Motor dies-then recovers-frequently

[Peter Brigg]

I have an E bike running a Q100 motor with an S-KU65 controller and the usual add ons-throttle and a charge indicator (pictured) All goes well for about 100 m and the system shuts down-I restart with the on/off charge indicator device and it goes another 100m then stops
I have done the following
Put a load on the battery 2X 24v truck light globes in series and they remain bright for a 10 minute test-lots of heat so i assume the battery is good. The battery is a 36v Li-ion type
Changed the controller to a new spare one same model
Changed the switch device pictured to a spare new one.
Have not substituted a new throttle for the existing one-I assume throttles are pretty robust devices
Can someone suggest what the problem might be??

 

[amberwolf]

Put a load on the battery 2X 24v truck light globes in series and they remain bright for a 10 minute test-lots of heat so i assume the battery is good. The battery is a 36v Li-ion type

what voltage is it while you have it under this load?
if it's dropping below the controller lvc under load, then even if it doesn't shut off at the battery the system would still stop working.

 

[E-HP]

Looking at the controller description, the lvc for the controller, using a 36v battery, is 30v. Does switching to the lowest power mode make any difference?

 

[Peter Brigg]

what voltage is it while you have it under this load?
if it's dropping below the controller lvc under load, then even if it doesn't shut off at the battery the system would still stop working.

Thanks for your answer-I put the battery under the 3 globe load and the voltage stabilised at 34v and held there for a timed 8 minutes. BTW I have been looking for the specs of the KU65 to determine the lvc among other things-where should I look? Thanks
Peter

 

[E-HP]

S-KU65 Controller for LED810 250W 15A 6mosfets Controller LED No LED

S-KU65 Torque Simulation Controller 6mosfets 250W 15A

bmsbattery.com

 

[Peter Brigg]

I am now thinking it is a connection problem-perhaps a joint that heats up and then fails but when cooled it works OK. E-HP suggests using a lower load and that does help-if I creep along on very low throttle the motor continues to provide some drive but if I gun it -it dies

 

[E-HP]

Likely voltage sag, as amberwolf stated.

 

[Peter Brigg]

It looks as though the problem is partly fixed-self inflicted on my part.
I was undercharging the battery possibly.
A few figures-the battery charger reads 43.7v unplugged .While charging the battery I saw some cells were 4.4v during the charge and believed that this was damaging to the cells so stopped charging at that point. So is the 4.2v maximum voltage per cell to apply only to a fully charged cell while not on charge? When I first connect the charger the battery voltage is 39.5v and after 10 minutes of charging the battery reads 40v and the pack is warm to the touch. Does this sound OK? The battery consist of a bank of 9 groups of 5 cells each in series. Would this be described as a 9S5?

 

[offGridDownUnder]

There is the general lore that most BMS circuits can't balance the cells until the battery is fully charged - it may be that your cells are not balanced and by not permitting the charge to reach 100% the BMS could not (yet) balance the cells.

However, I want to hear from someone more knowledgeable about the cells at 4.4v myself before I trust that answer.

 

[amberwolf]

A few figures-the battery charger reads 43.7v unplugged .While charging the battery I saw some cells were 4.4v during the charge and believed that this was damaging to the cells so stopped charging at that point. So is the 4.2v maximum voltage per cell to apply only to a fully charged cell while not on charge?

4.2v is the highest the cell should see. Sometimes a BMS will allow another few hundredths of a volt above that before it hits HVC, but not usually two tenths or more.

If the pack is not turning off charge on it's own, it's BMS is either defective, not present, or bypassed, or is being charged from the discharge port instead of the charge port. (or it is so badly designed that it qualifies as defective)

When I first connect the charger the battery voltage is 39.5v and after 10 minutes of charging the battery reads 40v and the pack is warm to the touch. Does this sound OK?

If the cells are heating up noticeably during charge, then it's likely the charge rate is too high for them, either because they are not designed to charge that fast or because they are aged so much that their resistance has increased (which also causes problems during discharge).

