That’s more based on whether your accessories can take the battery pack voltage directly, as @ShutterShock says, his lights are directly behind the pack voltage, so they most likely have their own built-in voltage regulation.
1 Like
Not always, I’d rather just have a switched discharge port and be able to put a buck wherever I like, rather than having the bms be larger.
Yes that’s right
For my other version of lights I was testing, I needed buck converters though, which are still connected to the discharge port, for making sure they will never kill my battery.
1 Like
I’m thinking on the same line myself on the issue.
2 Likes
Here’s a quick mock-up with no traces routed yet of a 18x65mm (18650 footprint) size PCB with 12 sense/balance circuits placed, just to check space wise is it doable and I’d say it is. There is also space available at the end for all the discharge/charge path control hardware. Probably gonna be a bit of a squeeze, but I’ve been squeezing a lot into PCBs before.
What kind of sense/balance connector would you want? 13-pin JST-XH? and probably solder pads for all the charger, pack and load connections?
7 Likes
hm, if one would skip the discharge port but keep the charge port.
It could end up purdy small.
some things i find useful that are easy to implement:
Dead charge port (you cant short it when charger is not present)
A series NFET for a simple polarity protection.
Just tossing this out there, but maybe make w/e charge connector+PCB like a panel mount?
That way the PCB only occupies the charge slot input. If that makes sense.
5 Likes
yeah nice. Then if you want a discharge stage it stacks on top.
Personally I’d wanna see the same one as Daly for simplicity
If this saved me soldering on a fuse holder for every dumb bms build that’s a no-brainer, I didn’t even know that was a thing you could do
The fact that the board he drafted there is 18650 size and has most of the basics already is impressive haha
3 Likes
I did some checking and there are multiple different connectors used in the different models from Daly, so would you have a specific model in mind?
I’ll look into the fully blocking FET setup to enable this. Mainly just need to check how to bias it correctly with a compact, non IC solution. So… charging FETs open when the charger is connected / voltage present at charging port and open when voltage disappears from the charging port… Might cause a problem with the particular chinese chargers that need a voltage present on the output to start charging.
Gonna need space for connectors for that. I’m not completely against the idea, but I need to develop the base PCB a bit further first to see how the layout develops.
2 Likes
I didn’t even realize that - I would think the size for 12s would be the most common?
1 Like
After more digging, it looks like they are JST PH -series (2.0mm pitch), aka the same connector as the VESC’s I/O-connectors.
Just to keep the footprint as small as possible, I guess you could place the connector such so the connector header and wire side housing don’t extrude so much out to the side.
9 Likes
This makes the most sense to me since a lot of us are stocked on 2.0 crimps
My thoughts on dumb bmsis that I wont ever use one that doesn’t have cell SOC readouts accessible.
Ive seen plenty of batteries on here get saved b4 perm damage was done because cell readouts, I myself have saved a pack because of your flexi.
I could go without all the perameter settings, but not cell readouts.
4 Likes
The older ones were standard, the newer ones are locking
1 Like
I’ll get a photo when i get to work
JST 2.54mm pitch locking connectors. I’m pretty damn sure it’s XH, but have to check in the morning…
Something along the lines of these:
(Metadata included so my filters apply)
It’s the pitch&connector that basically all LiPo balance chargers use w/their silicone cables…
There is however a BMS that floats around this forum that has a 4pin connector & 9 pin connector separate to plug in… I think some refer to it as a D140, but unless my memory is failing me (it is) the D140 was different & has a singular 13 pin JST connector of the favored/more common pitch. Or there are subtypes.
Anyways, the 4+9 balance connector D140 uses the wrong pitch. It is slightly fatter than the usual . We do not want that… genuinely fuck whatever engineer over there that made the choice to use the chonkier connector.
I would also be a fan of having screw down terminals instead of solder pads for the main bms wires… I know that sounds dumb for esk8, but a lock/nord washer & blue loctite should be more than enough to hold down a ring terminal under the vibrations we experience.
Assuming PWR_IN is charger+ and PWR_OUT is pack+
How do you stop current flowing from the charger to the battery with that circuit?
OR
Assuming PWR_IN is pack+ and PWR_OUT is charger+
How do you stop current flowing from the pack to the charger port with that circuit, when the charger is disconnected, aka the charging port is therefore not “dead”?
AFAIK, JST makes only metric pitched connectors. JST XH -series is 2.50mm pitched, not 2.54mm (100mil) pitch. JST XA, is the locking latch variant of the XH. I know this because I have XH, XA and PH and PA connectors in my own shelf.
1 Like
You dont, its simply just a polarity protection 
if positive is present on GND, FET is not conducting why would you want to stop current from charger? 
(assuming you’re using the appriopriate charger)
edit, ooh you mean for the dead charge port function? yeah that needs abit more circuitry
Yea, been thinking about it… Obvious would be a back-to-back mosfet setup, but just need to think about the biasing of it… and then figure out how to get it to work correctly with the charger plugged and unplugged…
Yeah, should be doable without a fancy IC controlling the back to back switch tho. I’ll play around abit in LTspice see if I can solve it