bluetooth BMS?

[Merlin]

jesus.....calm down dude...

if cells are IN BALANCE....you dont need "balancing"...
for sure we dont want to check all cells after every discharge, so a BMS is a must. priority is at first just safety. some videos are online what happens overcharging 18650.....they dont burn and are safe?....check it out...
if one cell starts to die, your "parallel" group starts to be weak because capacity is less.
iam no friend of stupid bms that even dont have LEDs on it that you never know whats goging on or how is your "health" of your battery.

so these BT BMS are the right step in the right direction. Adaptto was ...sry...atm IS the only one i know that shows your parallel groups cell voltages in your riding display. the "Pseudo" BMS from Adaptto is crap...but its the best way to check your cells and catch the early bird if something goes wrong.

anyway...izeman talks only about the difference that THIS BMS offers to balance "all time" or just when connected while charging.

 

[ElectricGod]

jesus.....calm down dude...

if cells are IN BALANCE....you dont need "balancing"...
for sure we dont want to check all cells after every discharge, so a BMS is a must. priority is at first just safety. some videos are online what happens overcharging 18650.....they dont burn and are safe?....check it out...
if one cell starts to die, your "parallel" group starts to be weak because capacity is less.
iam no friend of stupid bms that even dont have LEDs on it that you never know whats goging on or how is your "health" of your battery.

so these BT BMS are the right step in the right direction. Adaptto was ...sry...atm IS the only one i know that shows your parallel groups cell voltages in your riding display. the "Pseudo" BMS from Adaptto is crap...but its the best way to check your cells and catch the early bird if something goes wrong.

anyway...izeman talks only about the difference that THIS BMS offers to balance "all time" or just when connected while charging.

Except they don't STAY balanced...hence the need for balancing. You have 2 options...
1. manually balance periodically, which is a giant pain.
2. automatically balance with the aid of a BMS which is painless.

I think I choose number 2. And a smart BMS that tells you there are problems is defenately the best plan.

 

[madin88]

not needing to balance is the biggest lie being spread around. sorry, but if that was true then why does every single battery that comes out of a factory has some form of balancing built in? either something as simple as a top balance bleed resistor on some low end power tool right up to tesla's active power distribution from texas instruments that is REALLY expensive shit.

even so for us, why the fudge would you spend hunderds or sometimes into the thousands for cells and not spend the 50~100 bucks it costs for a half-decent balance board? why play the "should-and-normally" game with a grand worth of batteries?

If we talk about a properly build battery made of 18650 cells than it is fact that balancing is unnecessary. I can tell you many examples about 3 or 4 year old battery which got never balanced (because i have set balancing voltage a bit higher as max charge voltage), and all stay within a few mV.
Properly means:
all cells from same charge (with no voltage differences)
connections which ensure equal current flow through all cells

HV and LV cut off for each group is a must, but balancing is just a nice feature IMO.
Beacuse if your battery starts showing a drift, something is wrong and you should repair it. No BMS with balancing can "repair" a faulty cell.

If you have a battery made of large cells, like 20Ah or such, than i agree that balancing in necessary beacause there can be much higher variations in capacity compared to a 10p 18650 battery for instance where all cells are from same charge.
Tesla uses about 7000cells with single cell fuses which is by far a differnt story compared to most ebike packs made of 50-300 cells.

 

[flippy]

If we talk about a properly build battery made of 18650 cells than it is fact that balancing is unnecessary. I can tell you many examples about 3 or 4 year old battery which got never balanced (because i have set balancing voltage a bit higher as max charge voltage), and all stay within a few mV.
Properly means:
all cells from same charge (with no voltage differences)
connections which ensure equal current flow through all cells
HV and LV cut off for each group is a must, but balancing is just a nice feature IMO.
Beacuse if your battery starts showing a drift, something is wrong and you should repair it. No BMS with balancing can "repair" a faulty cell.
If you have a battery made of large cells, like 20Ah or such, than i agree that balancing in necessary beacause there can be much higher variations in capacity compared to a 10p 18650 battery for instance where all cells are from same charge.
Tesla uses about 7000cells with single cell fuses which is by far a differnt story compared to most ebike packs made of 50-300 cells.

scale is not a problem here. even the slightest difference in tab weld on a singe cell can offset the resistance ever so slightly and killing the entire deal. that is just plain stupid. it might not go out of balance in days or weeks but it WILL go out of balance and it cascades so once a slight imbalance happens the pack is gone pretty fast. this is what kills regular lead acid batteries. once cell goes slighty bad and takes the rest with it and lead acid is fairly naturally celf balancing like lifepo4 but us mortals generally stick to 18650 cells wich have zero tolerance for overcharging.


