Well the Unity is dead, and thats the only ESC with an eswitch I know of. I know how to install an external anti-spark so thats not the issue. I want to know that if one of my p-groups undervoltage that the BMS will save it. Can I do that without limiting the discharge?
IMO discharge BMS provides very important safeties. Short circuit and cell low voltage protections are the ones that pop out at me. But, the way BMS implements protection is to cut power, which is far from ideal for eskate. So pick your poison, at least for now. Hopefully soon weâll have BMS with communication that can tell ESC to ramp down.
5 Likes
A little speaker that makes a horrible sound when a p-group falls undervoltage would be nice. Shorts can be made safe with fuses.
1 Like
You have valid points which I didnât mention. Easy solution that I see here is to setup low voltage cut off a little bit higher.
@jack.luis bms will balance cells during charging, discharge is just separate circle without protection.
Exactly, so if one of the p-groups is undervoltage it will allow the pack to continue discharging, permanently ruining that p-group
1 Like
Well if one group goes so much lower than others in each charging circle, I would consider replacing it ASAP
I wish it was this simple. Cells degrade over time/use, so behavior of p-groups will change over time. Since voltage drops in a knee shape when a cell is empty, itâs really harsh on the cell to over discharge (possible fire?).
Iâve been kind of banging on the âwe need cell level monitoring not BMSâ drum here and there. Knowing whatâs going on in the pack seems to be the most important thing. So many BMSs fail and kill a p-group or a few⌠And well built packs with good quality cells tend to stay in balance. And charging to 4.1v/cell can double cycle life.
This depends on the age of the rest of the pack, right? If you pop one fresh group into a 300 cycle pack, I dunno if that helps or hurts. @tinp123 did you change your avatar?
3 Likes
True, true, everything you say is true. But @jack.luis will have BT so he could check if any group drifts. If user takes care of his battery, I donât see big risk using charge only bms.
And believe me I know how it is with bad bms out of wrap. You rarely see it immediately after installation, thats biggest problem.
I did not change my avatar, I have different one on old forum.
1 Like
been reading this off and on while working and waiting for things. just a few points iâd like to make based on experience.
- if one pgroup drops below the good zone LVC in the ESC wont know or care
- If one pgroup drops below the good zone the BMS will know and shut it down.
- if you can catch the lower capacity pgroup before all the cells in it drop below the good, check for broken welds and fix them, if that was the issue the group will rebalance itself out and then you can use a balance charger to get it back in line with the other groups.
- if one pgroup drops below the good zone it must be removed regardless of cause or cost or inconvenience
- if the cycle life of the entire pack is less than say 10 charges you can drop new cells in there and replace the pgroup with little or no consequence.
- If the cycle life of the entire pack is higher than that I wouldnât do it unless you have an external balance charger, youâll be fighting drift forever because
- most BMSs suck at re-balancing and one groups in your pack now has a slightly different capacity
- most BMSs suck at handling regen voltage spikes in a way thatâs freindy to the rider
- a BMS wonât prevent cell death within a pgroup, just keep you from using the pack while it dies slower
- discharge protection events while riding can be lethal
- over charge protection events while braking can be lethal
So, to sum up, in my opinion, fuck BMS protections on the charge/discharge path between the cells and ESC and if you have a dead pgroup in most cases you should rebuild your whole pack with new cells unless you lucked out and just had broken welds on otherwise good cells.
22 Likes
@longhairedboy , are you not worried about those braid connections in this?
https://www.instagram.com/p/B7g40kTpawI/
I feel like @thisguyhere had an issue with that style of connection breaking
If you look at in the enclosure, you will see its not to be used as a flex pack. So no problems.
2 Likes
awesome post!
Iâm one of those ornery bitches that likes to do things his own way. The direction Iâm heading is, bring the balance leads out to a multipin connector (I think SP20/SD20) and use an external balancer like this (It charges low cells from high cells. I have one. Itâs awesome but slow since it deals with 1 pair of cells at a time, but can do any series number and has a nice app. Also there are a few different models with different number of amps).
Bulk charge to 4.1v/cell unless I need the capacity that day. Check balance once a month or so, balance if needed.
In the future, I want to buy (or heaven forbid spend 5 years making) a monitor board, with jst passthrough, canbus, maybe a speaker, maybe a display, maybe bluetooth.
EDIT Iâm also considering building that frankencharger thing, with isolated TP4056 for each p-group. CC/CV charge each group separately. No drain-to-balance.
2 Likes
Amen! This all day. A battery pack is not worth the cost your health and thatâs exactly what discharge BMS does, cuts your power unexpectedly. Less of that the better.
not in this specific application. I wouldnât use this style over the segment wall of a segemented enclosure though, youâd want superworm with some slack there.
3 Likes
The middle of the braid has no solder or tin, so it can flex just as much as pure copper wire. The issue is those braids tend to soak up solder and turn stiff, but looks like that was done tight and not wicked all the way through.
1 Like
Aaahaha fight!
8 Likes
Iâm not a battery builder, but Iâve been soldering for over a decade now. Iâm sure Damon has been soldering for longer than I have.
I will however refer to @Arzamenable to voice his feelings on superworm after having used turnigy.
I think superworm wicks a little more than it should. This can be solved by using a hemostat style heatsink, but that adds a bit of time. Turnigy is also undeniably more flexible.
4 Likes
maybe I should have said âflexible high strand count tinned copper silicone insulated spaghetti wireâ
I donât actually buy the superworm brand wire, I get it on 50-200 foot spools from amazon from a company called BNTECHGO.
6 Likes
You can see here the way off soldering the copper braid better. Due this the pack can still flex a bit.
12 Likes
What BMS is that? Does it balance?
2 Likes