What about NIHM cell packs (like 1000+ AA cells)

Hello everyone,

I just joined here because I started to get curious about building a powerwall from NIMH AA cells.
Specially because I might be able to get 1000+ decent cells for free.

I was searching the net on it, but couldn't find a lot. Only packs that go up to like 10 cells, or 12V. Have found one guy trying to build pack for his e-bike, but did not see him actually using it.
I'm also kinda getting mixed responses to NIMH pro's and cons vs Li-ion...
In general I'm more listening to the statements that say : "they should contain more energy compare to NIMH, butreceive and give less current compare to Li-ion". What should make it more usefull for high capacity Home-energy-Storage systems. But seeing nobody building them yet, I guess they have major drawback why they aren't used.

Is it because they are difficult to keep in a good condition duo to no giving a good easy measurable "i'm full" characteristic?
Or is it about the life cycles? I hear really different story's about that. I alway's though they could do more cycles then Li-ion, but in practice they suffer to much from damage due to bad real life use cases like charging or memmory effect from partial discharge or deep discharge.


Can someone shed some light on this and help me prevent spending time and money in building a huge NIMH pack that is not going to work (or just for only 6 month?).

1. They have memory effect - if you don't discharge them completely and start charging, they'll lose some capacity.

2. They require a good ?t/?V charger to not kill the cells

3. They get hot at charging

So they don't have any serious advantage compared to Li-ion, that's why they are going to become history. Nowadays they are mainly used in low-cost devices.

In one way or the other 18650s are already too small to build a powerwall. It's a big effort to build a large battery from such small cells. AA cells make it even worse, do you really want to do this? Good AA NiMh cells store about 2.5Wh so you need 400 cells per kWh. Just insane. NiMh chemistry is perfectly fine for a powerwall but the AA format isn't. There are bigger NiMh cells available, used in electric bikes and cars, which would make it much more viable but they are hard to get in big numbers. The times of nickel based batteries are long gone in most areas.

I'd advice against doing it. Your time and money can be used much more efficiently.

You might be able to test and re-sell on eBay or elsewhere. They're still commonly used in low-power consumer devices and are still relatively expensive to buy new. I'd be interested in buying 10-20 to get rid of all my alkalines (peeve of mine, I've had like 4 devices destroyed just this year from leaky alkaline cells).

rev0 said:

I've had like 4 devices destroyed just this year from leaky alkaline cells).

Click to expand...

You are the second poster to bring up the subject of leaky alkaline .... Its a major annoyance forme too ... many appliances which I don't use regularly I don't leave cells in ....

It can be no coincidence the sellers of many electronic items also manufacture most of these cells ...Panasonic for one ... How hard can it be to make a cell that doesn't leak ???

This is all by design and won't end until the bastards are sued!!

The easiest (though expensive) fix is just to use NiMH cells in everything. I've nearly eliminated alkalines from all my stuff, picked up a bunch of those Ikea 2500mAH AAs and 900mAH AAAs, and some more on Amazon of different capacity. Just got a Li-Ion 9V battery set also.

rev0 said:

The easiest (though expensive) fix is just to use NiMH cells in everything. I've nearly eliminated alkalines from all my stuff, picked up a bunch of those Ikea 2500mAH AAs and 900mAH AAAs, and some more on Amazon of different capacity. Just got a Li-Ion 9V battery set also.

Click to expand...

Me too. It's just non-eco to use primary batteries nowadays.
The 9V ones i don't replace with Li-ion 9V - i convert them into rechargeable devices with integrated li-ion battery with protection, MT3608 DC-DC step-up converter and a charge level indicator




Push the button to see the remaining charge level. The charging cable can beeasily hidden inside as on the photo above




I didn't implement USB charging because i prefer to charge the li-ion directly and to 4.1V, not 4.2.

thunderheart said:

1. They have memory effect - if you don't discharge them completely and start charging, they'll lose some capacity.

2. They require a good ?t/?V charger to not kill the cells

3. They get hot at charging

So they don't have any serious advantage compared to Li-ion, that's why they are going to become history. Nowadays they are mainly used in low-cost devices.

