Dual Battery Connection Adapter

[saneagle]

We are miscommunicating.

I was talking about my weak cell group which i am convinced is caused by a single dodgy cell inside a group of six that are in parallel. My point was about how that parallel connected group must be flowing current into the weak cell.

I know how series charging and balancing works.

That would only happen if the one cell had an internal short, but then it would go all the way to zero.

You guys come up with all these clever theories. Why don't you try actually opening a battery and measuring things, then you find out stuff? I have never found even 0.01v difference between cells in a parallel group while charging. They're all held in parallel by plates of nickel sheet, so they can't be at different voltages to charge each other. it doesn't matter which one you measure, they're all the same.

 

[Ghost1951]

I have never found even 0.01v difference between cells in a parallel group while charging. They're all held in parallel by plates of nickel sheet, so they can't be at different voltages to charge each other. it doesn't matter which one you measure, they're all the same.

Of course they stay the same voltage. They are connected together.

That doesn't stop my dodgy group being dragged down in voltage by comparison to the ones it is connected to in series. The whole group is affected.

This morning, after being left unused for three months while I have been riding the Argos folder, the overall battery voltage was 40.2v while my bad group was 3.82v. It isn't that the weak group is powering the BMS while it is sitting unused because it falls quite rapidly when fully charged and put to use. Even when put back in balance, that group drops about 0.2v after five miles by comparison with the others.

Generally, the BMS puts it back to near balance and now and then I charge that group on its own when it gets too far out of line. I have two fused wires leading to insulated contacts on the outside of the case which I can use to monitor and charge the group when needed.

 

[eas2lv]

Part of the confusion here is not distinguishing between the emf of the battery and voltage in the circuit. Batteries connected in parallel will have same voltage in the circuit but can have different emf under load because of the drop across the internal resistance. When load is removed, no current is flowing, and emf could be different and will try to equalize.
That said, if you start with batteries having same emf, account for the emf vs capacity relation and note that capacity change is equal to integral of the current, they have to be in dynamic equilibrium at all times. That does not allow an emf mismatch to develop. I think this @saneagle 's point.
But something catastrophic can happen where one battery loses all capacity to generate an emf and acts like a passive resistor. Use of a diode protected Y will guard against such occurrences.

 

[Nealh]

Sorry to be a party pooper, but I must ask...

Why on earth would you want two batteries on your bike???

I had two batteries for longer rides , 160km from one battery wasn't pssible for me.

 

[Nealh]

I had two bike set ups with parallel batteries and at the end of each journey both batteries on each bike was in much the same ball park voltage wise.

My Boardman had a PF 14.5ah cases battery and a softpack of PF 's 10s 8p a simple Y lead connected them but always made sure charge voltage was within 0.02v before connecting up.

The Swizzbee ended up with a bespoke PF pack I made and fitted in to the frame case, 23.2ah and a HE2 pack of 17.5ah sat in the rear top bag . Like wise as with the above final SOC were much the same.

 

[Jodel]

My knowledge of the electronic / electrical side of things is rather less than comprehensive, so I like to keep it as simple as I can.

I've never really had a range problem on the tandem as we rarely go more than 30 miles these days, but having access to a second battery to swap over is quite comforting if I want to put the power up a bit to combat headwinds. I can fully understand the comments about voltage sag, but again that's never been a rate limiting factor for me (yet).

One of the plus points of a tandem is the size of the frame - mounting a second battery isn't much of an issue.

 

[Bikes4two]

If someone is considering connecting two batteries in parallel via schotky diodes, maybe take a look at the benifits of 'ideal diodes' first as their voltage drop and power consumption is better that schotky diodes.

This link HERE, one of many, gives an insight into this.

I'm like some others in this thread in that I run two batteries in parallel (no diodes of any sort for me) when going on a ride long than the range of a single battery and in this way I'm not lugging the extra weight around, and I 'parallel up' at the start of the journey with each battery pack at the same SOC (within 10mV or so of each other).

