YES PLEASE
2 Likes
That sounds pretty awesome. I think I would be shot of I gave something too big. I think some battery builders would have a good idea of what is realistic size. I would love something between a 3-4p 18650 group. Iām sure @longhairedboy @psychotiller @Winfly etc could probably chime in on dimensions for the packs/holders that they build.
within a 140mm by 60mm rectangle would be acceptable.
it needs to have/do the following in order to make my pants uncomfortable:
- active balancing, not just half ass balancing. If a p-group goes over 4.185v divert power to the lower voltage cells and bring everyone else up, and shut off the charging for anything at the desired charge level
- optional discharge bypass through the fuse of our choice, preferably a compact fuse type in the 50 volt range that will actually fry at 70 or 80 amps instead of cooking way before that
- built in eswitch with at least 6 massive fets that can also serve as over discharge protection when enabled. Yes the unity has one, no the og focboxes do not, plenty of ESCs donāt. Iām not talking about a temp switch trick, iām talking about support for halo push buttons with the option for an always on jumper in case we do have an ESC with a switch
- port for a fuel guage to plug in
- 5v accessory port of reasonable power levels. I need to light up the world with it.
7 Likes
Perfect for overnight charging or charging at work/uni 
1 Like
Added onto this list, Metr Functionality like the current DieBieMS has, and Canbus connectivity for sure
2 Likes
I think the diebiems almost has all of that built in except the fuse, active balancing and USB-PD. Even the size lol.
I think these are the desired changes to be added then
That and just make it super compact. Itās usually size that holds people back from using it currently (and availability). Also it works with metr already @totalgeek9224
1 Like
But there is no need for active balancing xD. Especially not in the sizes we are in right now (and when using quality cells). And active balancing is really expensive / takes lots of space 
.
8 Likes
Ohhhh my bad hahaā¦ Itās late here.
I think the bypass option would be great too but itās easy to read wire anyway.
@jtag thatās fair enough. The more I think about it the more it doesnāt seem to phase me when we talk about active balancing. As long as everything balances properly when fast charging.
If size could be the same or smaller than a 4p pack it would fit any kind of board, staked makes it more expensive but think it is worth it if it can make it not being so long.
Iād like to add to the list, 13s.
3 Likes
How can I give this all of the likes? And 13sā¦ and a partridge in a pear tree.
2 Likes
Exactly why I am not entirely surprised we donāt see much of them. No free lunch, etc.
1 Like
my perfect bms would be 4p 18650 or unity sized, fused, Metr Functionality.
with option for a partially populated charge only variant to use with Unity. Unity could be configured to do low power cutt-of based on cell voltages received from BMS. no point in doubling up hardware 
What about isolated DC-DC converter, or like few of them, that steps battery packās voltage down to 5V, switches it through transistors to reach multiple cells and charges the lowest one(s) with 1A? 50V >> isolated DC-DC 4.2V CV and 1A CC >> multitransistor packages to switch into different cells >> lowest cell(s) This way instead of wasting all 100% of energy into the heat, you recover like 60-70% There are these tiny isolated Recom DC-DC modules that are like 12x8mm so donāt take much space. I think you would not need more space for that than you already use for all the high power resistors in your diebiems where you canāt even use all of them anyway because of thermals
Most often when a battery needs balancing a single cell is to high preventing further charging, with the method you describe a single DCDC will have to charge multiple cells, which is ofcourse possible but takes some effort on the algorithm and possible some extra time, the biggest challenge is designing a DCDC that accepts a wide input range (because the bms should already work from at least 6S and hopefully in the future work up to 18S).
But I might ofcourse be missing something and there might be DCDCās around that are capable of doing this, than please share them and wāll give it a go!
1 Like
I got everything set and working but while making some changes, i broke it( the cell node part of the circuit for 3 of them) The voltage readings were crazy(for the first 3 cells). Debugged to found that i had destroyed few of the R36 resistor. replaced, it got a lot better, but not quite. The voltage readings were still a bit(0.2 volts) off. Also noticed that it changed with temperature. After probing the board at several places found that I had corrupted CP voltages. It was because those vias had increased resistance(like 2.5ohms, while the good ones were 0.2ohms)). Not that i have designed a lot of circuit in my life, but @JTAG should the vias be beefed up or doubled given that they feed the balancing resistors?
I have ordered some pcb drills(0.3mm) to reconnect the vias. Letās see.
All cells above are at 0.08 volts(measured with multimeter)
It could be that they indeed got damaged, having 2.5ohm will indeed be bad when balancing is active (which in your case it is).
I will look into beefing it up a bit, in cases like this where something went wrong with wiring always something will fail to death, beefing things up will likely move it to the next weak point.
2 Likes