this goes for all you "pray-balancing" people:

please stop talking in theory and give the impression to people that it is fine to have a pack without at least the option to balance by having a balance capable charger or a balance board. theory is fine in a classroom, not when people come here for information and help and spend hundreds of their currency on cells and expect them to last more then a year but their pack died because someone told them "it's fine" without a basic balance board or balance charger. it does not matter if someone builds a pack of 50 or 5000 cells. smaller pack are actually even worse then bigger ones as bigger packs have more resistance to a single cell or connection doing bad things.

 

[madin88]

please stop talking in theory and give the impression to people that it is fine to have a pack without at least the option to balance by having a balance capable charger or a balance board. theory is fine in a classroom, not when people come here for information and help and spend hundreds of their currency on cells and expect them to last more then a year but their pack died because someone told them "it's fine" without a basic balance board or balance charger. it does not matter if someone builds a pack of 50 or 5000 cells. smaller pack are actually even worse then bigger ones as bigger packs have more resistance to a single cell or connection doing bad things.

Hey, please don't get me wrong. I am not saying that BMS should be build without a balancing function! What i am saying is a battery which is in balance doesn't need to be balanced

How accurate are the voltage monitoring inputs on most BMS used for e-bikes?
I can tell you, they are not very accurate, which means the BMS will slightly unbalance your pack which was perfectly balanced before.
Adaptto BMS for instance have about +- 10mV accuracy if not calibrated manually, while one new 18650 battery normally has less drift as the variations of the BMS.

Before i started building my own batteries out of 18650 cells, i also thought that balancing is important and that for a large Ah battery you need big balancing currents, but it showed that this was not necessary.
As mentioned, i know about many batteries which do have a BMS, but the balancers NEVER did touch any cells as the balancing voltage was set higher as the max charge voltage, and these batteries are already lasting 3 or 4 years, not only one
for instance: 4,15V charge cut off and 4,16V balancing voltage.

As long as you have LVC and HVC protection, no battery will die, and if you have blutooth voltage monitoring like the BMS discussend in this thread, or a BMS like Adaptto, you always see whats going on with your battery.
And again: balancing will NOT repair any defect and poor quality cells, or solve other problems caused by improper design or mixed charges.

As mentioned, if you are using cells with large Ah (big LiPo's cells, or cheap stuff with lots of variations), it makes sense to look for a BMS with high balancing currents and let them work all the time, but not for 18650cells from known brands, or other cells with similar characteristics.

edit:
see what liveforphysics (which is one of the smartest guys here in the sphere) is saying about a BMS with 1,2A balancing current:
https://endless-sphere.com/forums/viewtopic.php?f=31&t=85357#p1247983
and he is right!

 

[izeman]

Not to forget that many ppl don't know how to properly build packs and use cells of unknown sources.
These packs will need regular balancing to give decent capacity.
All of my packs got balance wires and a BMS. But I would never assume a pack to be properly built and sized if it needs regular balancing.

 

[madin88]

I was just thinking about following situation:
one has a 10s battery and there is one group that has a bit lower capacity due to a weak cell, or cells from a different batch which have just 95% of the capactiy of the cells from the other batch.

Now you set up the BMS to balance all the time during charge and discharge.
What will happen?
At the end of charging:
That one group will reach max charge voltage sooner as the rest, and the balancers will bleed down this group to prevent a further rise of the voltage.
Ok, that makes sense.

But during discharge, or when almost empty:
The voltage of this group will sag qicker and what does the BMS do? It will bleed down all other 9 groups to the same voltage which would be senseless and counterproductive.

So it would be wise to disable balancing below a given voltage.

 

[ElectricGod]

I was just thinking about following situation:
one has a 10s battery and there is one group that has a bit lower capacity due to a weak cell, or cells from a different batch which have just 95% of the capactiy of the cells from the other batch.

Now you set up the BMS to balance all the time during charge and discharge.
What will happen?
At the end of charging:
That one group will reach max charge voltage sooner as the rest, and the balancers will bleed down this group to prevent a further rise of the voltage.
Ok, that makes sense.

But during discharge, or when almost empty:
The voltage of this group will sag qicker and what does the BMS do? It will bleed down all other 9 groups to the same voltage which would be senseless and counterproductive.

