Throttle drawing 800+ watts - shuts off unit. Help?

[Zarathustra1984]

Hello everyone. Long-time lurker and first time poster here. I appreciate the wealth of knowledge you all have given and it's helped make my dream bike (cheapo) come true. This is my first build after a period of about 10 years not owning a bike and want to give thanks.

The lowdown is that it has a 500W hub motor, KT48ZWSRKT-SJT02L, and 50A BMS 10AH battery and I'm getting intermittent LCD-3 shutoffs under load when I apply the throttle - even gently. It accelerates as fast as a car and I set the speed limit to 30 mph to stay within the law as it will do just that. The LCD-3 shows ~800 watts being put out when under full throttle load and the Voltage drops to ~43-44 and slowly climbs as speed increases and thus load decreases. I did not think to see the Voltage when the issue occurred, only now.

The first night I took it out, I went about 15 miles stop/go driving in the city with mostly full-throttle rips and speed limited; had a great time. Last night, mostly the same, but didn't drive it as hard, and around the similar mileage, the system shut down when applying throttle from a stop - it made it about 10 feet and the LCD-3 shut off. I unplugged and plugged back in the battery and I was able to use P.A.S. mode to limp it home. Any time the throttle isn't babied (like 1/10th of full twist) and it eased up to speed, it would immediately shut off the system. P.A.S. #5 will cause a shutdown. Once, I was able to get the LCD-3 to flash the battery meter being empty and then a slow fading shutdown - this on at least 75% battery. I gave it a full charge overnight and the same thing this morning for ~10 miles. Now, here the kicker: I had it on the charger for a while just now and decided to make a quick run to the store and it read ~52v and was able to apply full throttle again - even from a dead stop as if the issue went away. Note, this morning the system was completely cooled and just now it was plenty warm. I'll add that the "Regenerative Braking" only kicks in for about 1 sec when it's on a stand and also that the battery meter gauge in the upper left-hand of the screen doesn't seem very accurate.

I'm really at a loss here as to what is happening and hope ya'll could give some insight. She's not much to look at, but here's a few pics of an uncompleted build.

Full-size images:
https://ibb.co/dW9YXHN
https://ibb.co/wcF0Dw8
https://ibb.co/xGWF7tR

 

[amberwolf]

The lowdown is that it has a 1500W hub motor, KT48ZWSRKT-SJT02L, and 50A BMS 10AH battery and I'm getting intermittent LCD-3 shutoffs under load when I apply the throttle - even gently. It accelerates as fast as a car and I set the speed limit to 30 mph to stay within the law as it will do just that. The LCD-3 shows ~1800 watts being put out when under full throttle load and the Voltage drops to ~43-44 and slowly climbs as speed increases and thus load decreases. I did not see the Voltage when the issue occurred, only when it works properly.

If it sags in voltage that much, presumably from a full charge (52v for a 48v battery, 58v for a 52v battery), then the battery has a problem, either in being old, or defective, or in being not good enough cells to run your system.

Possible defects include cells taht are not well-matched, so they become imbalanced, at different states of charge / voltages, or cells that are not even the same kind / age / etc., poor interconnects within parallel groups so some groups don't have the same number of cells working, bms balance channels taht are stuck on and constantly draining cells, etc.

If it works sometimes and not others, there is probably a poor connection between battery and controller, or inside the battery somewhere, causing high resistance and excessive voltage drop. This may cause obvious physical damage (melting, discoloration, etc), depending on the severity of the problem.

 

[Zarathustra1984]

This is the battery (10AH version sold out).
https://www.amazon.com/gp/product/B07L56QNFJ/ref=ppx_yo_dt_b_asin_title_o09_s00?ie=UTF8&psc=1

Picture shows Li-on, but mine has a LiFePO4 tag on it. They sent a cheapo 3A charger and I had to get them to send the 5A, so maybe it's a bad company with junk cells? Paid $200 and am using xt60s.

