Chargings Multiple Packs in Parallel

3287 Views 20 Replies 9 Participants Last post by Swogm, Jun 11, 2021

tscott8201

Discussion Starter · Mar 5, 2018

I've got a golf cart that I am retrofitting with 48V 45Ah Li packs salvaged from a Chevy Volt. I am currently testing the cart with just one of these 12S packs and so far so good. I'm using a programmable charger and a BMS on the pack. I'm charging to 4.1V per cell for a pack voltage of 49.2 when fully charged.

45Ah won't give be enough range so I plan to add 2 more of these packs and run them in parallel. Each will have its own BMS but I'd liek to run just one charger. Can I charge all 3 packs in parallel with a single charger, allowing the BMS on each unit to regulate when it can an cannot charge?

Tom

1 - 20 of 21 Posts

PAXtss

· #2 · Mar 5, 2018

To add another question:

Can you monitor all packs with one set of BMS that are tied with small wire jumpers into same level cells on other packs? Only 12 BMS ties but each connected to 3 cells.

For your question- you should have no problem with the charging - it will just take longer. I am planning on doing it with two halves of a Leaf for 180V. I just don't know how many BMS I will be using - 48 or 96- $600 difference.

brian_

· #3 · Mar 5, 2018

PAXtss said:
Can you monitor all packs with one set of BMS that are tied with small wire jumpers into same level cells on other packs? Only 12 BMS ties but each connected to 3 cells.

I don't think so - any pair of cells which are not electrically connected at their positive and negative terminals are not really in parallel, and cannot be expected to be at the same voltage, even if they are nominally the same because they are the same number of cells up the stack. If the cumulative voltage of all cells "below" that level in each stack is different, the little BMS monitoring wires would be taking substantial current to equalize those voltages.

PAXtss

· #4 · Mar 6, 2018

Thanks Brian,
I can see where it could open issues if there was an imbalance in the two large packs.
Having each cell (actually two parallel cells) on its own BMS would be safer.

brian_

· #5 · Mar 6, 2018

PAXtss said:
Having each cell (actually two parallel cells) on its own BMS would be safer.

Yes, or in the case of a Leaf re-configure the packs to parallel at the module level (instead of parallel high voltage packs), which involves electrical connections using threaded studs. This is not so easy for the Volt modules, which would need their welded cell-level connections to be modified.

PAXtss

· #6 · Mar 6, 2018

I feel like I hijacked this topic-
but- one last question - I looked at re-configuring the cells- these are gen 3 leaf - double modules and the cells are glued together and even though I think that I could separate them I fear doing damage as I have 24 doubles so I would have to separate 48 cells- scary

So each pack is is series with heavy cables at negative and positive terminals going to controller. I could connect all cells at same level with connecting wires- any size and thus parallel the cells on that level. The question is how much current could/would flow through these wires? It seems that due to same level of voltage there would be minimum but-- I did flunk out of engineering school and that was not EE.

dain254

· #7 · Mar 6, 2018

I've been charging my Volt Modules in parallel for several months now with no issues on my Tomberlin Vanish UTV - my top balance cheap BMS simply has a parallel harness that plugs into the two balance ports. I've seen this done with as many as 6 modules. The only thing to watch out for is before plugging it in make sure they start at the same voltage so there isn't a huge surge of current to balance them upon plugging in!

See less See more

brian_

· #8 · Mar 7, 2018

PAXtss said:
... So each pack is is series with heavy cables at negative and positive terminals going to controller. I could connect all cells at same level with connecting wires- any size and thus parallel the cells on that level. The question is how much current could/would flow through these wires? It seems that due to same level of voltage there would be minimum....

If the centre tap terminals of paralleled Leaf modules are connected, the only current between them would be due to imbalance of cells which are at the same potential. That must be a small fraction of the current drawn through the main terminals, or the cells would be severely out of balance at the end of discharging or charging. With a couple hundred amps (per module if in parallel) on the intended path, it's hard to imagine more than a few amps through the ties between the centre taps. But that's with the cells paralleled.

Connecting just BMS taps across packs is essentially paralleling them through BMS wires.

PAXtss

· #9 · Mar 7, 2018

Actually I was thinking about paralleling the whole cell or module. Not just the center tap which is positive for one cell and negative for the other. There would be three wires tothe first module - neg,center tap, positive and two to each following module (-,+,-,+,-,+,-,+) same as a BMS would do. so the cells on each level would be parallel from one pack to next while each pack is in series and parallel as a pack. The question is how much current flow from cell to cell under different circumstances - current draw of motor and a weak cell situation being two I can think of. The wires connecting the cells could be larger than regular BMS wires if needed. It seems feasible but electricity tends to trick my brain a bit. I thought that Tesla does something like this with fused wires in case a cell goes out.
I do know that the safe way is to just BMS all lower parallel cell - 96 in total-$600 more than 48 if l did the wires, it is just my cheap nature that I am looking at options.

