Having a dead cell in p group does not change it’s voltage, does it ? How would you tell something from the voltage only ?
It changes, but not right away, the voltage will drop faster in this group, and also the sag will be bigger
I didn’t get the problem with the BMS and cell level fusing, or are we talking about the main fuse blowing and all current count through the balance wires? If that’s the concern just add fuses to each balance leads, there are some smd fuses that are really small
Should inuse 2 12 guage wires or 1/2" tinned copper flat braid for the series connections?
I chose to use 2 12 gauge wires as I don’t have any braided copper wire. I read someone say that the braided copper wire also wicks a lot of solder making it rigid. I don’t know what the amp ratings are for the braided copper wire but when choosing a solution for my batteries I wanted to try to reduce the amp draw over the nickle strip so I put 2 pieces of 12 gauge wire between the cells so fewer amps were being pulled across the entire nickle strip.
No it would be better if the main fuse blew. But if all the cell fuses blow in one P group then the balance wires surrounding that breakage will be exposed to full pack voltage instead of 4.2V. the current is irrelevant (but inevitably more) and the extra voltage is what will do the damage. The components on the balance board portion of a BMS are not rated for something like 42V. And last I checked, there’s no such thing as a fuse for voltage 
There would be 0 voltage. The battery would be splitted in two separete pieces.
So rather burn the whole board, and not the BMS?
Actually, if a load is connected to the battery (like a vesc) then the two P-groups that have been separated effectively become the new + and - terminals of the entire pack. So the voltage will not drop to 0, it would drastically increase. A vesc always uses power even when idle and is therefore a load.
I would argue that allowing something to internally short your battery is far more of a risk, which is the only time you would actually need cell level fusing. Why not just seal the pack?
I wasn’t planning on using nickel strips at all for series connection just the 2 12guage wires. I’ve seen others use 2, 14 guage wires for a 5p set up.
… yeah but is going from 4.2V to 50V inherently dangerous if, as you say, current remains low?
What is the internal resistance of the BMS balancing circuit? And what is the internal resistance of an esc with no throttle applied?
Well, unfortunately the most likely time for breaking a series connection is when you’re pulling a lot of current through the vesc, which is what would blow your cell level fusing (theoretically). So that resistance I would say is negligible.
It doesn’t matter what the resistance is between those contacts is. normally, it’s infinite actually, when the cells aren’t being actively balanced. But the voltage is far exceeding any spec for components, a transistor can only prevent a certain amount of voltage from passing through before it blows up.
So yes, going from 4.2 to 50V is inherently dangerous. You may not start a fire with that, but at the LEAST you’re going to brick your bms.
Yeah also remember the ratings on these wires are at way longer lengths, we use mere inches of wire and run short burst of high amps not really constant. My battery will be capable of 160 battery amps constant but I’ll never encounter a situation where I’ll use 160 amps for more than a few seconds at a time.
There is – it’s called a “spark gap”
But you need to be in the thousand-volts-plus range to use it
1 Like
Well, true. But AFAIK that’s usually for ESD in sensitive electronics and is really only effective for really high voltages
E: well you beat me to it, damn
1 Like
No. it won’t. It will break. The components on the balance leads are not rated for 50V. Everything else you said is completely irrelevant due to that fact.
I suggest you convert your 50V esk8 charger to have a USB output. Go ahead and plug it into your laptop, see what happens.
A BMS will be damaged by a series connection break within the battery if there is a load on the main battery terminals, full stop, end of question, no other possible outcome, period.
Here’s an explanation better than I can do:
More evidence: How you blow your BMS when you don't first disconnect all the loads | DIY Electric Car Forums
But you haven’t explained how in a real world scenario there would be a load on the main battery terminals? Which is exactly what the rest of my post was about (i.e. not irrelevant).
Using your example of a 50V battery capacity meter. If you put this in series on a 50V circuit, the voltage would be 50V across the battery meter, yes.