This one I'm looking at here says it has a 30A continuous BMS - am I going to get the same performance from the 22A continuous controller assuming the cells are good (Li-on this time)? Says "For 1200w" but I'm putting out 1800 and don't want to fry anything but would like to keep the same speed/accel.
https://www.amazon.com/Battery-Charger-Electric-Bicycle-Protection/dp/B086KKX4B9/ref=pb_allspark_session_sims_desktop_468_32?_encoding=UTF8&pd_rd_i=B086KKX4B9&pd_rd_r=80796dba-83ad-46a6-b439-2404f5de3dcb&pd_rd_w=MgOdd&pd_rd_wg=6m0ng&pf_rd_p=b4f7b0e0-9829-4c13-ba2d-476d670e1a97&pf_rd_r=55TCR5HF5ZF83G2YQNP3&psc=1&refRID=55TCR5HF5ZF83G2YQNP3

Thanks

 

[NCC1941]

This is the battery (10AH version sold out).
https://www.amazon.com/gp/product/B07L56QNFJ/ref=ppx_yo_dt_b_asin_title_o09_s00?ie=UTF8&psc=1

Picture shows Li-on, but mine has a LiFePO4 tag on it. They sent a cheapo 3A charger and I had to get them to send the 5A, so maybe it's a bad company with junk cells? Paid $200.

This one says it has a 30A continuous BMS - am I going to get the same performance from the 22A continuous controller assuming the cells are good (Li-on this time)?
https://www.amazon.com/Battery-Charger-Electric-Bicycle-Protection/dp/B086KKX4B9/ref=pb_allspark_session_sims_desktop_468_32?_encoding=UTF8&pd_rd_i=B086KKX4B9&pd_rd_r=80796dba-83ad-46a6-b439-2404f5de3dcb&pd_rd_w=MgOdd&pd_rd_wg=6m0ng&pf_rd_p=b4f7b0e0-9829-4c13-ba2d-476d670e1a97&pf_rd_r=55TCR5HF5ZF83G2YQNP3&psc=1&refRID=55TCR5HF5ZF83G2YQNP3

Taken from that listing for the LiFePO4 pack:

Recommend the following
300W-350W Motor choose 10-15AH battery
350W-750W Motor choose 15-20AH battery
750W-1000W Motor choose 20-25AH battery
1000W-1800W Motor choose 25-35AH battery
1800W-2000W Motor choose 35-40AH battery
2000W+ Motor choose 40-50AH battery

Sounds like your problem is that you picked a battery only rated for 350w to power your 1800w+ system. I'm not at all surprised that you're running into these issues.

The li-ion pack you linked to also looks underpowered for your kit. It's got a 30A BMS, but your controller can, will, and does pull 45A at full throttle. You need a battery that can handle not just your continuous loads, but your peak loads as well.

 

[amberwolf]

Some thoughts:

The most common LiFePO4 cells I've seen in packs posted up around here are 2C types, meaning they will handle about twice their capacity rating as amp draw. This means that for a 10Ah pack, 20A would be pushing the limits of the cells, and voltage sag will be large (which is what you see).

At 48v, 1800w would be just under 40A, so about twice as much as many of these cells can actually handle without serious voltage sag, often enough to cause the BMS to shtu off to protect them from overdischarge, causing system to power off, or the controller to disable itself due to low voltage, causing cutouts without powering off.

There *are* cells that can handle more than that, but the likelihood is that these are not those.

Based on the description of :

48V E-Bike Battery 10AH - 50AH Lithium ion / Lifeo4 Battery Pack with 5A Charger,50A BMS for 500W-3000W Motor (48V 28AH)

on that listing, and that they have several sizes, with the first available one of 28Ah being listed as a 1000w-capable pack, I would say that the 10Ah pack probably said that it was a 500w-capable pack (the minimum size of this listing), and is not nearly sufficient to run your system.