So it would be wise to disable balancing below a given voltage.

I don't think this would happen. Balancing is about the fully charged voltage of the cells, not about partly discharged voltages. I seriously doubt a BMS would ever start discharging less discharged cells that were under full voltage to bring them down to the level of more discharged cells. That makes no sense at all. Balancing just doesn't work that way.

 

[ElectricGod]

this goes for all you "pray-balancing" people:

please stop talking in theory and give the impression to people that it is fine to have a pack without at least the option to balance by having a balance capable charger or a balance board. theory is fine in a classroom, not when people come here for information and help and spend hundreds of their currency on cells and expect them to last more then a year but their pack died because someone told them "it's fine" without a basic balance board or balance charger. it does not matter if someone builds a pack of 50 or 5000 cells. smaller pack are actually even worse then bigger ones as bigger packs have more resistance to a single cell or connection doing bad things.

I have one question for you...
Have you ever ran without balancing?

Admittedly I am MUCH more comfortable with a BMS in place, but I also ran for 3 or 4 months with no BMS on my LION pack. It is 20S12P. Every 2 or 3 times I charged, I checked cell voltages to be sure they were NOT getting over voltage. I didn't have any problems, but never the less I still use BMS so that I don't have to check all the time.

 

[madin88]

I seriously doubt a BMS would ever start discharging less discharged cells that were under full voltage to bring them down to the level of more discharged cells. That makes no sense at all. Balancing just doesn't work that way.

If it does or not depends on the settings.
A good setting would be to start balancing only above 3,6V (or nominal voltage) and only during charge.
Or on a very well balanced pack it makes sense to use the balancers just as an "over-voltage conductor" by setting balancing voltage a tiny bit higher as max charge voltage -> thats how i do with with programmable BMS and i had no issues so far.

 

[flippy]

i only use the tinybms for all packs for 16S or smaller. i always set the bms to start balancing as soon as a positive input is detected, either acutal charging or using regen will trigger the balancing feature during the charging process. it only triggerts beyond a 16mV difference so it isnt that neurotic but it does keep them real close and you can see it happening with the proper software.
and yes, my pack is 3 years old and made from single batch cells and it does require balancing, but even a 4.2kWh pack like mine is no problem to keep balanced with 100mA balance current because it does it every chance it gets so the inbalance remains very small. 15mV is nothing for a pack that size.

 

[ElectricGod]

I seriously doubt a BMS would ever start discharging less discharged cells that were under full voltage to bring them down to the level of more discharged cells. That makes no sense at all. Balancing just doesn't work that way.

If it does or not depends on the settings.
A good setting would be to start balancing only above 3,6V (or nominal voltage) and only during charge.
Or on a very well balanced pack it makes sense to use the balancers just as an "over-voltage conductor" by setting balancing voltage a tiny bit higher as max charge voltage -> thats how i do with with programmable BMS and i had no issues so far.

What BMS does this? I have several programmable BMS and I'm pretty sure NONE do this. They discharge cells when they exceed a threshold which is usually at the full charged voltage...whatever you set that to, but definitely NOT below it.

 

[ElectricGod]

I got 2 more of the 16-20S smart BMS. As usual...craptastic copper reinforcing that doesn't make remotely close to P-.

I thought this was hilarious. So much for bonding the copper down to the board...this is pretty useless. The last 1/3 of the copper on the B- buss is not even attached.

 

[ElectricGod]

2 watt shunts arrived today. I spaced 10 of them side by side on the B- buss and I can fit 9 of them across the space. I'll have to see if 1.5 watt shunts exist. 2 watt shunts are too big.

 

[Inwo]

How about a copper heat sink down each side? Top of existing shunts.
It would also beef up the traces.
Even those cu bus bars might work.

 

[ElectricGod]

How about a copper heat sink down each side? Top of existing shunts.
It would also beef up the traces.
Even those cu bus bars might work.

The copper is fine...if it was actually soldered down and long enough. I think I'll pull up the copper on these new BMS and just lay down new that reaches as a single piece. I patched it together on the previous 2 BMS to make it reach the full length of the buss. There's no good reason to do that. I have copper...it's silly of me to patch it together like I did previously.

 

[izeman]

my 15s BMS doesn't even have any copper ...

 

[Inwo]

How about a copper heat sink down each side? Top of existing shunts.
It would also beef up the traces.
Even those cu bus bars might work.