If they have a low charge rate, they may also have a low discharge rate, and may be being used beyond their abilities there, too.

The battery consist of a bank of 9 groups of 5 cells each in series. Would this be described as a 9S5?

If there were only 5 cells in series, it would only be 5 x the cell voltage for battery voltage total.

If instead you mean there are 9 cells in series, then that's 9 x the cell votlage for battery voltage total.

Normally there are 10 cells in series for "36v" batteries, so 10s. If yours only has 9, then a 43.7v charger can seriously overcharge the pack. Even for a 10s pack, it should only be 42v.

A 9s pack should be maximum 4.2v x 9 = 37.8v full, so it's charger should be no higher voltage than that.

If a pack's BMS is working correctly, it will stop charge as soon as any cell reaches full. But if it is not working correctly or not designed properly or you don't have a BMS, or you have bypassed it by charging thru the output (controller) port, etc., then any charger that can go higher than the full pack voltage will overcharge the pack.

If a pack is not made of matched identical cells that are all at the same state of charge SoC (voltage), then it will be "unbalanced" and without a functioning BMS a higher-voltage charger than it should have would make it even worse, and have even greater risk of catastrophic cell damage (and potential for fire).

Most packs are not made of matched cells, so most packs require a functioning BMS with a balancing function to prevent imbalance and assorted problems (including catastrophic failures).

Many BMS do not have a balancing function, so while they will still stop charge when a cell (the worst one) gets full, they won't drain off that charge so the rest of the pack can continue charging and fill up the better cells. Eventually the pack will get farther and farther out of balance until there is very little usable charge in the worst cells.

A balancing BMS will help an unmatched (or aging, etc) pack continue to be useful for longer (usually a lot longer).

 

[amberwolf]

There is the general lore that most BMS circuits can't balance the cells until the battery is fully charged - it may be that your cells are not balanced and by not permitting the charge to reach 100% the BMS could not (yet) balance the cells.

Most BMS that have a balance function (many don't) will begin balancing at some point near full (which is normally 4.2v for the typical Li-Ion cell; depends on the cell chemistry and the BMS design; you have to use a BMS that matches the cells used in it, or is programmable to make it match them).

When a cell actually reaches the full point of the BMS (HVC) it will cut off the charge port, preventing further charge of any cell.

If it has no balance function, all charge ends at this point.

If it has a balance function, it will drain down the high cells to some cutoff point and resume charge, and cycle this way until all cells are at HVC.

If there is no BMS (or it isn't working or is bypassed), the pack will continue charging all cells until the total cell package voltage equals the charger voltage, regardless of what each individual cell voltage is. So the worst cells with the lowest capacity will have the highest voltages (above their HVC, probably), and the best cells will have the lowest voltages (as they aren't full yet). That can damage the already-not-as-good-cells further and the problems grow worse with every charge/discharge cycle. Eventually a catastrophic failure can happen (fire, etc) because the worst cells can be massively overcharged (and massively overdischarged, or even reversed in voltage), once they have a small enough capacity relative to the rest.


There are some BMS that work by shuffling charge, and some of these work all the time, not just during charging, so they may continue to shuffle charge around between cells to keep them all equal even while you are riding, or if the pack is just sitting.

That has a dangerous side-effect if they're not designed correctly, in that a pack with a defective cell that "leaks" internally so it always drops in voltage constantly even just sitting there, will keep feeding that cell and drain all the other cells too, until they are all at LVC and the BMS shuts off. But, if the pack sits like this, the BMS is still powered by the pack (even if minimally), and will keep draining them. LVC is there because the cells don't have much (if any) useful charge left at this point, and if it sits a long time the drain may pull them down to a non-rechargeable state. (meaning, not safe to recharge, vs not able to do so).

 

[Peter Brigg]

4.2v is the highest the cell should see. Sometimes a BMS will allow another few hundredths of a volt above that before it hits HVC, but not usually two tenths or more.

