max current draw on standard bike

[flippy]

first post, please be gentile. (or use lube)


while scouring the forum for all kinds of data (johnny must have more input!) i seem to drawing a blank on a few.

situation is as follows: got 2 bikes (identical, parents bought the secondhand) with bog standard 36v crappy chinese hub drives and ditto shitty lipo battery pack. cells are totally shot (3Ah of 10 left) so i can't do any solid current draw measuring because the cells go out to lunch when asking full power.
is there someone that knows what the max current draw is on these standard chinese 250w hub systems? i need that info in order to choose the correct cells to replace the busted ones.
case in point is that i can replace the 10 prismatic 10Ah cells with 6 18650's in order to get the capacity way up. current cells are physically the size of 7 stacked 18650 cells but with the cell holders i can only get 6 in.
so i am stuck with 10S6P pack size. for full power support i am running at a solid 7A or 1.3A per cell. that is no problem for any cell, even the NCR18650B can handle that. but the max draw when accelerating is something totally different. the only value i found was in the 650W range and that means 18 amps (3A per cell) wich is borderline on the 1C value of the B panny's. it would still be fine but the higher internal resistance would heat them up on long runs and i would lose a lot of capacity and it would wear them out faster.
better choice would be the PF's with half the IR (0.06) but a tad less capacity and considerable nicer price tag and would take the high current with room to spare.

is 650W a reasonable max current draw or do 250W hubs really use a lot more when accelerating? on one side i would love to use the 20Ah capacity of the B's, but on the other side the PF's would give me 17Ah and safer current handeling and lower price.

ps: here is a pic of those weird prismatics:

do not ajust your television set, these cells are really puffing.

fun facts:
it seems they are about 10Ah and they are in these 5 packs (2 of them) in the battery box. each one has a bleed resistor PCB on them for overvoltage protection, wich is nice but kinda stupid considering the protection board has balance leads and simply cuts off the charger when a cell even gets near 4.2v so they never fire and that also freaks out the 42v charger because the protection board (no actual balance capacity, just monitoring) simply cuts the charger off and releases again after a second or so, so the pack can never balance itself out. talking about stupid design.

 

[Nobuo]

is your hub direct drive or geared? in the first case it could handle sustained 500W probably. could you hsare a pic or link of the motor?

For the battery discharging rating it would be better if you don't pass over 2C on NMC / NCA 18650 cells if you want more than 90% of battery efficiency

 

[Joseph C.]

Welcome to the forum.

A 400 watt hour pack drawing 650 watts is just over 1.6C.

Your potential new pack has to be 10S6P? Okay, let's say that the Panasonic NCR18650B sell you mention puts out 3.6volts rather than 3.7 and gives you the full 3.4amp hours that's a 720 watt hour battery pack. That's less than .9C. I would be sceptical of those specs.

Either way your new battery should be fine. I'd recommend if you can at all to buy a new charger that has a switch allowing you to charge the battery to 90 per cent. The pack will last an awful lot longer if you do this.

Some people on the forum also like these Pansonic cells - http://www.amazon.com/Samsung-INR18650-29E-2900mAh-Lithium-Battery/dp/B00RIRIJXI
Though I don't know if that links sells the genuine stuff or not.

EM3EV rates that battery at 3.6 volts and 2.75 amp hours. That's still a good 594watt hour pack and the cells are rated for 2C - which should probably be taken as the peak rating. At a peak of 650 watts you wouldn't even be hitting 1.1C.

Edit: You need to check the controller and see what the amp rating is to find out how many watts it is capable of pulling from the battery.

 

[flippy]

Thank you.
Problem is that the controller seems a standard 250w ebay job but any numbers have worn off from standing in the sun, so no specs on that thing. Motor is geared.

Also a problem is that i cannot do a proper current rating as the existing pack is totally shot, it simply goes into low voltage protection when drawing any decent current.
I plan to upgrade stuff like the cables as those seem quite dodgy but not knowing what actual power goes into the motor makes it considerably harder to make proper decisions. Are there any "basic/average" specs on those controllers?

 

[dogman dan]

If your controller was truly limited to 250w max, then it's a 6.9 amps controller, 36v.

But really we are just still guessing. far more likely, it's a 10 amps controller, and you actually have 350w.

