Functions of a Battery Management System

[irie]

Seems to be some confusion about 'flat' batteries so thought this info from Google would be interesting. Specifically, a BMS protects a battery from complete discharge because Lithium-Ion batteries (always?) require having some residual charge to be able to recharge.

A battery management system (BMS) ensures a battery pack's safety, longevity, and optimal performance by monitoring voltage, current, and temperature; estimating the battery's state of charge (SOC) and health (SOH); balancing individual cells; managing thermal conditions; preventing overcharging and over-discharging; and communicating data to external systems. These functions prevent damage, protect users, and maximize the battery's lifespan and efficiency.

Core Functions of a Battery Management System

Monitoring:

The BMS continuously tracks key parameters like cell voltage, pack current, and internal/external temperatures to understand the battery's real-time status.

State Estimation:

It calculates essential information about the battery's condition, such as:
State of Charge (SOC): The remaining charge in the battery.
State of Health (SOH): The battery's overall condition and remaining capacity.

Protection:

The BMS acts as a safeguard by preventing hazardous operating conditions that could damage the battery or pose a safety risk to users. This includes protecting against:

Overcharging
Over-discharging
Overcurrent
Over-temperature

Cell Balancing:

For multi-cell battery packs, the BMS ensures all individual cells remain at similar voltage levels by moving charge from higher-voltage cells to lower-voltage ones, which is crucial for consistent performance and longevity.

Thermal Management:

It regulates the battery's temperature by controlling cooling or heating systems to keep the cells within their optimal operating range.

Power Management:

The BMS manages the flow of power to and from the battery pack, optimizing energy usage and maximizing the system's efficiency.

Communication:

It exchanges data with other devices or systems (like a vehicle's main computer) to provide real-time information, report faults, and facilitate overall system control.

I think @cream can add to this including do's and dont's.

 

[Ndanger]

On a Bosch Gen 5 eBike system, when the display shows zero bars (empty battery), the pack is not truly at 0 volts —
For practical purposes a battery is flat at 30 volts not Zero full_stop. And fully charged at 42 volts.

lithium-ion batteries are never fully drained because that would permanently damage them.


Bosch for example (like most e-bike makers) builds in a battery management system (BMS) that cuts off power before the cells reach a harmful state of discharge.

• A Bosch Gen 5 battery is usually a 36V nominal pack (10-cell series, 18650 or 21700 cells).
• Fully charged: ~42.0V.
• Empty (display shows 0 bars): usually around 30–31V under load.
• The BMS will cut off completely somewhere close to ~29V to protect the cells.

So when your display shows 0% or no bars, the battery still has a safe reserve charge. It’s not actually empty, but it’s below the usable level Bosch allows for motor assistance.

In practice:

• 42V = 100%
• ~36V = 50%
• ~30V = 0% (display empty, cutoff imminent)

 

[cream]

Indeed, those are the functions of a BMS, in general. Complexity vary with application or manufacturer.

As for ebike world, there's no active thermal management and also no active cell balancing. At best, they have passive balancing with few mA or tens in some cases. Active balancing is not worth in low power electric vehicles. A properly designed battery with quality cells should last 3-400 cycles without serious degradation. In practice, most won't reach that number because of external causes (shorts, water ingress etc). In case of uneven wear of the cells, BMS can't do anything to compensate other than to report an error/warning or complete power cut off when reaching a certain limit set in firmware.

Now, for do's and dont's...

If you want to get the most of your battery, DON't rely solely on the BMS protections/limits.

DO use your battery, rather than storing. If you need to store it, DON't do it at higher than 80% state of charge and lower than 30%. In many cases, full charge storage will be worse for battery health than store it at low level charge.

DON't charge your battery immediately after draining it at high load (turbo/boost/etc). Let it rest/cooldown for at least 1h.

On cold weather, try to warm it up before charging. Even if BMS has thermal protection, charging it close to that limit will degrade your battery faster. This might not matter to the manufacturer after warranty period is over...but it matters to you. I still see batteries properly maintained with >300cycles and >5 years and cell health(available capacity) between 92-94% (of course, no significant imbalance).

Avoid washing your bike with pressure in the battery area/electrical connectors/drive unit. It might take a while longer but in the end, you'll enjoy your ebike for a lot longer without the need to service the electric system.

DO check the connectors for dirt/moisture every time you remove/install the battery. On magnetic connectors (like Rosenberger), it's mandatory to check for metallic debris that can interfere with electrical signals potentially doing damage. Even if the debris is not electrically conductive, it will still prevent the connectors from seating properly, creating a higher resistance which in time will destroy the contacts (by heating, oxidizing). I personally DON't recommend using contact cleaners or any dielectric grease. That will attract dust/dirt much easier, creating a substance you don't know the properties of. Cleaning should be done with Isopropyl alcohol or other substance that will clean and evaporate with no trace or chemical reaction with metal alloys the contacts are made of.

DON't use 3rd party cheap chargers with cheap connectors. You'll risk damaging the whole electrical system. I've seen cases with Bosch gen2/3/4 where using a $10 charger resulted in total system failure. Blown motor control PCB, Display PCB and Battery BMS PCB.

