It can’t control individual amps on P groups, however, it can monitor their voltage. So discharging using a BMS has the following extra layers of protection:
Offers protection against short circuits: a short circuit will cause high Amps to be drawn from the BMS and that will cause it to shut down immediately. A fuse might do the same, but a fuse is sometimes much slower to react
Offers protection against low voltage on individual groups: if for some reason one of your parallel group discharges at a faster rate than others, it will shut down all the power once that particular group reaches the BMS over discharge protection voltage. This could happen if one of your cells goes bad for example.
It will protect the pack from overcharge: if you start at full charge and go down hill for a long time, the bms will cut out the power due to the overcharge voltage limit.
That’s basically it. It’s also why a lot of people (including myself) prefer to bypass it for discharge: you can use a small one and you don’t need the extra room in the enclosure. The over discharge protection is done by the VESC (not on a per group level but on a pack level). Same goes for protecting the pack from pulling too many amps, the VESC limits that (with the battery max setting). Over charge during regen is still a risk, same when a parallel group goes bad: unless the overall voltage drops dramatically, the VESC will have no clue it happened and will carry on.
Also, a BMS balances when a cell/group reaches 4.2v. I’ve seen people suggesting you will save some charge cycles on the cells by only charging to 4.1v. While that is true, if you do it too many times and your cells/groups start to drift, the BMS will never have the opportunity to balance them, since you never allow the voltage to reach 4.2v.
There has to be be a way to solve the BMS issue altogether. I want it… but outside the board or not at all.
No BMS type solutions (and no embedded in your build type BMS)
Option #1 - use a 12S charger like the EVPeak A9 – yes it was about $200 but it is rapid and well worth the investment. The only issue I have is that I have my (2)6S cables that tie into the 12s charger’s balance board hanging out of the case (in picture below I don’t have the enclosure on but they would go into the 3d printed holes for the xt90 and the other rectangular hole/bracket to the right of it).
Option #2 – Why not use something liike @Nowind cell pack monitor to charge and keep the voltages the same. Balance charge when they differ greatly.
Option #3 – put BMS leads outside the board like I have but have the BMS on the outside of the build then hook balance cables only when charging. This saves you from buying a 12S balance charger.
None of my ideas are gr8 because well how do you deal with portable charging? Maybe #3. There has to be a way to get rid of the BMS inside the build… it’s a space waster and if you do like 4-5 boards… it just gets annoying… each build has to have an AS switch, each a BMS, each an x and a y and a z. We need to find solutions to delete it from our builds. Maybe an engineering solution like a circular balance wire power plug that leads the balance cables outside to the BMS, Inside the circular plug are the (12) balance cables. The input and output of the power plug re-extracts the 12 wires. And the BMS is next to the power brick charger. I think that would be a cool invention. Then it’s just two quick plugs to plug in – both shaped like circular power charger ports. You only need buy one BMS + charger could solve all your builds! This shit would literally cost like $2 to manufacture.
Option #3 is what i’m using. a circular plug that has all the wires, and i’ll just plug it in my external charger.
But that brings problems, like OP said. Risk of individual cell over-discharge despite ESC safe limits.
I had that happen when i don’t have any BMS on board. I ran out to about 33V(10S pack), which in reality was one cell zero volt.
all that does is shift money around, and not in a good way. A good BMS and charger costs less than a multipin milspec connector ( to send all your balance leads to an outboard balance charger) and an outboard balance charger. Especially at 12S.
You’re still doing the same shit just on a different day.
Ok but we can change the world… the BMS doesn’t have to be in every one of your 10,000 boards right?
The component that @Sebike has linked + combined with Jenso’s cell monitor lets you from having 10,000 bms’s right for your 10,000 boards? Now it will look ghetto with the BMS and charger on the outside… I get that part.
I suppose we could also switch a single drivetrain to different boards but that is not the scope of this thread. We need to be implement “best practices” now and avoid being over regulated in the future.
For science i just charged up my 30Q 10S3P to 41.50V, the BMS is set to 4.125 end/balance voltage. I will time how long it’ll take to bring down to 41.25V. Considering it has to bring the whole pack down it’ll give a good estimate in terms of max time to balance.
We are severely lacking a quality programmable high-amp BMS. JTAG’s looks so promising. Hopefully they become more readily available. Especially since BattMan is effectively dead
I totally agree on this point of view. For this reason I believe IC bmss are in a whole different league.
For example there was an old firmware for @JTAG’s bms where a “bug” allowed to balance the cells WITH THE CHARGER DISCONNECTED.
Just imagine how cool would be to have your cells start balancing whenever you like with or without charger. This is Tesla kind of stuff…
If it’s between one of p groups going below voltage and power cutting out while on the throttle. I’d rather let the p group go low voltage and let the pack potentially get damaged. Cells are rated for 2.5 volts aren’t they. Rather save some skin
I would rather manage my cells properly and save my skin and my cash. Of course most/some of us know there is a cliff at 3v, once a cell gets there it plummets unpredictably and they turn into an expensive radiant heater. Good for warming all that skin you save on these cold nights.
Why is it BMS mfgs don’t describe exactly what their devices do. So frustrating having mystery black boxes interfacing with the most dangerous and expensive part of our boards. (If I’m wrong and there are docs, plz post a link, I’ve yet to see docs beyond some basic basic specs)
Float charging is bad for lithium ion. Our phones and laptops don’t float charge. If BMSes need hours with power to the cells to balance the cell groups, I’d argue that’s another strike against BMS. If BMS can keep balancing after charger is removed, that’s better for the cells, but makes me worry about malfunctions.
In theory cell groups should not be far out of balance so time spent balancing should be minimal. But we need some indication if things are not going well. Things break, cells wear out unevenly due to imperfect pack builds and heat distribution.
If BMS had a display or at least some flashing LEDs to indicate exactly what was going on that I could bring to the outside of the board, I’d feel a lot better.
To this day I don’t understand why people insist on having a BMS at all - it seems so unnecessarily complex and expensive plus is an additional part that might fail. Especially with larger p-groups the chances for drift should be even lower than on smaller batteries - why should a BMS be more important on a large pack then?
No BMS is the way to go for me - check your cells/groups after a ride and you’ll be fine and even save some money in the process. Seems like a good hourly rate if you save 100$ on a BMS for a 10 second check with a lipo buzzer or whatever.
curse you though for only starting this thread after I bought 2 charge only bms :-/
now I went from feeling pretty secure about doing a good build to having bad dreams and not being able to sleep!