See less See more

brian_

· #10 · Mar 7, 2018

PAXtss said:
I thought that Tesla does something like this with fused wires in case a cell goes out.

This Tesla configuration is simpler than any of the scenarios which we are discussing.

All current Tesla batteries have many cells parallel at the lowest level (number depending on battery capacity), then purely series beyond that. The fuse wires (just wire, so fine that it acts as a fuse) connect each cell to the bus plate for it's level of the module, and there is one BMS connection to that bus plate. If a cell passes too much current, the wire melts and that group of parallel cells just has one less cell; it makes no difference to the voltage seen by the BMS.

brian_

· #11 · Mar 7, 2018

PAXtss said:
There would be three wires to the first module - neg,center tap, positive and two to each following module (-,+,-,+,-,+,-,+) same as a BMS would do.

You need to connect to every centre tap, so the BMS can manage all modules at the cell level. Whether you tie the centre taps of modules directly together (with the bus bars as you've shown before) or run separate BMS leads to them, they all need to be connected. Have I missed something?

dain254

· #12 · Mar 7, 2018

One other important aspect to consider if paralleling through the BMS taps is to have modules that came from the same batch (so the same original pack). As Brian mentioned, when you load the cell the voltage sags and if you have slightly different cells there will be some amperage between the BMS leads depending on how the different cell groups handle the current draw.

tscott8201

Discussion Starter · #13 · Mar 8, 2018

I plan to run the charge and discharge in parallel, but I'm probably going to run each pack on it's own BMS. I'll set both BMS modules the same with an over voltage cutoff set just above the charge voltage I set on the charger. For now I'm charging to 50V, so I'll probably set the BMS to cutoff charging at 50.4V.

Tom

Moltenmetal

· #14 · Mar 9, 2018

dain254: whose "cheap top balance BMS" are you using?

Personally I'd want a fuse on each of those BMS lead jumpers between cells to protect the conductors. But, given the small current flows which I would expect between pairs of paralleled cells arranged like this most of the time, it looks like it should be OK. You might get in trouble if you ever start scraping the bottom of your SOC though- one cell could easily go 0.5 volts below its neighbour and the current flow between them could become pretty high (hence the desire for the fuses).

It's a darned sight easier than what Yabert did with his- he TIG welded bare copper wire to the terminals of the cells between groups of Chevy Volt pack cells to put them in true parallel.

dain254

· #15 · Mar 9, 2018

Molten - I saw how Yabert tied his in parallel a while back which led me to thinking there has to be an easier way! There is a company that makes a clip that grabs onto the tabs on the cell groups easily but I still don't trust it. I never run the vehicle below 3.2V/cell which is another reason I think I'm fairly safe. The small wires used could also somewhat act as a fuse themselves - but I agree the absolute safest way would be to use a fuse and a proper BMS that is using that harness to measure individual cell voltages.

As for the BMS itself, it is one of the eBay "SUPER HAPPY FUN TIME BMS 60A" that can be bought for about $30 I'm seeing. Since I only have it tied into my charge circuit which is somewhere around 15A, I can easily get away with the low current BMS. I randomly check my cells from time to time and they have never been off even slightly - again likely due to the fact they came from the same complete Volt pack.

See less See more

boekel

· #16 · Mar 9, 2018

How would you detect a blown fuse?

dain254

· #17 · Mar 9, 2018

It would have to be wired differently - and I'm not even sure how it could work! Someone smarter than me might be able to figure it out. If the fuse blowing meant the voltage would show zero, the BMS I have would stop it from charging so you would know there is an issue somewhere - other BMS could just show a 0V.

boekel

· #18 · Mar 9, 2018

Probably one fuse would blow between cells, leaving the bms connection intact, only connected to one less string on that cell-tap

Yabert

· #19 · Mar 10, 2018

dain254 said:
I saw how Yabert tied his in parallel a while back which led me to thinking there has to be an easier way!

Yes, just plug the chevy Volt cells lead into my small PCB with resistor

http://www.diyelectriccar.com/forums/showthread.php?t=190906

Jimbo69ny

· #20 · Mar 10, 2018

Im doing the same thing. Many people will say it cant be done but I know people who have and it works.

One thing to keep in mind is that the pack with the lowest resistance will provide more power than the other pack. That matters in my build because for me, one pack is a couple years old, the other pack will be brand new.

1 - 20 of 21 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.