Dual batteries in parallel (different capacities)

[JuiceMeUp]

I come to this forum as the purest source of online information. Apologies if it has been asked before.

I have a Talaria MX-4 with the stock 60volt battery (16S, 45AH: 20700 cells: Max rating of pack 150+ Amps)
I have just built an auxilary 60volt battery (16S, 10AH: 18650 cells: Max rating of pack = 30 Amps)
BIke pulls well over 150 amps under full throttle. Stock battery BMS is not bypassed and handels it OK.

I plan to parallel them together for additional range and am considering the best way to do this. Aux pack will live in a little tank bag
Both batteries would start at exactly the same voltage

Option 1:
Make the parallel connection at the controller DC inputs.
Concern - WIll the load at the controller overwhelm the AUX battery pack as both packs split the same load? Say 75 Amps each? which would trip / destroy the Aux pack, which is only rated for 30 Amps. Or would the Aux pack max out, voltage sag and the main pack take the bulk of the work current draw needed.

Option 2:
Plug the Aux pack into the charge port of the main battery.
As the main batteries voltage comes down, the Aux pack voltage would come down to match.
Concern - unregulated current coming in through the charge port, potentially blowing the main packs BMS.

I know enough to be dangerous. But you don't know what you dont know and trying to avoid a fire between my legs is the reason for this post.

Anyone with direct experience or a proper engineer could weigh in on this plan. Appreciated - thanks

 

[From-A-To-B]

I would definitely not plug the aux battery into the charge port of the main pack, for the reasons you already listed. Plus, the charge port wiring was likely not set up to pass more than a few amps.

I’d Y them into together near the controller. Be sure to use good connectors and sufficient gauge wiring. The two batteries should share the load well, but there’s a chance the aux battery sees more than 30A draw from time to time. That depends on the particulars of the cells as they share the load. In that case, either the BMS will shut down the aux battery, or the BMS could cook — depends on what the BMS was built to do.

Paralleling two batteries at low loads (20-30A) is much more straightforward because all the gear can easily be over-specc’ed. good luck!

 

[amberwolf]

The capacity of the batteries in parallel isn't really relevant, but the load capabilities of them are, as well as all of the wiring between them and inside them if they are connected via charge port.

Do you know what that wiring is? And those limits? If you don't, you must find out *before* connecting things up this way.

If the ports *and* all that wiring, *and* the BMS if current is going to flow thru that from charge to discharge on the pack being used as the conduit, are not capable of the full current that the system will ever draw, worst case, then you dont' want to connect things this way.

If something goes wrong in the conduit pack so that it's cells stop supplying current, then all of the current will be drawn thru that wiring/etc from the aux pack.

If any part of that can't handle the load, things can go wrong from fuses blowing to BMS tripping and ending your ride, to overheated and blown BMS charge FETs, to fire from overheated connectors or wiring, etc.

If instead you directly parallel their outputs, leaving charge ports separated, then neither pack is part of the current path from the other, and the only concerns are wehther the packs thesmelves can individually handle the full system load, and the added wiring/connectors to do the paralleling.

The packs will not split the load evenly, they will split it based on their internal resistance / voltage sag under the load at any instant. If the less capable pack is a low enough resistance vs the better one, it will take more load than it should, and probably shut it's output off. (this is also all true of connecting the aux pack via the main pack's charge port).

If either pack shutsdown for some reason, and the other can't handle the full load of the system, then it's BMS should gracefully shutdown it's output, ending the ride but preventing problems, assuming the BMS functions correctly and has suitable overcurrent protection.