I'd just ball park calculate for a 10amps controller. A 15 amps controller they'd have been more likely to call 500w. But they will call anything at all 250w, to sell to Europe.

For wires and connectors, can't go wrong with 12 g wire, and connectors that handle up to 45 amps. Andersons, or 4mm RC hobby bullet connectors are popular for the battery to controller wire, and the big power wires to the motor.

Eventually, you could find out all, by adding a cheap blue watt meter to your rig. Ebay, less than 20 bucks.

 

[Joseph C.]

If the controller is pulling 15 amps from the battery at six cells in parallel each cell won't be even experiencing 1C using the 2.75 amp hour cells - even less if he goes with the 3.4 amp hour batteries.

Even if it is pulling 20 amps it will still be well within peak spec even with the E cells. That's just a peak of 1.2C.

As long as the batteries that he chooses are genuine Panasonics they will be fine. And they should last a long time if they are only charged to 90 per cent or lower. The only thing I would be worried about is the state of the BMS on the old battery. For all we know it could the reason why the battery is ruined.

Edit: Meant 1C not one amp.

 

[flippy]

i know how amps and series/parralel cells work, it's kinda my job.

for proper 90% SOC i need a modified BMS/protection board that triggers on 4.1v otherwise it will never balance properly and kill the cells. that shit is going to be expensive as those don't exist as far as i know. manually replacing the trigger resistor network and reflowing a board is not the nicest way of spending a weekeind.

the bMS in the old pack is working fine, the protection board triggers at 4.21v and 2.74v but the balance boards trigger around 4.23 so that is pretty stupid. i plan on simply rewiring the packs so the charger bypasses the protection board so it can charge and balance the cells properly without the protection board constantly going into overvolt mode. this is imho the main reason why the cells died, and the fact they are really crappy quality and a few years old.

considering nkon is just a 10 min drive from me i plan on getting my cells from there. never heard of nkon supplying fakes.



what is the actual difference between the panny PF's and the samsung 29E? the 29E is 45 cents cheaper in bulk 153 or 180 euro for a pack is quite a difference, not to mention the 240 a pack with B's will cost me. personally i tend to go with panasonics considering the brands reputation and consistenty in cells.

 

[d8veh]

Those standard Chinese 250w controllers are normally set to around 15 amps. That's no problem for 10S4P 18650s, which is a common configuration for OEM 250w Chinese ebikes. The modern ones with Sony, Panasonic or Samsung cells can go as high as 20A, which will almost certainly be enough for your bike. If you want to go as high as 6P, even the cheapest crappiest 18650s will easily be able to cope.

 

[flippy]

Those standard Chinese 250w controllers are normally set to around 15 amps. That's no problem for 10S4P 18650s, which is a common configuration for OEM 250w Chinese ebikes. The modern ones with Sony, Panasonic or Samsung cells can go as high as 20A, which will almost certainly be enough for your bike. If you want to go as high as 6P, even the cheapest crappiest 18650s will easily be able to cope.

thanks for that info, but i plan on using proper 18650's. i don't plan on replacing 150 euro's of crappy cells every odd year, i'd rather build a 200 euro one that will last me a good time. if those 15 amps limit is true i can get away with using panny B's without breaking them and maximize range or just crank the assist on high all day long.

 

[dogman dan]

I mis read and thought you meant per 10 ah cell, or cell group of 10 ah. Lots of people write cell, when they are really talking about a group of cells in parallel.

My apologies. Now I'll go delete that stupid comment I made.

Spend the 15 bucks or so, and find out what amps your controller actually pulls. But it sounds like you are on track for a battery that could handle a 20 amps controller with ease.

 

[flippy]

I have better stuff then those watt meters, i have proper fluke meters (yes more then 1, it's a real problem for fluke owners ) and a few amp clamps. I used amp clamp with inrush mode and a trusty old fluke to register the lowest voltage drop and the highest current but the existing packs i have are totally shot. They go into low volt mode as soon as i load it with anything more then 5 amps on the bench.

Its just that i would like to know the limits the pack will see but i cant duplicate the packs 36v on the bike. My bench power supply maxes out at 31v and is quite difficult to mount on a bike with a proper half mile extention cord to test the power draw.

