monitor, balance and charge the cells within each of the 48 modules. I believe their BMS treats each pair of parallel cells as one which requires them to be kept well balanced. I'm not certain but this may have been designed as such to provide for a simpler more cost effective BMS which has provisions for only 96 cell pairs. The latest 62kWh LEAF battery pack actually has modules with three cells in parallel stacked serially (96S3P). The moral of this story seems to be that if you want better monitoring and control of your battery pack (necessary for Li-ion), then your bus across the middle is likely necessary.
 

daba1955

Joined Apr 27, 2019
130
I don't see much of a problem, providing you don't get a bad battery. These are gel batteries, so are rechargeable, and assuming they are all healthy, everything will balance out, the batteries or cells with higher terminal voltages will effectively "charge" the others, so you end up with everything balanced out, which it has to be since things are electrically connected at all nodes.

But put a bad battery anywhere, and all the others are going to be constantly trying to charge it, leading to a total drop-off in performance of the whole thing.

Proper battery management circuitry would prevent that happening, and would signal to the overall controller a bad battery or cell.
 
Top