Click to expand...

The memory effect can be reversed to a reasonableextent, but it does require cycling the pack a few times. Also, NiMh do suffer self discharge as well.

That said NiMh have been used in hybrid cars. I really have no idea what the success rate.

Having used NiMh 5 cell 'hump packs' for receivers in RC cars, they really do require cycling every few charges, as I don't like to push voltage too low. Now I use 2s LiFe, a far better choice.

If I could get large format NiMh cells for free I may consider trying them for some kind of home storage battery, but again, I think you would be better off selling them and going with Lithium chemistry.

Thanks for all the replies!
I agree with some of the down sides, but there are also a few I'm not so sure about. I will share a few thoughts, maybe some people can give more insight in some of them,

- AA to small, lotsa work to make a decent kWh pack.
Agree! but if NIMH turns out nobody really tried it yet, I might try it if it has decent succeed changes.

- NiMH has discharge
I think there are multiple types of NiMH. The newer cells you can get sometimes advertise to lose way less change over a year then older NiMH do in a few weeks. Even to the rate that the newer NiMH you buy in the shop are already pre-charged just like a alkaline battery. I use the, myself now, and looks like they keep there charge good enough that I don't notice it when I use them half a year later. Did not do measurements on this yet though. But the discharge is far from what I remember form old RC toy things.

- NiMH has memory effect and need to be discharged fully before charging, so not good for those environments
I can't get my head around this one. Specially when I found out my Prius 1.8 actually uses NiMH for the quick regenerative breaking system. I have the one without the external pokey bit, but the once that do those use an extra battery pack (beside the NiMH one for breaking) and this one is Li-ion! Why? My knowledge about both types of cells would let me to a design that was opposite, Li-ion for fast charge/discharge of large currents, no memmory effect is perfect for dynamic breaking/throttle environment, and OK with the less capacity. And use NiMH for the ~20KM drive pack that can be charged with the pokey bit. As it is more likely it will be fully charged/discharged and has a higher energy density.
But in the prius the NiMH is dynamic and gives only ~2.5KM, while the same sized extra Li-ion battery option gives ~20KM but does not get charged from breaking. I can thing of a few things like NiMH can keep playing a resistive load for breaking even if they are full, but yeah. I mainly don't get this, as NiMH seems to work perfect in my car and did not notice any drop in performance/range even after having done already ~80.000KM in it while its total is almost 200.000KM now.

-NiMH get hot
As far as I know, they mainly get hot when overcharged. When not over charging I hear and guess have noticed myself that they don't get hot. It is only when I'm laying in bed and to lazy to get them out the feel a lot hotter the next morning then they usually do. That might be one of the problems tough. A lot of my NiMH charges don't cut power when full, and I learned NiMH is pretty tricky to sense when they full. I heard they drop ever so slight in voltage when they get full. But I have an good RC charger (Battery Charger Imax B6AC) that does sense when it is full... So the hot problem can be solved with a good BMS I think.

When searching on the internet I also get lots of mixed results. some say Li-ion have far more cycles then NiMH, but an other post/graph shops NiMH having way more then Li-ion. Are there different types of NiMH? and if so, is there a (sub)name to know what kind of cells your dealing with?

NiMh cells with low self discharge (LSD) are the default for about a decade now. If you buy NiMh cells today they have to be LSD. If someone tries to sell you non-LSD cells, don't buy. Only buy good quality, name branded cells. Preferrably only buy Eneloop.

NiMh cells don't have a memory effect. They suffer from a similar effect called lazy battery effect but it's far less serious and can easily be reversed.

Like all cells, NiMh might get hot during high current charge or discharge, but that is true for most other chemistries as well, so it's nothing special.