The copper is fine...if it was actually soldered down and long enough. I think I'll pull up the copper on these new BMS and just lay down new that reaches as a single piece. I patched it together on the previous 2 BMS to make it reach the full length of the buss. There's no good reason to do that. I have copper...it's silly of me to patch it together like I did previously.

I meant copper in place of larger shunt.
It would do dual function as heat sink.

 

[ElectricGod]

I think if you did that, that the BMS would not detect current.

 

[Inwo]

I'll have to make a picture.
Two strips. One down each side, but not touching in the middle.
They could be on edge or even angled.

 

[ElectricGod]

Has anyone tried running these smart BMS from just the balance connections only? If you do and don't care about over current or LVC cut-off, then this BMS could make for a decent smart balancer. Since no current passes through the mosfets the power section would be totally irrelevant and multiple BMS could be ran in series. If you needed 48S, run 3 16S BMS in series. Set each BMS like you would for a 16S pack. Since each BMS "sees" only the cells it is connected to and therefore is operating relative to those cells only, it never exceeds it's operating limits. Isolate each BMS from the others so there's no way for voltage cross contamination to happen and it should be 100% OK. Each BMS won't read current and it won't read total pack voltage, but if all you care about is balancing and cell status, then this would work pretty well.

I'm not ready to do this test yet, but soon. I have 4 of the 16-20S smart BMS and 32S of LIPO at 16,000mah. I can use this to test out series connection. It shouldn't be a problem at all. you will connect to each BMS, check it's cell status, connect to the next BMS and repeat. My end goal is to see if I can get these BMS to operate at an LVC of 2 volts and HVC of 2.8 volts and then build an LTO pack with 48 cells. The spec sheet for the TI charge controller says it can't go this low, but a friend of mine sent me an email saying he had messed with LTO voltages on this BMS and got it going, but doesn't remember the details.

 

[izeman]

even if it would work (don't forget to connect MAINS GND wire, or your voltage reading of cell #1 and #2 will be completely off) those BMS are really bad candidates for balancing only. even the seller's description states that.
balance current is so super low, there are a lot more BMS that do much better that those (bestechpower hcx-131 for example - even though this one lacks BT connectivity, but it's super efficient, and has much higher balancing current)

 

[\ (•◡•) /]

Wow, such an informative thread, i was actually looking at getting on of the 300Amp smart bms's for my lg hg2 20s10p pack about a month ago but went with this bms instead:

At the time I couldn't find any information on the smart bms so I went with this one instead, now I'm regretting that. Is it worth switching to the smart bms? I haven't even used this bms yet so it might be a bit sad to already retire it but with the extra features of the smart bms it's tempting. Also does the bms i purchased look particularly good anyway? I paid 200 aud for it shipped from alibaba, Would love some opinions, i defiantly don't mind spending the extra 160 aud for the new bms if it will help my packs lifespan. Thanks

 

[izeman]

what pro/cons do you ask for? why do you think that your bms is worse than the others?
your's looks like some decent hardware ...

 

[flippy]

Has anyone tried running these smart BMS from just the balance connections only? If you do and don't care about over current or LVC cut-off, then this BMS could make for a decent smart balancer. Since no current passes through the mosfets the power section would be totally irrelevant and multiple BMS could be ran in series. If you needed 48S, run 3 16S BMS in series. Set each BMS like you would for a 16S pack. Since each BMS "sees" only the cells it is connected to and therefore is operating relative to those cells only, it never exceeds it's operating limits. Isolate each BMS from the others so there's no way for voltage cross contamination to happen and it should be 100% OK. Each BMS won't read current and it won't read total pack voltage, but if all you care about is balancing and cell status, then this would work pretty well.

I'm not ready to do this test yet, but soon. I have 4 of the 16-20S smart BMS and 32S of LIPO at 16,000mah. I can use this to test out series connection. It shouldn't be a problem at all. you will connect to each BMS, check it's cell status, connect to the next BMS and repeat. My end goal is to see if I can get these BMS to operate at an LVC of 2 volts and HVC of 2.8 volts and then build an LTO pack with 48 cells. The spec sheet for the TI charge controller says it can't go this low, but a friend of mine sent me an email saying he had messed with LTO voltages on this BMS and got it going, but doesn't remember the details.

it only draws current from the last positive balance input. that is were i meaure all the current being used by the board.

is it possible to just dump the C- side entirely and just bypass the entire mosfet array on that side? it would bypass the shitty mosfets low voltage ones on that side and reduce rdson as a bonus.