 

[irie]

Interesting about the lack of active battery thermal management.

@cream

Is it right to say that current more tightly packed batteries in downtubes (to make eMTBs more attractive) increases the likelihood of performance degradation or system shutdown from an overheating battery?

Edit: a second problem with heat dispersion of carbon fibre compared with aluminium is illustrated by the below info from Google:

"Aluminum has significantly higher thermal conductivity than carbon fiber, transferring heat about 40 times faster, making it excellent for heat dissipation and suitable for heat sinks, while carbon fiber is a much better thermal insulator due to its low conductivity."

So perhaps a potential double thermal whammy from tightly packing batteries in carbon fibre downtubes for weight saving and cosmetic purposes?

 

[cream]

I would say battery overheating in normal conditions is unlikely (unless an internal problem with the battery). More likely motor will overheat on high loads which will cause performance degradation because of throttling.

Most serious battery packs manufacturers (for ebikes) will use a configuration that won't stress cells too much. Of course, there are exceptions...

For example on Bosch older systems with 18650 cells. They're using 10s4p configuration with 10A continuous discharge rating. That would give a 40A max continuous discharge for the whole pack. Even if you run the motor full power all the time (which you can only do it on a bench test...or not if the motor will overheat and start thermal throttling) you won't be using more than 20A. So this is half of max discharge permitted, in these conditions no overheat will occur.

Some are using 10s5p configurations or 21700 cells(with higher discharge current rating than 18650) in 10s4p configuration.

There were cases on Bosch smart system 750Wh batteries that would give false error about battery temperature/overheating but in reality the cells were within normal temperature limits.

There are recent 21700 cells with insane current and capacity ratings. 20-30Amps/cell at 5500-6500mAh, in a 10s4p pack that would result in 60-90Amps max discharge rating. You could easily build a lighter pack in a 10s3p configuration not worrying about thermal problems on most full power eMTBs out there.

In any case, using Alu for battery housing will improve thermal transfer but if there's an air gap between cells and metal case, the improvement will be minimal.

 

[Winford]

if BMS has thermal protection

Amflow does. Not only does it have a battery temperature monitor, if you get it hot from excessive current draw, it will go into a limited mode, and or may power itself off.

Shimano has a slightly different method, for excessive draw, but that battery also will shut itself off from overheating, and takes time before it will boot back up, which varies depending on how hot you got it.

I believe most BMS are all computer controlled to some extent, some more than others. Shimano has quite the list of codes the battery will spit out with blinking leds

 

[Winford]

I would say battery overheating in normal conditions is unlikely

Shimano is terrible for this, it is so finicky they just throw parts at it blindly, but it never fixes the problem.

My battery has shut off multiple times, and thrown error codes so you know its not the motor, but they are tied together, meaning if the motor has any issue, it can overheat the battery or visa versa.

I was told the battery and the motor has temp sensors, on Shimano you can tell its the motor because it shuts off while climbing a hill, and wont turn back on until it cools, but the battery will power up without codes. Most of the time these Canyons shut off while coasting after a 3 mile 1000' climb. I think we all got around 1000 miles on the bikes before we had shut down issues, and Shiman just replaces the Rosenberger main power cable, which only helps for a few weeks. I have had the motor replaced and finally the battery on recall. Now after 14 months, which is only 12 months riding im on my 3rd EP801 motor. 5000+ miles and 700,000' of climbing

 

[Tom Sellers]

"As for ebike world, there's no active thermal management and also no active cell balancing."

JK and JBD both offer advanced Battery Management Systems (BMS) that include active cell balancing and thermal management features, distinguishing them from simpler or older e-bike 'passive' BMS units. I use these whenever I build eBike battery packs.

A properly designed battery with quality cells should last 3-400 cycles without serious degradation.

• Some studies show cycle lives around 250 cycles for LiCo chemistry (usually found in eBikes) before capacity reduces significantly, consistent with degradation rates observed in real-world applications. sciendo
• Note: Charging a lithium-ion battery to about 80% state of charge (SOC) can significantly extend its lifespan, often by 2 to 3 times or more, compared to charging to 100% regularly. I use a CCCV bench power supply for this purpose.

 

[randycpu]

On a Bosch Gen 5 eBike system, when the display shows zero bars (empty battery), the pack is not truly at 0 volts —
For practical purposes a battery is flat at 30 volts not Zero full_stop. And fully charged at 42 volts.

lithium-ion batteries are never fully drained because that would permanently damage them.


Bosch for example (like most e-bike makers) builds in a battery management system (BMS) that cuts off power before the cells reach a harmful state of discharge.

• A Bosch Gen 5 battery is usually a 36V nominal pack (10-cell series, 18650 or 21700 cells).
• Fully charged: ~42.0V.
• Empty (display shows 0 bars): usually around 30–31V under load.
• The BMS will cut off completely somewhere close to ~29V to protect the cells.