Based on the choices and info in that listing, you would at minimum need the 35Ah version of that pack to support 1800w. (and that is when the pack is brand new--as it ages, voltage sag gets worse, so getting *at least* the next size up (50Ah) would be adviseable, as even that is only rated to 2000w). They don't list the 3000w version, so that's probably out of stock like the 500w version you already have.




If you want to see if it's just imbalance (probably not), you can leave it on the charger whenever it is not in use for at least several days (may take weeks if it really is an imbalance problem that's very bad).

If it sags less and less each time you do this then it means there are cells inside that are problematic (mismatched or defective), and the problem is just going to come back unless you always leave in on the charger when not in use (and as it ages, it'll get worse).

If the sag doesn't change, then the pack is simply incapable of handling the load your system puts on it.



So what you really need is a good-quality battery, made of good cells that are well-matched. The cheaper you go, the bigger a battery you have to use to support your systems' needs, so you end up paying more for physically huge and heavy low-quality battery that then reduces your system's performance and is hard to mount to the bike,

 

[Zarathustra1984]

Sounds like your problem is that you picked a battery only rated for 350w to power your 1800w+ system. I'm not at all surprised that you're running into these issues.

The li-ion pack you linked to also looks underpowered for your kit. It's got a 30A BMS, but your controller can, will, and does pull 45A at full throttle. You need a battery that can handle not just your continuous loads, but your peak loads as well.

According to the Amazon seller, the current battery cells I have are 3C. I assumed the AH rating (their suggestions seemed to be predicated on this) has no bearing on actual performance (outside of distance) as evidenced on my first night's ride (and today). Just that the run time would be lessened by the lower capacity. Does an increased # of cells lessen the sag? I will note that when the shutoff issue occurred, it would happen at just over 1000w of power input, yet now it pulls 1800w just fine.

The 2nd battery I linked had a 30A BMS rating for continuous and I messaged them to find out about the peak. I assume if the controller can do 22A cont and 45A peak, then a BMS with 30A cont and ~50A peak would be sufficient, yes? That's 5 extra AH.

The smaller size of the triangle due to the rear shock has made this a PITA. Will an el cheapo charger be able to balance cells if this were by chance a part of the issue?

 

[amberwolf]

Well, I assume the AH rating (their suggestions are predicated on this) has no bearing on actual performance (outside of distance) as evidenced on my first night's ride (and today) - do correct me if I'm incorrect.

This is incorrect, as explained in the previous posts, *and* by the listing itself. Unfortunately this is a lesson that is commonly learned the hard way; most of us did so. :/

Some cells can handle higher discharge rates than others, but waht you have are low-current cells, so you need MUCH higher Ah to provide the A you need to run your system.

Note that you have fairly extreme voltage sag in your measurements: If it is fully charged at 52v, and drops to 43v under 1800w load, then that is 9v of drop, across (probably) 16 cells in series.

If it's 52v full, and 16s, then that's 3.25v per cell (which is a little low for LiFePO4, usually 3.3v "full" and 3.6v hot off the charger), 9v across 16 cells is 0.57v of drop per cell, which is quite a lot for a LifePO4 cell that's basically already in it's "flat" section of a discharge curve. That means each cell is dropping down to about 2.7v under load. LiFePO4 cells are pretty empty at 2.8v, and usually LVCs are around 2.5v or above.

So your pack is sagging so much under load that it is very near shutting off, so when it discharges a little bit it's actually doing that, either at the BMS (if it actually powers off and display blanks) or at the controller (if it doesn't turn off but just stops responding)



FWIW, the BMS rating isn't a measure of what the cells can output, it is only a measure of what the BMS can handle before it fails from overheating the discharge FETs, or shuts off to protect itself (if it has that functionality--not all do).

 

[amberwolf]

Will an el cheapo charger be able to balance cells if this were by chance a part of the issue?

It's not really up to the charger, as long as the charger outputs enough voltage to fully charge the pack up to the point the BMS is setup to begin balancing. (typically around 58v for a 16s LiFePO4 "48v" pack).