If the pack is not turning off charge on it's own, it's BMS is either defective, not present, or bypassed, or is being charged from the discharge port instead of the charge port. (or it is so badly designed that it qualifies as defective)


If the cells are heating up noticeably during charge, then it's likely the charge rate is too high for them, either because they are not designed to charge that fast or because they are aged so much that their resistance has increased (which also causes problems during discharge).

If they have a low charge rate, they may also have a low discharge rate, and may be being used beyond their abilities there, too.




If there were only 5 cells in series, it would only be 5 x the cell voltage for battery voltage total.

If instead you mean there are 9 cells in series, then that's 9 x the cell votlage for battery voltage total.

Normally there are 10 cells in series for "36v" batteries, so 10s. If yours only has 9, then a 43.7v charger can seriously overcharge the pack. Even for a 10s pack, it should only be 42v.

A 9s pack should be maximum 4.2v x 9 = 37.8v full, so it's charger should be no higher voltage than that.

If a pack's BMS is working correctly, it will stop charge as soon as any cell reaches full. But if it is not working correctly or not designed properly or you don't have a BMS, or you have bypassed it by charging thru the output (controller) port, etc., then any charger that can go higher than the full pack voltage will overcharge the pack.

If a pack is not made of matched identical cells that are all at the same state of charge SoC (voltage), then it will be "unbalanced" and without a functioning BMS a higher-voltage charger than it should have would make it even worse, and have even greater risk of catastrophic cell damage (and potential for fire).

Most packs are not made of matched cells, so most packs require a functioning BMS with a balancing function to prevent imbalance and assorted problems (including catastrophic failures).

Many BMS do not have a balancing function, so while they will still stop charge when a cell (the worst one) gets full, they won't drain off that charge so the rest of the pack can continue charging and fill up the better cells. Eventually the pack will get farther and farther out of balance until there is very little usable charge in the worst cells.

A balancing BMS will help an unmatched (or aging, etc) pack continue to be useful for longer (usually a lot longer).

Thanks Amberwolf for that really comprehensive answer. I have had bad experiences with BMS-probably because i bought cheap ones so the battery setup you saw was without one. From info provided on this site I thought an answer would be to add another bank of 5 cells in parallel making the setup a 10S rather than a 9S-this makes the charger more compatable with the battery. The charger is a relic from charging a 36v LiFePo4 battery. After changing to a 10S system all works fine so apparently the voltage drop caused either the controller or the throttle device to shut the system down.. Each group of cells is led to a terminal strip on the outside of the battery so I can manually balance when need be. So all the contributions helped. My next step is to buy a quality BMS with Bluetooth which will be a big improvement. I only charge the battery on a timer and avoid any cell exceeding 4v .Can you recommend a brand of BMS? One of the issues with getting info on this site is miscommunication-if you could see what a contributor is doing I think problems would be solved more easily. For eg many GP's in Australia do a telehealth appointment for a medical consultation so big mistakes are made Thanks again

 

[amberwolf]

Thanks Amberwolf for that really comprehensive answer. I have had bad experiences with BMS-probably because i bought cheap ones

That's not an unusual experience--most batteries come with cheap ones, and most replacements people choose are no better. A good BMS can cost as much or more than the rest of the pack if it's a cheap pack....

The best solution depends on budget, usage, space, user-capability, etc., but generally getting a pack made from matched (identical, not similar but *identical*) cells that are significantly more capable (by at least half-again) than the application needs, so that it will still meet requirements as it ages, will do more than getting a superb BMS for keeping a pack working as needed for the life of the system.

The problem is that it is virtually impossible to get a pack that's made like this unless you do it yourself, and can find a *trustworthy* :lol: cell vendor that will sell you pre-tested guaranteed-matched cells, or at least all come from the same batch and you also have or get the equipment to match them....but you'd have to buy many more cells than you need so you can find enough identical ones among them to build your pack from.

