help for building NIMH battery pack

[rompperi]

Hello ES!
I'm a noobie on electrics and e-bike are quite new to my country and there is not too much info on Finnish sites so I thought to try my luck here. I happen to got hold on hundreds of commercial rechargeable AA sized 2600mAh NIMH batteries and started to do research on how to make battery pack out of them since being a student I have to rely on DIY experiments . Seems like it is possible and would be decent option for my e-bike plan which is putting a 36v 250w geared hub motor on front of old single speed classic style steel bike. I'm looking to build 10ah 36v battery pack.
What i understood the main problem would be charging the battery safely when paralled so how could this be done correctly? Is it enough if I purchase smart charger with thermostat to prevent over-charge and PCB to prevent over-discharge?
Your site is enormous and seems to focus on lithium batteries so any guidance will be much apprecciated!

 

[mark5]

You could begin your research by reading the thread below.

Re: So is there any way to safely charge NiMH in parallel? by amberwolf
https://endless-sphere.com/forums/viewtopic.php?f=14&t=53223#p796792

Building a BIG NiMH pack and trying to charge in parallel? by SHARKBITEATTACK
https://endless-sphere.com/forums/viewtopic.php?f=14&t=53223

 

[Sunder]

AA? Ouch. Hope you have a lot of patience and an eye for detail. To make a 36v, 10ah battery, you are going to need a 30S4P battery: 30 cells in series, strings of 4 in parallel. Thats a lot of connections as potential assembly faults and future failure points.

If your cells are absolute top of the market cells and brand new, you might make it work, but even then it won't work well. Capacity and voltage are only two of the critical stats. Discharge rate is also important. The best cells on the market can discharge up to 2C (5.2A if your cells are genuine and still in good condition) so about 20A for a string of 4 in short bursts. Thats more than you need for a 250w motor, but only when the pack is new. When it starts aging, you don't have a lot of reserve to rely on.

This is before we get into overdischarge cell reversal, and as you pointed out, charging safety. NiMh does not use simple voltage as end of charge detection. It uses a delta v termination. So you can't just hook it up to a power supply, you need a charger that can detect an end of charge voltaged drop. The recommended charge rate is also between 10 and 20 hours to maintain battery life.

I don't want to be all negative, but to be honest, even if someone gave me top of the market brand new cells, I'd take a dozen with thanks to run my bike lights, but no way would I make an ebike battery with them. Too much work and the end result would be too unreliable.

You should note that electric bikes "skipped" Nickel as a chemistry largely. It went straight from Lead to Lithium, and for good reason. Nickel is hard to manage.

If you are going to do it anyway, spot welding 4 in parallel first, then connecting those into 30 series would be the recommended method. Cell holders are not recommended. Good luck.

 

[Sunder]

Just giving a bit more thought about my last post, I thoight I'd be a bit more clear, so not as sound so negative.

1. I am refering to using AAs only when I criticise complexity and reliability. My electric motorcycle was originally NiMH. In fact, 102 of them. But they are 20Ah and designed to be connected in series with quality connectors for the part. Not 2.3Ah and designed for spring contacts.

2. You could consider charging to 1.4v per cell, to avoid the delta v issue, but 36v smart chargers with built in delta v detection are a lot more available and cheaper than I thought.

3. If you have been given "hundreds" I would make the biggest pack I could, rather than targeting 10Ah. If you can do 30S5P or even bigger, that would mitigate some of the issues I raised earlier.

 

[amberwolf]

Don't parallel NiMH.

Not at cell level.

Not at pack level.

Because of the way NiMH works, you're potentially asking for a fireball if you parallel them.



FWIW, those little AA NiMH cells are only meant for low discharge rates. If you use them at really high rates, they'll heat up and potentially catch fire. If they don't catch fire, they'll be damaged by the heat, losing capacity and ability to provide current. Some of these cells are actually good and can handle the current, but not the ones I can recall that I've had.


There are two ways to avoid high discharge currents from a single cell (or series string of them), and the first (to parallel a lot of them) isn't possible (not safe at all).

The second is to do what Vectrix and Honda did, which is to series a whole lot of cells for a very high voltage, so that even at lower currents you still get enough watts out to be useful as an EV motor drive.

However, at only a few hundred mA (we'll be generous and say 500mA to 1A) out for the series string, to have useful wattage (say, 500W out of the battery peak, maybe half to 3/4 that continuous), you'd need a string of several hundred cells for around a thousand volts at full charge (half that if you get double the current). Not practical, I don't know of any existing controllers that would be useful to you on an ebike, and wiring, insulation, and safety would be a serious concern.


If you only need say 300-350W out of the battery for a 250W-at-the-wheel motor system, and the cells can handle an amp each, then you get down to the 300-350V level, which is still pretty high voltage.


I can't think of a safe and practical way to use those cells to make an EV pack from.




BTW, I also doubt most of these cells actually provide the Ah they claim. I don't think any of the ones I've ever used did; the highest said something like 2300 or 2500, IIRC, and I might've gotten most of that when they were new, even though they were just in flashlights and flashing bike lights, etc. That's partly counting self-discharge, as they wouldnt' keep a charge on their own for more than a month or two even when not in a device, so I had to always charge them up right before using them. Some of them, as they aged, within just a couple years, got to be nearly worthless. (Some I had on the A/C thermostat would have to be swapped out every week! that was the lowest-drain item I used them on, I think)




If these were bigger, like the D-cell or F-cell sizes, you could just use them outright as a simple series pack for "36v" on a typical low-power ebike kit. I've used packs made like that, and they work fine, for the most part.

It's just really hard to safely use the really small cells like yours for this.