The BMS, if provided enough voltage from the charger, and if it has a balance function (many but not all do), will eventually fix a balance issue.

But that's probably not the problem you have here. Easy, if timeconsuming, to test, though.

 

[Zarathustra1984]

Well, I assume the AH rating (their suggestions are predicated on this) has no bearing on actual performance (outside of distance) as evidenced on my first night's ride (and today) - do correct me if I'm incorrect.

This is incorrect, as explained in the previous posts, *and* by the listing itself.

Some cells can handle higher discharge rates than others, but waht you have are low-current cells, so you need MUCH higher Ah to provide the A you need to run your system.

Note that you have fairly extreme voltage sag in your measurements: If it is fully charged at 52v, and drops to 43v under 1800w load, then that is 9v of drop, across (probably) 16 cells in series.

If it's 52v full, and 16s, then that's 3.25v per cell (which is a little low for LiFePO4, usually 3.3v "full" and 3.6v hot off the charger), 9v across 16 cells is 0.57v of drop per cell, which is quite a lot for a LifePO4 cell that's basically already in it's "flat" section of a discharge curve. That means each cell is dropping down to about 2.7v under load. LiFePO4 cells are pretty empty at 2.8v, and usually LVCs are around 2.5v or above.

So your pack is sagging so much under load that it is very near shutting off, so when it discharges a little bit it's actually doing that, either at the BMS (if it actually powers off and display blanks) or at the controller (if it doesn't turn off but just stops responding)



FWIW, the BMS rating isn't a measure of what the cells can output, it is only a measure of what the BMS can handle before it fails from overheating the discharge FETs, or shuts off to protect itself (if it has that functionality--not all do).

Thanks for your help, Amberwolf. Assuming I buy that 2nd listing I posted of the 5 extra AH in Lion this time, would I get any benefit to the sag assuming a good C rate? Battery life isn't really an issue, the space of the triangle is. I don't want Lion to explode between my legs doing 30 mph

 

[amberwolf]

This one I'm looking at here says it has a 30A continuous BMS - am I going to get the same performance from the 22A continuous controller assuming the cells are good (Li-on this time)? Says "For 1200w" but I'm putting out 1800 and don't want to fry anything but would like to keep the same speed/accel.

Then you need to get a pack that is designed to handle *at least* the power levels your system will demand of it.

If you get a battery unable to handle the power levels, then voltage will sag under load, and you will either expereince the same problems you have now (although at a lower state of charge than at present, meaning when your battery is closer to empty than full), or you'll overheat the battery (and BMS) and damage it, over time, or cause the cells to become unbalanced easier and more often than they otherwise would.

More likely, all of these will be a problem.


Your other option is to see if you can go into your controller settings and limit it to what the battery is actually designed to handle. But this will reduce your accleration and performance, which you said you don't want to do.

 

[Zarathustra1984]

From the controller pdf: "P3 parameters affect the power assist control setting. When the value is set to 0, the throttle is dependent on the PAS Gear Ratio. This means when the PAS gear ratio is 1, the throttle will provide the least amount of power, whereas when it is 5, the throttle will provide the most power. If the P3 parameter is set to 1, the throttle will provide the maximum power regardless on the PAS gear ratio chosen."

I'm going to change the max throttle power to PAS level 3 or 4 and run it like that. That Lion battery (5 more AH) won't be any more likely to overheat than the LiFePO4, will it? I really don't have the space for more cells unless I were to find someone to make a custom to-fit pack for my triangle in order to squeeze out a few extra AH.

Thanks again

 

[amberwolf]

Assuming I buy that 2nd listing I posted of the 5 extra AH in Lion this time, would I get any benefit to the sag assuming a good C rate?

Unfortunately, "good C rate" doesn't mean anything by itself.

None of this is really what you probably wanted to hear...but reality often damps our expectations.