Insert two leads of a BMS in series into that circuit. Nothing will happen. The internal resistance of the battery capacity meter is so high, it’s effectively as though the circuit is broken. It has to be this way, otherwise it would blow up in everyday use. V=IR
I don’t know what would happen if I put 50V across the usb terminals of my laptop - I’ve never tried it. But is it really the slam dunk you’re going for? Would the laptop actually complete the circuit? Regardless the point is irrelevant - the USB circuitry of a laptop is not a battery management system designed for use with a 50V battery. And before you get all sassy on me, yes I understand a BMS is not designed to handle 50V through its balancing circuit. Just… redundancy etc etc… I’m not saying the internal resistance of the BMS is high enough to withstand 50V at 80A… but at 1A, would the current be strong enough to overcome the resistance?
An ESC with no throttle must have a fairly high internal resistance. I appreciate people disconnect these when not in use because they always draw a current.
Lastly, maybe dial down on the churlishness? This is a discussion in a thread for people asking for help with battery packs. I’m not claiming to be an expert. There are plenty of people who seem like experts in this area who are using cell-level fusing. I’m only trying to work out whether I should bother, because it does seem like a lot of hard work.
If a vesc is connected.
If there is a break in the series connection inside the battery, there would actually be 0V across the battery meter. Did you even watch the video clearly explaining this?
This does not affect anything. The resistance can be as high as you want, but so long as it is not infinite, you will read voltage across the load. Try taking a 100k resistor, in series with a multimeter, touch the other end of the resistor to the positive terminal of a 9V battery, and the negative of the multimeter to the negative of the battery. You will read 9V. Even if it’s a million ohms, even if it’s zero ohms, if there is not current flowing then you will see full voltage.
Yes like you said right after this, this portion of the BMS is not designed for 50V.
There is no redundancy in this part of a BMS.
The current would be 0 amps in this scenario (until it possibly fails closed or something). Current doesn’t matter, it is still being exposed to high voltage. Transistors, especially small ones like these, are VERY sensitive to overvoltage.
The resistance of this does not matter. So long as a voltage can pass through, that’s enough to damage something.
I’m no expert either, but I know a thing or two about electronics, enough to make my own BMS from scratch if I had to.
Yes, which is why I’m trying to explain how people making battery packs should be cautious of this.
If for some reason your fuses blow inside of your battery, don’t assume that the BMS is functional. You could end up causing more damage by using it again.
Know the risks. You reap what you sow.
I’ve explained the risks. It is okay if you don’t comprehend WHY this is a risk. Just understand that it IS a risk. I’ve done my best to try and explain this in simplest terms, and I can’t do more. The youtube video is a GREAT simple explanation.
just because you don’t understand something doesn’t mean it’s okay to potentially misguide new users who want to learn. It is a risk. BMS’s have been broken this way. It doesn’t just apply to cell level fusing, it applies to bms’s in general. it’s simple. Don’t disassemble your battery while the BMS and the vesc are connected, simple as that.
Anyways, this is getting too off topic. I’m done discussing in this thread, if you’d like to start a new one just for the sake of arguing a moot point be my guest.
Cool, thanks. Sorry if I’ve misguided anyone. I’ve been careful to frame my comments as questions for this reason.
I think I’ve more or less figured out my last question as to why this is specifically a risk for packs using cell-level fusing. In packs which don’t use this fusing, a bad 18650 will effectively act as a short for the others in that p group… eventually the whole p group will go bad, and just act as an extended series connection between the p groups either side. Is that correct?
So… cell level fusing - you are suggesting the risk outweighs the reward? Anyone advocates for cell level fusing care to weigh in?
Cool, I’d be keen to for the cell-level fusing discussion to be moved to a new thread. Could be interesting.
I imagine the discussion is going to get side-tracked often with specific battery details like this though, going off the many attempts at “general” battery building threads I’ve read through on this forum. I’m at least a month off starting any building so no pics just yet sorry.