So when your display shows 0% or no bars, the battery still has a safe reserve charge. It’s not actually empty, but it’s below the usable level Bosch allows for motor assistance.

In practice:

• 42V = 100%
• ~36V = 50%
• ~30V = 0% (display empty, cutoff imminent)

Battery SOC must be accessed not "under load", but at quiescence. No load after a few minutes of no load. The BMS and display do this automatically so as to prevent premature shutdown and wildly fluctuating SOC levels

 

[Just gan]

JK and JBD both offer advanced Battery Management Systems (BMS) that include active cell balancing and thermal management features, distinguishing them from simpler or older e-bike 'passive' BMS units. I use these whenever I build eBike battery packs.

• Some studies show cycle lives around 250 cycles for LiCo chemistry (usually found in eBikes) before capacity reduces significantly, consistent with degradation rates observed in real-world applications. sciendo
• Note: Charging a lithium-ion battery to about 80% state of charge (SOC) can significantly extend its lifespan, often by 2 to 3 times or more, compared to charging to 100% regularly. I use a CCCV bench power supply for this purpose.

If battery is at 20 or 30 percent after a ride and I am going out again in two days I normally top it up to 80 ish . Would it degrade if I left it and just topped up before I went out.

 

[Tom Sellers]

That's pretty much what I usually do:

• Resting at 20-30% SOC for two days is safe and unlikely to degrade the battery noticeably.
• Charging to ~80% before riding is the best practice to maximize battery lifespan and readiness.

Note: On my older battery packs that have a passive BMS I roughly charge to 100% SOC one in ten charges because I read somewhere that with the old style passive BMS's they don't do any balancing otherwise. (Passive BMS systems typically perform cell balancing by dissipating excess charge as heat to equalize cell voltages, but this balancing often happens only during full charge cycles when the pack reaches 100% SOC). But when I am using a new pack that I built myself using a better BT Active balancer BMS, I don't adopt this practice. (Active balancing systems actively redistribute charge between cells during charging and discharging cycles at all SOC levels, rather than only dissipating excess charge at full charge as in passive balancing.)

 

[Just gan]

That's pretty much what I usually do:

• Resting at 20-30% SOC for two days is safe and unlikely to degrade the battery noticeably.
• Charging to ~80% before riding is the best practice to maximize battery lifespan and readiness.

Note: On my older battery packs that have a passive BMS I roughly charge to 100% SOC one in ten charges because I read somewhere that with the old style passive BMS's they don't do any balancing otherwise. (Passive BMS systems typically perform cell balancing by dissipating excess charge as heat to equalize cell voltages, but this balancing often happens only during full charge cycles when the pack reaches 100% SOC). But when I am using a new pack that I built myself using a better BT Active balancer BMS, I don't adopt this practice. (Active balancing systems actively redistribute charge between cells during charging and discharging cycles at all SOC levels, rather than only dissipating excess charge at full charge as in passive balancing.)

Brilliant thats what I wondered about . Last battery probably treated the same way . It lasted June 2019 till March 2025 with 10,000 miles then I sold bike. Last ride eco all the way i did 39 miles and had 26 percent left .

 

[Just gan]

Thanks

 

[BushLevo]

Specialized Levo range have batteries and firmware that provide automatic charge limit to 80%. Every 10th charge the 80% limit is overridden to provide a 100% charge followed by cell balancing. You can disable the 80% charge limit and manually time charging if you wish. My 2019 700Wh battery has had 64 full charge cycles and is still at 99% health (according to BLEVo app on my phone)

 

[Tom Sellers]

I'm guessing it will not be too long before we see most manufacturers going to active balancing given that JK and JDB active BMS's are only in the range of 30-40 dollars more than their passive counterparts. Although I use a CCCV bench power supply for charging, the Meanwell NFC version of the NPB-450 series has caught my attention:

 

[Ndanger]

I'm guessing it will not be too long before we see most manufacturers going to active balancing given that JK and JDB active BMS's are only in the range of 30-40 dollars more than their passive counterparts. Although I use a CCCV bench power supply for charging, the Meanwell NFC version of the NPB-450 series has caught my attention:

How does the power supply interface with a battery pack that has a BMS inline and downstream of the power supply

 

[Tom Sellers]

On a typical BMS, the charging port and power output lead to the same connections on the battery pack.

 

[Fangs2k]

‘Thermal management’ of battery packs ranges from nothing to extensive depending on the pack and its application.

E-bike battery pack thermal management is pretty basic, ie battery temp info informing the bikes ECU so that in can ‘throttle’ the draw of the motor to allow the pack to cool down, or even just turn it off until it’s back within the designed operating range.

It’s clear the Google article is referring to electric cars when talking about the BMS being able to actively heat or cool the pack, most modern EV battery packs are liquid cooled/heated so that they spend as much time as possible in the ‘Goldilocks zone’ for performance and lifespan reasons.

We don’t and likely won’t have anything like that on a bicycle as it’s heavy, expensive and ultimately unnecessary for the low current draw in any e-bike that complies with the relevant legislation.