Since the aux pack is so much less capable, you may want to fuse it for 30A or even less, so it can't be pushed too far beyond it's abilities, but it can deliver that 30A as much as possible. (a 30A fuse is "guaranteed" to not blow at or below 30A; the current it does blow at, after how much time, will depend on the curve from the manufacturer's data sheet)


FWIW, there are lots of threads on paralleling packs, if you want more info on the various methods, problems, etc. many of them are in this list but not all (and not all in this list are relevant)

Search results for query: parallel

endless-sphere.com

 

[lnanek]

If instead you directly parallel their outputs, leaving charge ports separated, then neither pack is part of the current path from the other, and the only concerns are wehther the packs thesmelves can individually handle the full system load, and the added wiring/connectors to do the paralleling.

That only holds under load. Once you stop at a stop light or something, one of the batteries will drain into the other depending on which lost more voltage based on their differring internal resistance and voltage sag. That's why people start adding diodes/battery blenders when the batteries to be paralleled have different capacities.

 

[amberwolf]

If the connection resistances are as low as they should be, there won't be any voltage difference between the batteries. There has to be a sufficient resistance between two voltage sources for there to be a difference between them.

If say, the BMS FETs are high enough resistance, and the connectors are (like with poor crimps/contact) as well, and perhaps a loose fitting fuseholder, you could get enough resistance to create a significant voltage drop under high enough currents.

It has to be a relatively high resistance; for instance a 1 milliohm shunt probably has more resistance than the contacts of connectors should have, and will only have a tiny voltage drop across it under load. FETs in a BMS are in the same order of resistance, generally at most a couple of milliohms. So neither one of those should be able to cause any significant voltage difference between paralleled packs.

 

[Bush mechanic]

Anyone with direct experience or a proper engineer could weigh in on this plan. Appreciated - thanks

Not an engineer and never had a system exceeding 2000w, but I've done this for years without issues with much lower power packs (14s drawing 40A total), using a mix of batteries. It can work perfectly well in principle as already stated above, but I don't think what you are proposing, at least certainly not via charge port, is a good idea. These wires are normally sized far below discharge rate, so unless you can confirm the internal wiring and BMS capability I'd be concerned about something even as simple as melting wires/insulation within the battery where they could then create a short.

I'd parallel them if I had to, but I'd still be concerned with such a significant mismatch in discharge ability unless you're confident the weaker battery can protect itself (including allowing for the possibility your main battery shuts down/loses connection etc).

I'd probably use a shunt to measure current flow from the smaller battery see what actually happens as you increase power.

 

[AnthonyC]

OP didn't mention a BMS for the smaller pack, that would leave it vulnerable to damage. To recap, there's no easy way to control the current sharing whether connected directly in parallel or using diodes, or with BMSs if they simply cut off at excessive current. So both batteries have to handle the maximum load current individually but can differ in capacity and they need diodes to prevent them re-charging each other. Unless the batteries are very unalike 2 x 150A may handle your bike's "well over 150A" but it's not guaranteed.

Suppose the internal resistances are 50 milliOhms and 160 mOhm and they're in parallel. Under a load that makes battery voltage fall by 5V they'd be supplying 100A and 31A: the larger battery's coasting but the smaller is overtaxed and its BMS cuts out. Now that 31A will come from the larger battery causing a further 1.5V drop. Say your battery was at 60V under load, you're doing 30mph and your motor's making 53V BEMF, your bike just lost 20% of its 7V drive voltage.

Battery performance varies with age (years), charge distribution (seconds) and everything bar mood, sadly you can't simply do the math to build a second pack to match the resistance and behaviour of a first that's seen some use. Hence needing another 150A at whatever Ah, or less if it's OK for the bike to lose power or cut out occasionally.

@AW there's no voltage difference where the packs join but they have their internal resistances between them. Chances are the batteries will drain unequally creating an internal difference when their voltages recover off-load, then one will drain into the other. They'd age faster and being uncontrolled we can't assume they'd be (and stay) safe.

High and low drain cells mixed

@ Wolf, The story of different ir makes sense. But the low ir (high drain cells) can be easy used with high ir cells(low drain cells) We use the cells at really low drain, so i think there is no issue/problem here. The other way around, if i would use the average laptop cell in my power tools...

secondlifestorage.com