Multiple battery packs in parallel

[melvin2345]

Let's say for example I have a 16p3s battery pack of 18650s for 11.1v 35.2Ah. Let's say later down the road I wanted to double the capacity. Could I create another 16p3s battery pack and connect that pack to the other one in parallel, each with it's own BMS (See Below), or should I just modify the original battery pack by adding more cells to it? In either case, I would be using the same cells (manufacturer, voltage, mAh).

If this is acceptable practice, is there a practical limit to the number of packs you can run in parallel like this?
Assuming this is acceptable, I would imagine amps from the charge source would need to increase as you increase total capacity to keep charge times from becoming excessive. Is this accurate?

I apologize in advance if these are ridiculous questions. I'm new to the 18650 game and have TONS to learn.

 

[Sunder]

The answer as always is "depends".

If capacity is the only thing you care about, then putting them under the same BMS is acceptable. The cells will self balance per series string, and there will be a little bit of intra-string charging. That is, the older cells will drain faster, since they would have lost capacity over time, and once the load is off, the newer cells will feed current back into them. This will make the pack a little less efficient than you think, but unless you were measuring down to the mah, it's doubtful you'd notice.

By using a full separate BMS, you can increase the total power output of the batteries, but the batteries can only balance as whole batteries, not as series-strings, and the balancing needs to happen through the BMS. The problem with this, is that if you were under high load for a long time, the difference in voltage between packs might be so high, that the current to rebalance those would be quite high, and could trip or even blow the BMS.

If you need no extra power, and just need extra capacity, I'd personally add the extra cells behind the BMS, rather than get another BMS and make two whole batteries. Of course, sometimes having two separate batteries can be very useful, such as having one out for repairs, or lending one to a friend, but if you don't foresee those situations, then I'd recommend just one bigger battery and one BMS.

Is there a practical limit? Yes - Eventually you'd need a charger bigger than the BMS can handle, or a recharge could take 2 or 3 days, (or months, or years, if you went that insanely big). You would also always be limited by the max current rating of the BMS.

Hope this helps.

 

[dogman dan]

You could go either route, but if you just make the battery bigger, you can't ride with half of it ever. But you can charge the larger pack with one charger.

You can parallel batteries that have a bms. Nothing special needed, just hook them together when they are fully charged. I do it all the time. This is for batteries of the same voltage of course. Both 13s, 48v. Can't parallel a 14s with 13s.

For me the plus side is that I can still run the half size pack when I'm just going a few miles. It's not like I need the larger size to pull more amps on all my bikes. One bike does need both to run without pounding the battery to death. Usually I parallel for the range, not the amps.

The down side is charging, either one by one disconnected from each other, or with two chargers both at the same time, again, while disconnected.

I need the ability to mix and match my batteries, so I go the route of two packs and parallel. I may need 10 ah one day, 20 another, and sometimes I need 53 ah.

If you have one bike, and need the amps, or just want sag, one bigger battery makes sense to me. But I have 4-6 bikes running usually, and need no less than two batteries. Some bikes need a small light battery, others need everything I have to go 100 miles.

 

[whereswally606]

An interesting read. I have some b&q macallister 36v 2.6 ah batteries which I intend to parallel. I have 5 and intend to get a 6 so I can do 3s2p as well. (think in that case I will need a diode) but since I cant without damaging the batteries access the internal cells I have to use the 6 individual bms for protection. I like the idea of having a battery weight appropriate for the journey. I intend to batch make the adapters to access and hold the batteries via plastic printing and my prototype is looking good so far. Just need to print the rest and cut some terminals for the +ve and -ve connections.

 

[melvin2345]

If it makes any difference, the max load at any given time would be 10 amps or less. I'm only going to use this to power some 12v LED lights in my workshop, and some outdoor 12v LED lights on my deck. Won't need a ton of power, I would just like to add capacity over time as I acquire more 18650s.

 

[Voltron]

Just an FYI.... The parallel banks have to be the same voltage, but not the same capacity, so you don't need to exactly double it to make sure they're identical or anything like that.

 

[melvin2345]

Just an FYI.... The parallel banks have to be the same voltage, but not the same capacity, so you don't need to exactly double it to make sure they're identical or anything like that.

My OCD begs to differ.

 

[Sunder]

Just an FYI.... The parallel banks have to be the same voltage, but not the same capacity, so you don't need to exactly double it to make sure they're identical or anything like that.

My OCD begs to differ.

Your OCD would have kittens, if you did any real testing.

I got 10 batteries brand new from the same manufacturing batch, and the lowest was 15.7Ah, and the highest was 16.7Ah, with most of them between 16.0 and 16.3.

Imagine the difference if they came from a different batch - with a year of wear in between them

 

[dogman dan]

It won't matter a bit either way, if they are paralleled. Whatever the real world actual capacity at the moment of one is, adds to the others. They will remain close enough to the same voltage as they discharge, regardless of size, and internal resistance. You only have to match the voltage and chemistry. So no OCD needed.

And,, if you really want to, you can even mix up chemistry, but some combos suck, while others are easy. That one does require a bit of OCD, for the combos that suck.

For a stationary application like yours, no need whatsoever to duplicate bms.

Build a separate pack, of the same voltage. parallel the main wires before the bms to the other. Parallel the small bms wires to the other packs bms wires, before the bms.

Now they are paralleled at pack voltage, and if any group of cells gets very low, the bms shuts it all off.

The tiny bms wires can handle 2 amps, and its very unlikely that anything will get so out of whack they carry any more than that.

Or,, you could just use fat wire, and parallel directly to the old pack. each cell group getting one fat connector wire, plus one more for the negative end. Either way fine. Either way, both packs charge through the one bms, one charger.