Need Advice on BMS System

lx_ss

1 µW
Joined
Jan 1, 2019
Messages
2
Hey Guys,

I need some advice/help on a bms system to suit what i have.
Its an electric motorcycle, 5kw hub motor, Sevcon Gen 4 Controller 72v-84v 140Amp cont/350Amp Burst.
The battery setup is 20s4p pack of LiPo's to give 72v / 64aH, Each pack is 72V / 16 aH. Batteries also rated 15C cont /30C Burst

Knowing the volatile nature of how LiPo's can be, i want some protection, mainly over charge / over discharge. balancing can be done externally but if done via BMS that would be a bonus.

I've looked at Zeva BMS etc.... however by the time i spend that much its worth more than the bike.

Anyone have any ideas of cheap alternative solutions ???

I looked at the chinese Bluetooth BMS and the only ones i can find that suit the 140A cont of the controller are the 150A(some 300A ones also) for approx $150-$200 each, which is fine but it will only do one pack and when you need 4 it all adds up.

Ideal world i'd run different batteries and such, but im just trying to get along with what i have and to run it safely.

Any help would be great.
 
Can't recommend one as I balance my lipos manually and they last months and a shitton of cycles before needing attention. I run 4.08 charge and discharge to around 3.7. Not discharging too far is the main thing that will keep your batteries balanced, and the margin from full charge serves to give you plenty of room for cells to get a little out of balance without consequences.

Do you really want something unnecessary hooked up to your battery all the time that can kill it if it frocks up? I sure don't, so I don't use a BMS. Note, I would use a BMS if it came from a reputable company like ASI.

If you did use a BMS, you would want to make sure the termination / balance voltage is programmable, so you don't end up hitting 4.2 volts / cell every balance.
 
flat tire said:
Can't recommend one as I balance my lipos manually and they last months and a shitton of cycles before needing attention. I run 4.08 charge and discharge to around 3.7. Not discharging too far is the main thing that will keep your batteries balanced, and the margin from full charge serves to give you plenty of room for cells to get a little out of balance without consequences.

Do you really want something unnecessary hooked up to your battery all the time that can kill it if it frocks up? I sure don't, so I don't use a BMS. Note, I would use a BMS if it came from a reputable company like ASI.

If you did use a BMS, you would want to make sure the termination / balance voltage is programmable, so you don't end up hitting 4.2 volts / cell every balance.

How do you monitor that you dont discharge the cells lower than 3.7 whilst using them ??
 
lx_ss said:
How do you monitor that you dont discharge the cells lower than 3.7 whilst using them ??

There are a few ways. After you get experienced you'll be able to get a general idea of how much energy a given ride will take and plan accordingly. To make sure the entire pack doesn't go below a given voltage you can use a Cycle Analyst, multimeter, power meter, or whatever to monitor the resting voltage. Just divide the pack voltage by the series count to determine the per-parallel group voltage. So a 20S pack = 20 * 3.7 = 74 volts target to finish your ride.

Cycle analyst is also a power meter / cycle computer and will count your power used in watt-hours. Very convenient.

To make sure my individual lipos aren't going out of balance I check them from time to time using my iCharger. Avoiding physical damage to the cells along with observing the charge / discharge guidelines I recommend will minimize any balance differences, and balancing should rarely be necessary. One additional factor that can put cells out of balance is if your C-rate is insufficient. That shouldn't be a problem in your case.
 
Yes, people think "a BMS" should be available as a single product.

Best IMO to look at it as functionality, shared across simple robust OTS devices.

If live cell balancing is removed from the req list, much less prone to failure.

That leaves LVD, OVD and maybe temperature protection.

Implementation at pack- or bank- level depends on the use case.

Redundancy required for mission-critical or very large expensive banks.

All in addition to charge control and monitoring by the user, which can in fact make many automated systems unnecessary, substitute time and diligence for capital investment.
 
But wasn't it an active bms will be much better to have it vs not ? Like here, 4 pack, so 4x active bms high current 200a (chinese do have.) So when in used, all cell are fully protected with thoses 4 bms.
And after used, like 1x time a week we plug the bms in for a day, then disconnect the bms the next day. So cell will be kept in balance, fully monitored and no chance to get a faulty bms by not leaving it on all the time... Running a bms is less likely to get a bad cell and way more less likely to catch a fire because 1 of the pack get too low or so..
 
Huge difference between homemade packs of secondhand cylindrical cells,

and brand new top notch big prismatics.

With the latter, thorough testing during commissioning will reveal substandard cells. As pack-level cutoffs will be sufficient "live" protection.

Avoiding high C rates and both voltage shoulders, they just don't go out of balance much.

But yes, whatever tool you use to periodically check per-cell voltages, does not need to be "a BMS", that is just one bit of required functionality.

For the LVD / OVD cutoffs, no setpoint adjustability is a deal-breaker for me, what's the point if too much damage has already been done?

And with close personal attention I think many not too risk-averse owners can do just fine with just bank level protection using OTS components.

Ultimately these are all owner judgment calls, I have yet to see "a BMS" I'd want, from a vendor I'd trust, that doesn't come close to doubling the cost of the bank.

The open-source Arduino or rPi or PLC approach seems more promising to me.


 
Whoops sorry, just realized where I was.

I use LFP, not LiPo.

And House bank storage use cases, not EV, so 180AH CALB units, .2C would be very rare, most loads under 10-20A.

Which may make all my comments here irrelevant
 
Hi
what you mean by :
For the LVD / OVD cutoffs, no setpoint adjustability is a deal-breaker for me
Was it: if there's no overvoltage and lowvoltage protection ajustable... the bms is useless, because it need thoses function ?

Cheap bms to only do the balancing and charging are ok, but a smart handling discharge... doesn't it sound safe on the long run ?

thanks again
 
docop said:
Hi
what you mean by :
For the LVD / OVD cutoffs, no setpoint adjustability is a deal-breaker for me
Was it: if there's no overvoltage and lowvoltage protection ajustable... the bms is useless, because it need thoses function ?

Cheap bms to only do the balancing and charging are ok, but a smart handling discharge... doesn't it sound safe on the long run ?

thanks again
Again, a BMS is functionality, not "a thing" that does everything you need.

The only functions I need are temperature protection and OVD/LVD.

All three of those, I require user-adjustable setpoints. And I specifically do not want anything "live" during normal cycling usage at the cell level.

I don't know what you think you mean by "smart handling discharge" or about charging, other than the three functions I listed.


 
lx_ss said:
I looked at the chinese Bluetooth BMS and the only ones i can find that suit the 140A cont of the controller are the 150A(some 300A ones also) for approx $150-$200 each, which is fine but it will only do one pack and when you need 4 it all adds up.

You don't want to pump your current through BMS at all. Find a BMS that uses contactors for load/charger switching, then discharge rating is only limited by the rating of the contactor. Given your configuration requires at least 20 channels, Chargery BMS24T is the only inexpensive one that comes to mind, but Chargery as a brand has a questionable reputation that has been brought up here recently.
 
Back
Top