Uuhh why are there 2 leads to the same V potential connected? Talking about the far end of your pack.
Maybe he has it wired as two 6s packs for charging, for some reason
Thats makes sense. Maybe he had to 6S laying around?
Any reason why you would want this over a smart BMS?
I thought a lot about BMS options, what job they are expected to perform, why we use them, etc.
What we expect a BMS to do:
- Disconnect if we overdischarge a pack or a p group.
- Disconnect if we overcharge a p group.
- Balance the cells as needed.
- Display cell voltages so we can take action if there is a big problem.
Reality:
Overdischarge protection:
- Using them on the discharge side is risky as our brakes depend on working connections. If the BMS either fails, or more likely cuts power because of some detected event, it can fry the ESC and/or prevent you from stopping. It seems like many people use them for charge only for this very reason⦠but this also means that they are basically using them as a cell monitor rather than a battery protector.
- What happens if p group 5 goes under 3v on an unbalanced pack but the rest of the pack is above the VESC low voltage cut-off? Ideally a BMS would tell the VESC that a cell is undervoltage and to initiate a reduced power mode anyway. As far as Iām aware, we donāt have a BMS that will do that (yet).
Overcharge protection:
- They actually work pretty well for this purpose. It will disconnect the charge if the pack or a p group is over voltage.
Balancing:
- Passive balancing BMSs only balance at the top of the charge. If I deliberately undercharge to say 4.15v to extend calendar life, they wonāt balance.
- Active balancing BMSs will balance all the time, but - the voltage reading isnāt all that accurate. If the BMS thinks that two p groups are off by 150mv (typical accuracy of an aliexpress Smart BMS), but they really arenāt, itās actually going to introduce an imbalance to a pack that may have never been out of balance in the first place.
- Packs made with new cells that are balanced before assembly likely wonāt drift out of balance in any significant way. Anecdote: After about 50 cycles on my old pack with no BMS, the voltages were almost dead on. My old pack was rough. Soldered cells and no parallel connections except at the battery +/-. Yet even with no parallel connections, and potentially heat damaged cells, all 48 cells were within 0.01v until some of those series connections failed due to poor pack construction. I think with new, high quality cells and solid battery connections, the cells really shouldnāt drift.
So if we donāt use the BMS to protect the battery during a low voltage p group situation and we may not need balancing, then why do we have one? To protect against an overcharged p group? To view individual cell voltages on our smartphone to see if something needs to be done? The $25 chargery will do that, with an added benefit of audible alarms if a p group drops under a set voltage. The voltages could be verified with a multimeter and manually balanced as needed.
The only drawback I can think of, is that I need to be within earshot of the board while itās charging, or at the very least check that the cell voltages are balanced at the start of the charge by looking at the screen before walking away from it.
However, these chargery units are not well tested on the forum. While it sounds good in theory, time will tell.
If we had a BMS with the following features, Iād use that instead:
- Charge only
- Communicate with the VESC to enter a reduced power mode when a p group is below a set voltage
- Configurable active balancing with ~10mv accuracy. Set it to not act unless p groups are more than X volts apart.
First of all, I would wager that 90% of us do not use the discharge function on BMSās nowadays, so no one is really depending on that. I guess that writes off half of the functionality.
We just trust the vesc for cutoff, and set our cutoff at a safe voltage like 3.2 to avoid issues due to sag or over discharge.
Second, many of us use LLT smart BMSās with a bluetooth module in order to be able to set and view our settings. In this circumstance, we are able to set a balance start voltage, from which, the pack stays balanced. For example, on mine, I have it start balancing at 3.6v, and then the charge cycle ends at 4.15. Most of the time it is done balancing pretty shortly after the pack is full
We have questioned the necessity of a BMS in the past but always just fall back onto using one it seems
Would rather just have one in there to keep it in check rather than be opening up the enclosure all the time to probe cells
I know nothing about the charger you are proposing but it seems interesting, will follow
Where exactly are you thinking this screen will sit?
I wont comment on the merits of your decisions (because I feel unqualified to do so) but I really like the thought you have put into this. Iāll be interested to see how that unit works out for you!
Does it sound an audible alarm for cell under-voltage as well? i.e. is there a chance that if a group is out of balance during discharge, I could hear the unit making noise and know to stop riding?
eBay will have some protections. But I ordered a Kweld and it showed up incomplete after I had to wait for his restock to ship them out, even tho when I placed my order he said he had them in stock. Now, heās been ignoring me about the missing parts now for over a month. So I have a useless second welder⦠wouldnāt recommend.
I ordered from him pre-covid, hard to speak for it now, probably should have mentioned that lol
On second thought, who in the world would ever order a kweld and not need the screen module, stupid that it even left the warehouse like that
My first order precovid wasnāt bad⦠this one tho. Fuck.
Iām thinking it will be inside the enclosure stuck to a polycarbonate window on the bottom side for viewing. It wouldnāt be visible while riding.
Yes, I think this is one of the big benefits. A 100db alarm will sound if any p group drops below whatever you set it to. You can also set a timer, to say that the voltage has to remain below that set point for X seconds before sounding the alarm. You can also rig it to turn on alert lights or other stuff too, but Iād probably just use the alarm.
Thanks for sharing your BMS thoughts!
Would you mind sharing in more detail why you prefer active (charge transfer) balancing? Itās expensive, large, and IMO doesnāt offer any benefits for the small packs used by the esk8 community. If you have a huge high power pack there can be a benefit to the lower power losses you get via active balancing though.
Both active and passive (resistor bleed) balancing can be accurate (or not), can occur at any voltage or time, and can offer the same protection. It only depends on the design decisions of the manufacturer.
Itās a side noteā¦I would also add over- and under-temperature protection features to any BMS. IMO they are critical.
I think I used the wrong terms with active vs passive. I think the resistor bleed style is fine as long as it can be configured to engage at any point, like start at 3.6v and end at 4.15v in the example above. Iād also want to be able to configure the amount a cell has to be off before the balancing is engaged, 0.02v for example. That would prevent the BMS from needlessly balancing cells that may āsettleā back to a balanced voltage, and address some of the voltage reading inaccuracies.
Agreed! Good features to have.
Thereās a place for a low-feature-count BMS that does one thing and does it very well but if itās going to be an adjustable full-feature BMS then it should be a full-feature BMS.
he paid someone $3K to make a website but got scammed⦠so take what you will from that.
shopifysayswtf.com
anyone have serious problems with this pack design? hmmm Pgroup isolation for one.
went in this⦠hope they store it safely.
With speaker wire⦠or @least that what it looks like to me.

start at 3.6v and end at 4.15v
Starting balancing to early can be counter productive as it rely on every cell having the exact same capacity to a mA. Cells are just not that consistent. If you start it early you will fined as the pack gets to a higher state of charge it will have to undo the balancing it all ready did

configure the amount a cell has to be off before the balancing is engaged, 0.02v
I want a few more settings than that
Charge cut off max voltage
Charge current termination
Minimum pack voltage to engage charge
Max p group voltage
Min p group voltage
Max charge current
All of these are bare minimum if your following manufacture spec and some need changing depending on what cells you have used or charger

speaker wire
Speaker wire is pretty thick copper, so no objections there, but arenāt the main leads kinda thin?
In my experience, the transparent insulation on speaker wire makes the conductors inside look bigger than they are. That whole pack looks awful, speaker wires included.