Charging is indeed the most tricky part. With low currents it's difficult to determine when the end of charge has been reached. With high currents it's easier but it might effect longevity. You have the find the balance in between the two. And it's important to do the charging properly as it's the easiest way to ruin a NiMh cell. Specifically the end of charge is important as they don't like overcharging at all.

dubble post from laptop restart, sorry

DarkRaven said:

If you buy NiMh cells today they have to be LSD. If someone tries to sell you non-LSD cells, don't buy. Only buy good quality, name branded cells. Preferrably only buy Eneloop.

Click to expand...

LSD cells have lower capacity compared to non-LSD and higher internal resistance. They are good for low power applications, but for power-hungry ones non-LSD is a better choice.
Eneloops are highly overpriced... that's why i prefer Chinese PKCELL brand. Last year i bought about 20 LSD and 30 non-LSD cells from them and they are the best choice for me.

DarkRaven said:

NiMh cells don't have a memory effect. They suffer from a similar effect called lazy battery effect but it's far less serious and can easily be reversed.

Click to expand...

Call it whatever you want, it doesn't matter. I call it memory effect.

DarkRaven said:

Like all cells, NiMh might get hot during high current charge or discharge, but that is true for most other chemistries as well, so it's nothing special.

Click to expand...

Really? Have you ever charged a Li-ion or LiFePO4??????????????

thunderheart said:

DarkRaven said:

NiMh cells don't have a memory effect. They suffer from a similar effect called lazy battery effect but it's far less serious and can easily be reversed.

Click to expand...

Call it whatever you want, it doesn't matter. I call it memory effect.

Click to expand...

How could it possibly not matter if it's a different thing which it clearly is? Just because it is a SIMILAR effect it is not the SAME effect.

thunderheart said:

DarkRaven said:

Like all cells, NiMh might get hot during high current charge or discharge, but that is true for most other chemistries as well, so it's nothing special.

Click to expand...

Really? Have you ever charged a Li-ion or LiFePO4??????????????

Click to expand...

Yes, really, and yes, I have.

DarkRaven said:

How could it possibly not matter if it's a different thing which it clearly is? Just because it is a SIMILAR effect it is not the SAME effect.

Click to expand...

It's the same effect, according to Wiki:

DarkRaven said:

thunderheart said:

DarkRaven said:

Like all cells, NiMh might get hot during high current charge or discharge, but that is true for most other chemistries as well, so it's nothing special.

Click to expand...

Really? Have you ever charged a Li-ion or LiFePO4??????????????

Click to expand...

Yes, really, and yes, I have.

Click to expand...

I've never had any Lithium cell which would get hot during charging. Never. They could get warm, but never hot if you don't charge them at crazy high rate.

Well, Wikipedia is wrong then. Which would not be the first time. Or they take into account that for most people, since Wikipedia is not exclusively made for experts, this explanation is good enough. The memory effect limits usable capacity while the lazy battery effect doesn't. It just lowers the discharge voltage of the effected cell so you have full capacity but less energy. To a certain extent this is even worse than the memory effect because it could mean that a device turns off immediately although you have put in a fully charged cell. But unless the memory effect it can be easily reversed. The memory effect usually effects NiCd cells while the lazy battery effect is a NiMh thing.

It depends on the definition of warm vs. hot. Lithium cells could easily get 20C above ambient when fast charged. Obviously they won't get hot when charged slowly. But it's the same for NiMh. The difference is their charging efficiency, NiMh reaches only 60% to 70% or so while lithium gets 98 or 99%. So it is easier for the NiMh cells, especially small AA size cells, to get warmer than a comparable lithium cell, but it generally isn't an issue. I think "NiMh gets hot" is not a valid disadvantage of NiMh cells.

I agree with DarkRaven on the warm/hot part. If you keep charging Li-ion (overcharging) they will get hot as far as I know. The only reason you might think Li-ion is not getting hot is because Li-ion is (almost) alway's charged with a charge controller of protection circuit. In my RC day's I also never had hot NiMH cells with proper charging currents, while the cheap NiMH AA/AAA chargers at home would get my NiMH's hot when I was to lazy to take them out in time.

