Protecting the input of a lithium-ion battery pack

[derGoldstein]

I have a lithium ion battery pack with a PCM which protects the pack from overcurrent and overvoltage. While I trust the PCM to do its job, some failure states of the PCM result in the circuit dying (to protect the cells) and needing replacement. How would you go about "externalizing" this failure state, specifically when charging?
Worded a bit differently, I have a known PCM but an unknown charger. I'd like to protect the battery pack and the PCM from harm. The three things I can think of preventing are overcurrent (continuous), overvoltage (continuous), and voltage spikes. Another thing to protect against is inrush current, but I'm not sure it's applicable here.
(there's also reverse-polarity, but that's a separate circuit)

I think I'm looking for an input protection circuit, but there are so many configurations.

The first thing I'd think to do is add a normal fuse at the input. If all else fails at least I've limited the maximum current flowing into the battery pack. However I'd like to add a circuit that kicks in before this happens.
Could I achieve this with only passive components? Would a PTC be appropriate here? To divert voltage spikes, should I use a MOV, a GDT, or a TVS diode?

Any advice would be appreciated.

[SeanB]

Protect the input with a TVS diode, selected so it will clamp at a value lower than the maximum input voltage applicable to the input of the protection circuitry. Though in any case you will be much better off using a LDO regulator as a prefilter to limit voltage, and the TVS diode to protect the regulator instead. Just choose a LDO that is usable with a voltage fed to the input, or add a low voltage drop shottky diode to the regulator output to isolate it when power is removed. Adds an extra volt drop to the charge requirement, but will save the cells, and provide some input current limiting and thermal protection.

[derGoldstein]

I didn't consider an LDO as a protection device, but it might work for the lower-voltage packs. Thanks.
The highest battery pack I'm trying to protect goes up to 7S, which means 29.4V peak, at up to 3A.
I've filtered these specifications in mouser. If the output were less than 28V then I'd have a few options, but at 29.4V output I come up with only one device: the LD29300. Its input voltage only goes up to 30V, however, so anything above that and it would fry, and I don't know if the fail state is open or short. Even if I knew that the voltage going into it would never exceed 30V, I'd be using it at its absolute maximum V/A rating, and I don't think it would survive long.

Would the schottky diode prevent anything apart from reverse-polarity on its own?

Is there a particular reason to choose a TVS diode over a MOV in this situation?

[SeanB]

How about a simple power transistor ( or in this case for 3A power darlington) and TL431 adjustable zener to give you a regulator. not as good regulation, but will drop out gracefully with a 1V5 drop or so, and with a shottky diode you will only have 2V drop all the time, but a 200V transistor will give you the ability to withstand temporary ( under the time it takes the transorb to blow the fuse from overcurrent) connection to the mains.

[derGoldstein]

Is this for the current-limiting or the voltage regulation, or both?

The first stage of the charging process is constant-current, and the second is constant-voltage. I'm trying to think if I'd have to tweak the charging voltage of the charger to compensate for the voltage drop... The voltage drop isn't constant, it will reduce when the current is low enough (like at the end of the constant-voltage stage), exposing the battery pack to a higher voltage than the 4.2V per-cell max. But if I don't compensate then the battery pack will always be charging at a lower rate, and will only reach 27.9V.
Also, at its peak (near the end of the constant-current stage) the power output of the charger is almost 90W. This means about 4.5W of heat dissipation with a darlington, which would require a heatsink and some path for the heat to escape.

That's not insurmountable, actually I'll probably build a circuit like that to test and see how much surface area the heatsink would need, but there do seem to be many downsides.

Would a crowbar with a PTC be a good solution?

[SeanB]

Crowbar and PTC would work, but then again you run into the issue of either you clamp short spikes and get spurious triggering ( and the PTC will likely keep the crowbar on until power is cycled, unless you really choose the SCR for high holding current, which makes it ineffective) or use a time delay, allowing the spikes through to kill the BMS.

Either way you can easily destroy the BMS on the cell.

[derGoldstein]

Yeah, either the BMS, or the charging circuit, or both... That's the problem I keep running into, both sides of the circuit can supply current, so a crowbar without a diode can fry either side.

This may be another type of overkill, but what if I put a polyswitch on *both* sides of the crowbar? When the circuit is shorted out, the first polyswitch to go off will be whichever side outputs the highest burst current, quickly followed by the polyswitch on the other side (because the PTC won't release fast enough, I think...). This way normal operation will see very little resistance across the the protection circuit, but it still has the ability to short in case of a spike. I'd still be left with occasional unnecessary triggering, though.

This could also deal with overcurrent conditions, but this doesn't deal with a continuous overvoltage condition (like if the charger keeps going beyond the 4.2 per-cell because it's calibrated slightly off). In that case I still have to either A) cut the connection, requiring a semiconductor in the middle or B) continuously convert the access energy into heat for as long as it's connected to the charger (I think I could do that with a zener that triggers a mosfet which then flows into a large wirewound resistor).