Replacing NiMH batteries with Li-Ion

[SubnetMask]

First, let me start by saying I know that NiMH chargers are not meant for use with Li-Ion and would likely destroy the cells at minimum (and best case) - my thought is to not charge the batteries with the original charge mechanism. Let's assume for a second that I want to make a pack where the cells are removable so they can be charged in a LitoKala or similar charger.

That being said, I have a WireScope 350 set with dead battery packs that use 3800mAh 4/3A NiMH cells in a 6S configuration for a 7.2v nominal pack. These packs are 'obsolete', and if you can find them, they're $200+ from what I've found. From what I gather, it seems like a fresh, fully charged AA NiMH is 1.4V, so since the 4/3A and AAs are both '1.2v' batteries, assuming this info is correct, then the fully charged voltage of a NiMH 6S pack is 8.4V - a fully charged Li-Ion 2S pack would also be 8.4V. From a purely 'power the device' perspective, and assuming that charging will never be done as it was, through the device, does it seem like replacing the 6S 4/3A NiMH batteries with a Li-Ion 18650 2S or 2S2P pack would be safe for the device? For discharge, I could get a small BMS to prevent the batteries from being over discharged, but if the voltages seem safely in line for the device, I'd look to find a way for the cells to be removable so I can charge them externally with a proper Li-Ion charger.

 

[amberwolf]

Generally you can replace the old chemistry with the new...but keep in mind that a "full" NiCd type battery doesn't stay full, because most of the chargers keep charging the cells until they reach their "voltage collapse" point, where the >1.2v portion of charge turns into heat, and the charged pack voltage actually ends up down at the #cells x 1.2v, rather than x 1.4v. This is how nickel batteries "balance" without a BMS. NiMH cells have less of a voltage drop, but they still have one. You'd need to check your specific cells to see what their drop is supposed to be.

If you look around, you'll find that nickel chargers generally use either Delta-T (change in temperature, where the pack gets hot at end of charge due the above), or Delta-V (change in voltage, where the voltage drops at end of charge due to the above). If the pack has a thermistor in it that hooks up to it's charger, then it almost certainly uses Delta-T, and may *also* use Delta-V (some use both), to terminate charge. It might be a "dumb" charger that simply outputs a constant voltage that is well below the total # of cells times the "voltage collapse" point voltage of the cells they used, so that it would never reach taht point and so they wouldn't have to worry about that in designing it. (is cheaper)

So the operating voltage of the device from battery is likely to be based on the "lower" full voltage, rather than the "higher" 8.4v (6 x 1.4v). It might not be (probably isn't) enough difference to make any difference (but it could be).

But, just in case, I would investigate this before building the new battery so that you don't damage anything in the device, just in case. If the device can operate while the battery is charging and *not* from a separate power input (or with the battery switched physically to a "charging" state, disconnecting it from the device), but actually from the battery/charger, then it can probably handle >8.4v and it'll be ok.

It is likely that it can indeed deal with it fine...but myself I'd rather be certain. The easy way to deal with it if it does turn out to be an issue is to simply not fully charge the new Li pack.


If you want to use your own external charger, and never charge internally, you'll need to locate the charging circuit inside anyway, and disconnect it's output from the battery positive terminal.

 

[SubnetMask]

Generally you can replace the old chemistry with the new...but keep in mind that a "full" NiCd type battery doesn't stay full, because most of the chargers keep charging the cells until they reach their "voltage collapse" point, where the >1.2v portion of charge turns into heat, and the charged pack voltage actually ends up down at the #cells x 1.2v, rather than x 1.4v. This is how nickel batteries "balance" without a BMS. NiMH cells have less of a voltage drop, but they still have one. You'd need to check your specific cells to see what their drop is supposed to be. <snip>

Thanks for the info. The way this pack is build, it has a thermistor between two of the cells, that, based on the markings, cuts the circuit at 70*c. Although I would like to to devise a way to power it with a 2S2P 18650 setup, which if I were to use my 3200mAh cells, would give something like 75% more runtime than the original 3800mAh pack, the problem is I did (just, like, tonight) rebuild one pack with 4000mAh NiMH 4/3A cells, and before assembling the pack, I charged all six cells up to full with my LitoKala, so the pack, after being built, read around 8.4v, which should be about full, but when I put it in the device and powered it on, it was complaining about the battery being at 0% and of course, critically low. For grins, I removed the battery and probed the positive and negative contacts on the device while it was connected to its 12VDC wall wart, and surprisingly, the contacts read ~11VDC. I would have expected something closer to the packs full charge voltage... So it seems it just shoves full voltage into the battery until the cells can't take any more, get hot and trip the thermal cutoff. Next interesting bit is these cells that were or should have been full charge, while reading 0% initially, after connecting the power supply and letting it sit, the charge percentage started rising - As of now, about two hours later, it's reporting 29%. When the pack reaches 100% per the device, I'll have to see what the pack says for voltage...

After your post, along with the voltage readings from the 'fully charged' pack, I quickly whipped up a 2S2P 18650 pack to see what it would do, and again, like with the fresh NiMH pack, it read ~8.4v and when I powered the device on, it said 0%. So I don't know what the 'logic' is - it's 'fuel gauge' doesn't seem to be based (solely) on pack voltage - maybe it needs to charge the pack to 'calibrate' and if you just throw a new pack in, it doesn't know what to do with it... So as much as I'd like to build an 18650 based pack for it where the cells are removeable/externally charged and easily replaceable, it seems like that may not ever be able to work with this device.