BTW, if you want enloop battery's for cheap... go to IKEA. I have not found a solid written confirmation yet, but they are produced in the same factory and are damn good. People even did testing / comparing the IKEA version with original enloop cells and found almost none difference between the two while only ~25% of the price. I discovered a bit to late, just after I bought ~30 enloop battery's >.<

But back to the NiMH wall packs :
One thing that kinda worry's me about NiMH though it that the current it can pass though is way lower then Li-ion. So while the density is higher, and you might need less cells per pack if you where to use bigger capacity NiMH cells, the pack does not only deliver less current, but might also have a problem forwarding the current from other packs in series. I haven't really thought this through yet, but does increasing the whole powerwall voltage (putting more battery packs in series rather then increasing the battery pack number of cells) help with getting more watt's out of the system without needing to make the packs bigger in order to transport enough current at lower voltaget?

And does raising the total voltage of the battery setup effect the cells in a negative way? is there a limit to it? I see most people build ~24V setups. But the UPS I have seems want 192Volts (it has 16x12V lead-acid battery's in series). Though these voltages are really dangerous, is there a difference in using Li-ion or NiMH for a setup that produces 192Volts, or is this not even possible? In the past I used loads of Alkaline battery's to make a high voltage spark thing, but the battery's didn't quite like it, they would get really hot. So I guess the voltage increase is just helping overcome the internal resistance and putting through more watts of energy then the battery is designed for.

If this is true, then NiMH might just not work because of its current limitation.
If Li-ion can handle 2,5 times it current, but in a full setup only delivers 0,25 times its current, it has enough current room left for 9 other Li-ion packs in series, giving a total of 10 packes without going over the max current each cell can handle
If NiMH can handle 0,5 times its current, and delivers the same 0,25 times its current, it has only enough current room left for 1 other NiMH pack.

I'm I thinking straight or am I missing the boat totally?

In typical applications it doesn't matter how much current a cell can provide. Most people test at 1A and use 500mA or less. Even small AA NiMh cells can provide these currents.

If you all haven't discovered lygte-info yet, it's a great resource for detailed Li-Ion and NiMH testing, here's one for the Ikea NiMH cells: https://lygte-info.dk/review/batteries2012/Ikea Ladda AA 2450mAh (White) UK.html Looks like even at 10A (~4C) they only rise 17.1 C, which is pretty respectable compared to Li-Ion cells.

DarkRaven said:

In typical applications it doesn't matter how much current a cell can provide. Most people test at 1A and use 500mA or less. Even small AA NiMh cells can provide these currents.

Click to expand...

But isn't the total load you put on a cell more?
Even though a cell will only deliver say 20% of its max current, when having 5 cells in serie each with a current draw of 20%, doesn't that make for a total current flow of 100%
I guess I'm not thinking straight >.<

Hmm I really need to do some calculations on this to improve my understanding . I might really be wrong, as I think now, higher voltage decreases the current flow when giving it the same load.
Yeah guess it doesn't matter if you would put the same number of cells more in series or parallel. you just need to have thicker wires with more cells is parallel, en more wire isolation if you put more in series.

The current doesn't add up in serial connections.

If you have a load across cells in series then 100% of that load is present at every single cell in series. The current is always the same and the voltage rises by adding cells in series, that's why you get more power out of the cells. It's very simple, power is voltage times current.
And power is a constant with a fixed amount of cells. And because it's voltage times current you can effect power by changing either of these. Adding cells in series rises the voltage, giving more power, and adding cells in parallel rises the current, giving more power as well. 5 cells in series will give the same power as 5 cells in parallel.
You would think that serial connections are better, because the current will stay the same and you don't need bigger wires to handle the current. But not everything can be run at high voltages, high voltages might be dangerous and also you need the balance cells in series. Cells in parallel don't need to be balanced but they provide power through larger current which means you need bigger wires. So every battery is always a compromise between voltage and current.