Or get EV-grade large-format cells (even used, or especially used from the same vehicle's pack) as they will already be tested and matched at the EV-pack factory. That's the solution I use, and even though the cells I have were used when I got them, and they are now more than a decade older than that with a lot of miles on them, they are *still* matched, even though their performance has degraded over time, it has done so identically for all of them. (so there are no balance issues; I don't use a BMS, just random checking now and then)

From info provided on this site I thought an answer would be to add another bank of 5 cells in parallel making the setup a 10S rather than a 9S-this makes the charger more compatable with the battery. The charger is a relic from charging a 36v LiFePo4 battery. After changing to a 10S system all works fine so apparently the voltage drop caused either the controller or the throttle device to shut the system down..

That sounds like expected results--remember that controllers are built with HVC/LVC votlages that match specific battery types, so if you use one that is not in the voltage range the controller is designed for, the system is not going to work quite as expected (unless you take the differences into account).

One of the issues with getting info on this site is miscommunication-if you could see what a contributor is doing I think problems would be solved more easily.

The biggest part of that issue is that the people asking for help don't usually provide all of the necessary information, even when it's asked for. Sometimes it's because they don't know to provide it, but sometimes it is for unknown reasons (to us), when insufficient responses (or none) are made to various questions.

 

[Peter Brigg]

Yes-some cell groups did briefly reach the 4.4v mark and Irealised that was not a good idea so I put a load on and reduced the cells to an acceptable range-so it was a brief overcharge and it appears no damage was done. Cells became warm not hot

 

[amberwolf]

Cells became warm not hot

When cells are charged at appropriate rates for their actual properties, they should not warm up noticeably. (RC packs not withstanding, as those are typically used outside their true capabilities regularly ).

It doesn't mean a cell is incapable of charging at that rate, but it means that it's probably not optimal for various reasons to use that rate.

If a cell heats up, then it's resistance is high for that current flow, so significant power is being wasted inside the cell as heat. In discharge it means that waste power from the pack is heating it instead of driving your wheel (or wahtever the pack is for). In charge it means that waste power from the charger is heating it instead of charging the cells.

Heat inside the cells also degrades them faster than they would otherwise (because chemical reactions generally occur faster at higher temperatures); by how much will depend on the specific cell design/chemistry/properties/age, and the amount of heat.

 

[Peter Brigg]

When cells are charged at appropriate rates for their actual properties, they should not warm up noticeably. (RC packs not withstanding, as those are typically used outside their true capabilities regularly ).

It doesn't mean a cell is incapable of charging at that rate, but it means that it's probably not optimal for various reasons to use that rate.

If a cell heats up, then it's resistance is high for that current flow, so significant power is being wasted inside the cell as heat. In discharge it means that waste power from the pack is heating it instead of driving your wheel (or wahtever the pack is for). In charge it means that waste power from the charger is heating it instead of charging the cells.

Heat inside the cells also degrades them faster than they would otherwise (because chemical reactions generally occur faster at higher temperatures); by how much will depend on the specific cell design/chemistry/properties/age, and the amount of heat.

 

[Peter Brigg]

A question-I have a 36v ebike battery LiFeP04 type a 38120 and one of the cells was reading low at about 2.5v I charged this cell by itself and bought it up to 3.2v then discharged it with a light bulb and charged it up again and it seems to be "rejuvenated" ie holding at 3.2v under load and remaining much the same as the other cells. Does this happen? Could it really be as good as the rest of the cells or will it just degrade to its former state over time.

 

[amberwolf]

How much capacity does it have?
how much capacity do all the other cells have?
What's the internal resistance of the cell vs the others?

Those will tell you what will happen with it.

There is always a reason a cell (or group) is different voltage than the others; some common ones:
--the BMS balancer is stuck on, and is constantly draining it.
--the cell is internally leaking, and constantly draining itself.
--the cell has a lower capacity than the others, so it runs down in voltage faster than the others (and rises faster during charge)
--the cell has a higher resistance than the others, so it rises faster during charge high current stage causing early shutoff of charging, and falls faster under load causing early shutdown of system during discharge.

There's a lot of battery and pack and cell testing threads that have much more detail on these and other things, as well as testing methods, if you're interested in poking around.