Something else to consider is that sellers often classify batteries with whatever C-rate they decide to, not what it's actually capable of, so you should generally be pessimistic with ratings, and never optimistic. Meaning, if they call it a 2C pack, you could consider that 2C is it's *maximum* capability, and that 1C is it's *continous* capablity. Then you will get a more realistic expectation from the pack, and if you're lucky, it will turn out even better than you expected.

If you go with optimistic ratings, and they call it 2C continous, but it really isn't, then you end up with a pack that sags greatly in voltage under those conditions, and makes your system perform badly, making you unhappy.

If sellers were realistic about what's actually in their packs, and built them all consistently and with good processes and parts, and matched all their cells well and used good quality cells (or at least rated the not-so-good stuff appropriately), we wouldn't have to worry about all this. But they don't. :/



So, to get the performance you want, you should get a battery that is capable of providing *at least* the full power you're going to use, as often as you're going to use it, even when the battery is nearly empty. This usualy means over-spec'ing the battery, getting one rated as capable of *more* than you need by some significant margin.

If it can't, then it will heat up inside (how much depends on the cells, interconnects, casing, external environment, etc, and whether that is a problem depends on the cells themselves), and it will sag in voltage enough to cause shutdown once the pack drops below a certain voltage level (when it gets empty enough), whenever the load is high enough.

In your case, this means something that can supply at least 1800w even when nearly empty, without significant voltage sag, and I would recommend something "rated" as say, 25-50% over what you think you need, at minimum.

Sometimes it's simply not possible to fit the battery needed in the space available, using the desired type of battery. Options in those cases include buying two smaller batteries and paralleling them, fitting one in the space you wnat to, and the other in some other space. This has other potential complications, so is best to get them with a single charge/discharge port rather than separate ones, to reduce those. Another option is to use one bigger battery placed somewhere else on the bike. Another option is to use a very high-rate battery, but these may be more volatile (like RC LiPo) with much shorter lifespans. Another option is to live with the problems an under spec'd battery will cause.

 

[amberwolf]

FWIW, using a middrive that you can shift gears on (like BBSxx, etc), if you are willing to actually do the gear shifting at all appropriate times during acceleration, etc., will use less (sometimes a LOT less) power to do the same acceleration rate, speed, etc., for the same battery. (if you don't shift gears to the right gears at the right times, it won't help as much, if at all).

Whether the battery you already have would do it, I don't know, but it would probably be easier on it.

However, a middrive like this is higher maintenance (can quickly wear out sprockets and chains), and harder to setup (chainlines, etc), and has more failure modes (potentially less reliable) than a DD hubmotor in a wheel like you have now.


Something to ponder for the next bike.

 

[amberwolf]

I notice you've now (after all of the above discussion) edited your first post and title to show you only draw 800 watts, instead of the 1800 you first reported, and that you only have a 500w hubmotor, rather than the 1500w motor you first reported.

If that's true, then your battery pack is EVEN WORSE than you first reported, if it's sagging so many volts at only 800w when closer to full charge, and shutting down at those power levels when further discharged.

Just so you know.


If you're editing it to make the problem report more accurate, that's helpful in order to help you troubleshoot your problem...but your controller, the KT48ZWSRKT-SJT02L, is still basically a 2kw controller, and is still going to draw ~40A+ at full throttle to give you the same acceleration/etc you got before, when using it in modes that let you get the speed and acceleration rates you want (what you get at the higher power levels).


But if you're just editing it to make the problem you have look better (which is something some people do, for whatever reason), it's not going to help us help you fix your problem, because it means your system will be drawing a lot more power (more than twice as much) than you are actually reporting, so anything new you get based on that lower power requirement will perform terribly at the real higher power usage, and any advice provided to you based on the incorrect information will not be helpful to you. Also, if your motor is only a 500w rated motor, for real, not the 1500w you'd originally reported, it's possible that eventually it will have problems at the much higher power usage (nearly four times it's rating) your controller can actually put it thru.