 

[Ghost1951]

If someone is considering connecting two batteries in parallel via schotky diodes, maybe take a look at the benifits of 'ideal diodes' first as their voltage drop and power consumption is better that schotky diodes.

This link HERE, one of many, gives an insight into this.

I'm like some others in this thread in that I run two batteries in parallel (no diodes of any sort for me) when going on a ride long than the range of a single battery and in this way I'm not lugging the extra weight around, and I 'parallel up' at the start of the journey with each battery pack at the same SOC (within 10mV or so of each other).

Interesting item. I looked on aliexpress and the xl74610L was the one I found and it is quite cheap, but they are either very near or underspec for even a 36V battery. The max uncooled current is 15 amps and the max voltage is 36v. To be honest, I wouldn't be drawing more than 15 amps, or not much more, but the fully charged battery is obviously at 42 volts, not 36. Maybe two connected in series would solve that problem, and the losses and voltage drop are negligible.

They can be sensitive to voltage transients - no idea whether that might or might not apply to this application.

Maybe there are tougher ones I haven't seen.

https://www.aliexpress.com/w/wholesale-ideal-diode-30a.html?g=y&SearchText=ideal+diode+30a

 

[lenny]

STMicroelectronics 100V 30A, Rectifier Diode, 3-Pin TO-220AB FERD30H100STS | RS

uk.rs-online.com

 

[AntonyC]

Yes, the voltage under load stays the same. But a higher proportion of the energy extracted comes from the pack with lower internal resistance, so its position on the discharge curve changes more than the other one.

When the load is removed, the packs therefore want to be at different voltages, and so a current must flow to equalise voltages. Size of current depends on internal resistance of the battery packs.

Above with mixed cell types more current comes from one than the other and the ratio of their contributions changes and reverses during discharge. This also happens below using identical cells with differing resistances, in this case interconnect resistances.


Below with mixed cell types when the load is removed the orange and yellow cells recharge the green and blue ones:

I think one of the tests on that forum showed one cell of an off-load P-group recharging another and later being charged itself from the second one as they reached equilibrium. At the end of the day there are times when the simple Rint or R || R+C models for cells don't do them justice.

In practice the more mismatched paralleled batteries are the faster they'll wear from 'invisible' current flow between them, and that's also a reason for matching cells closely in a P-group. With diodes we can mix batteries of different capacities or ages without that issue. With or without diodes the load sharing can be very unequal which means the Ah ratings add up but not necessarily the peak A ratings, and it's why you never charge batteries any faster in parallel.

 

[Ghost1951]

This site has some serious expertise on it. I learn things here on a regular basis.

 

[harrys]

At the end of the day there are times when the simple Rint or R || R+C models for cells don't do them justice.

Thanks AnthonyC for stating a complex topic so simply!

There was a time when I would think ..."Posh, They're in parallel. No way they can have different voltages." Then I paralleled a 2AH and a 5AH tool pack and the resultant capacity was 3AH. I blamed it on not understanding how their BMS worked.

The kicker was when I built a 10S-5P battery, 50 cells needed, and only had 48 of the same type cells. I put mixed cells into one group, adding two higher capacity cells. Gotta work. They're all in parallel. First test ride, I got half of the expected capacity out of the battery. That mixed group had fully discharged and tripped the BMS, when the other nine were at half charge. Acted just as you described.

Still, when you parallel batteries, if the current draw is low relative to their capacity. they will probably discharge as expected.

 

[Benjahmin]

Still, when you parallel batteries, if the current draw is low relative to their capacity. they will probably discharge as expected.

And I think this is the main advantage. Current draw is split between the two batteries. Therefore, assuming that either one battery was capable of running the kit by itself, the current draw stress on each battery is approximately halved, so lengthening the batteries lives. It also means that steep hill sag(there's plenty of them here) is far less pronounced.
Also, for me, the discharge level is not so deep, again stressing the battery less. My body has a distance limit of around 25 miles, maybe 30 on a very good day. Having two batteries has not changed this.