TeenyBMS, a small form factor 16S smart BMS

Guess a simple diode would do the trick, but its abit unsexy

Had the same idea and the same thought about it too, just a bit unsexy :smiley:

Well, I guess if we’re not charging with alot of amps, it kinda makes sense. Should be a fairly cheap solve tbh

I suppose something like this could solve it.

Need to figure out the biasing although :smiley:

Lets see who can solve it with the least amount of components :joy:

1 Like

And no ICs

4 Likes

Managed to round it down to 4 components, gate might need some help tho

I feel like a dead charge port might be nice, but there are a LOT of different chargers out there who don’t work with those. I have a pretty good balance charger who does not like it myself.

My vote is on live charge port and the space used for something else, like a small discharge parth.

Talking about overcurrent protection, would it make sense that if a overcurrent event is triggered, that it it would reset by disconnecting and reconnecting the charger?

Although as @Linesflag pointed out

This would not be possible with that setup…

Maybe the overcurrent event should trigger a small delay, stop charging and then restart?

Imo that’s what the fuse is for. What’s your target charging current?

think its hard to solve it all with just discrete circuitry, atleast if you dont want a vast amount of components. DieBiems does most of these things, but it uses a clever IC and an opto to detect the charger if i remember correctly.

With the blocking diode you still get the full bat voltage over the port if you remove the 10k resistor. leakage thru the diode brings it up to roughly Vbat. You can’t draw any current tho, then it plummets.

Wonder if thats enough for the chargers, probably talkin micro seconds.

Been looking at different hardware implementations and I should probably specify that I meant no programmable ICs. Gonna need some ICs to implement the wanted features, is the conclusion I also ended up at.

For fault cases I think I’ll use a 555/556 timer in monostable (one-shot) configuration to have triggerable delay for stopping charging, like in an overcurrent event or too high pack temperature measured via NTC.

Opinion about putting a fuse on the BMS itself?

It it’s easily replaceable? Otherwise I’d say no.

1 Like

Size wise I was thinking of a SMD fuse like this

https://www.digikey.fi/en/products/detail/bel-fuse-inc/0679L9200-05/5844045
But then the question becomes, that if it pops, how easily can people get a new one and it isn’t something that you can easily source from a nearby hardware or autopart store, so would it be better to not add one to the BMS itself, and let users use their own external fuse?

1 Like

I would not add one unless it’s easy to replace without SMT soldering, and space constraints make that very unlikely.

Even if it was easy to replace, I’d still say no. I’d prefer an inline fuse in the wiring. You don’t know if folks will wrap this in fishpaper or shrinkwrap and bury it inside a battery. In fact, for this form factor, that seems the intended use.

3 Likes

Yeah this. Makes the design less safe, but a fused charge port should be standard by now.

4 Likes

Cant you use resettable fuses?
Like a couple of these parallel https://hu.farnell.com/bel-fuse/0zcf0110af2a/pptc-resettable-fuse-aec-q200/dp/3864268

Look into “efuse” several smart silicon devices that provide the fuse functionality without actually beein a fuse. Only good for up to 30A or so tho

1 Like

Not gonna help, because the problem is that to allow the BMS to work with the chargers that expect to see a voltage on their output for them to start charging, you need to get the battery connected to the charge port and the FETs on the BMS in that configuration are only able to detect the overcurrent event in the charger → pack direction and cut it off, but not towards the charge port, aka charging port short-circuit cannot be protected against on the BMS.

One problem though is that I don’t own any of those types of chargers, so I can’t test how sensitive they are to start charging. How much voltage do they need to detect on their output? Or is it based on current flow?

Should the user be responsible to place an external fuse to protect against the possible charging port side short-circuits and call it a day?

1 Like

Yes. :smiley:

2 Likes