Remember, we're trying to help you get a system that does what you want it to, so accurate and complete information from you is very important to getting you stuff that meets your goals. It is also important that rather than just editing in new information into your old posts, that you instead correct that information in a new post to advise us of it. (I only noticed it because I was looking for information for someone else with a similar problem and remembered this thread).



Quoted below is what we had already quoted of your original post:

The lowdown is that it has a 1500W hub motor, KT48ZWSRKT-SJT02L, and 50A BMS 10AH battery and I'm getting intermittent LCD-3 shutoffs under load when I apply the throttle - even gently. It accelerates as fast as a car and I set the speed limit to 30 mph to stay within the law as it will do just that. The LCD-3 shows ~1800 watts being put out when under full throttle load and the Voltage drops to ~43-44 and slowly climbs as speed increases and thus load decreases. I did not see the Voltage when the issue occurred, only when it works properly.

Quoted below is your new full post as of
"Last edited by Zarathustra1984 on Jul 03 2020 4:43pm, edited 4 times in total."

Hello everyone. Long-time lurker and first time poster here. I appreciate the wealth of knowledge you all have given and it's helped make my dream bike (cheapo) come true. This is my first build after a period of about 10 years not owning a bike and want to give thanks.

The lowdown is that it has a 500W hub motor, KT48ZWSRKT-SJT02L, and 50A BMS 10AH battery and I'm getting intermittent LCD-3 shutoffs under load when I apply the throttle - even gently. It accelerates as fast as a car and I set the speed limit to 30 mph to stay within the law as it will do just that. The LCD-3 shows ~800 watts being put out when under full throttle load and the Voltage drops to ~43-44 and slowly climbs as speed increases and thus load decreases. I did not think to see the Voltage when the issue occurred, only now.

The first night I took it out, I went about 15 miles stop/go driving in the city with mostly full-throttle rips and speed limited; had a great time. Last night, mostly the same, but didn't drive it as hard, and around the similar mileage, the system shut down when applying throttle from a stop - it made it about 10 feet and the LCD-3 shut off. I unplugged and plugged back in the battery and I was able to use P.A.S. mode to limp it home. Any time the throttle isn't babied (like 1/10th of full twist) and it eased up to speed, it would immediately shut off the system. P.A.S. #5 will cause a shutdown. Once, I was able to get the LCD-3 to flash the battery meter being empty and then a slow fading shutdown - this on at least 75% battery. I gave it a full charge overnight and the same thing this morning for ~10 miles. Now, here the kicker: I had it on the charger for a while just now and decided to make a quick run to the store and it read ~52v and was able to apply full throttle again - even from a dead stop as if the issue went away. Note, this morning the system was completely cooled and just now it was plenty warm. I'll add that the "Regenerative Braking" only kicks in for about 1 sec when it's on a stand and also that the battery meter gauge in the upper left-hand of the screen doesn't seem very accurate.

I'm really at a loss here as to what is happening and hope ya'll could give some insight. She's not much to look at, but here's a few pics of an uncompleted build.

Full-size images:
https://ibb.co/dW9YXHN
https://ibb.co/wcF0Dw8
https://ibb.co/xGWF7tR

BTW, we don't delete threads (unless they are pure spam), even by specific poster request, because future readers will learn from previous posters' problems and solutions.



EDIT:: added this quote from another thread, showing that the edit to the title and first post lowering the power usage must just be to mess with this thread, or those of us trying to help, vs correcting information, because it is in the originally-posted power range. Not really very cool. Oh, well.

Thanks again Amberwolf, and LeftieBiker. FWIW, regarding yesterday's thread; took it out today and it immediately died with throttle, so I ended up reducing the controller amps from 22 to 20 on the LCD3 and accel/speed remained the same with no more shutoffs :wink: @ ~47v sag. 1800w became 1500 with a 30 second fix. Rode the 10AH 15 miles basically 30mph non-stop in the city until the battery (lifepo4) dropped at the expected time - then it shutoff on a throttle rip. $200 battery; Can do