But so dar i really appriciate the input. I know my electronics and batteries/cells but i lack the experience in the ebike/scooter department that is present here.

 

[dogman dan]

Well, since you have a meter, measure it. Get your max amps by grabbing throttle and brakes at the same time, very briefly. You can see no load amps by letting the wheel spin in air, but that number won't tell you a lot really.

Not sure how complicated it will be for you to ride around with your meter, and get readings for load while you cruise full speed.

A watt meter on your bike will allow you to see amps on the fly, or look at max amps later. The best bike watt meter is the cycleanalyst. I didn't suggest a watt meter on your bike because it was dumb to know what your battery is doing while you ride.

 

[d8veh]

If it's a standard 6 FET controller, which it most likely is, I can almost guarantee that your max current is no more than 15 amps. Show a photo of it and give us the measurements if you're unsure how many FETs.

 

[flippy]

Well, since you have a meter, measure it. Get your max amps by grabbing throttle and brakes at the same time, very briefly. You can see no load amps by letting the wheel spin in air, but that number won't tell you a lot really.
Not sure how complicated it will be for you to ride around with your meter, and get readings for load while you cruise full speed.
A watt meter on your bike will allow you to see amps on the fly, or look at max amps later. The best bike watt meter is the cycleanalyst. I didn't suggest a watt meter on your bike because it was dumb to know what your battery is doing while you ride.

problem is that as soon as i draw full power or anything above 5 amps it trips the low voltage of the pack. the existing pack simply has packed it in. (get it? pack..packed.... aaah,your own jokes....)

it's a 6 fet controller. the chinese kid that solderd it was in a hurry so i plan on doing some free upgrades, beef up the tracks and mess a bit with the shunt to get the amps flowing freely. as someone that is used to work with high end millitary spec electronics i am slightly going nuts from the poor quality of these electronics. it makes you think how the chinese ever got this far as making all our stuff. i feel the thing will end in a great ball of flames if i put power on it.

 

[d8veh]

You can take it to about 18 or 19 amps by soldering about 25% of the shunt. Don't go any more unless you can keep it cool.

 

[dogman dan]

Ahh, can't measure if it stops running. You should get a max amps in the first half second, but it might be way low if the battery just can't deliver more than 5 amps before it quits. You just get max amps of the battery.

I agree, it's just not going to be over 15 amps. and likely enough even lower.

 

[flippy]

great stuff, guy's. tomorrow i will recieve 120 new 29E's from nkon and later this week i will get 4x5 cell holders and nickel strip from ebay and hopefully a few days later my brand new cheap ass aliexpress sukko 709 spot welder (with welding pens!) so i can make some nice looking packs.

this upgrade is mainly a dry run for a 390 cell 4kwh lipo upgrade i will do to a scooter (qwic emoto 87) that i bought last weekeind for 100 euro's

i will make pics and stuff about the upgrade and pack building. i know you guys seem to love posts with lots of pretty pictures with expensive cells on them.

 

[flippy]

good times:

the spot welder got caught by customs so that is delayed a bit cause as their customs payment systems is from 1776.

 

[wesnewell]

Just got around to reading this today. In reference to your old pack and it stopping and starting during charge, that's normal for an out of balance pack. And if it's way out of balance, it may take days of charging cycles to get it back into balance. If you're pulling the charger off when it starts cutting on and off, you're not letting it balance the pack. Unless that's a 10 year old pack that's been really abused, I can't see it just having 30% of it's original capacity. I'm just guessing here, and you may already know this, but it doesn't sound like it to me. Once the pack is balanced, the charger should not cycle much more. BTW, those cells don't look that puffed to me.

 

[flippy]

i actually tested 3 cells independently from the pack with a imax b6 and also balanced the entire pack with my bench power supply. 2 of the 3 cells were at 30ish percent and another was at 50%. there was also massive difference in "puffyness" from cell to cell. 2 cells in the pack actually have high self discharge. most of them sit around 3.8v full but those drop to 3.3v after a day or two.

 

[d8veh]

They're Phylion cells. They're pretty well always a lost cause when it comes to battery problems. They seem to lose capacity for no reason. I've had several of them for repair, where the cells were all at 4.2v. As soon as I put a load on, the voltage collapsed and then bounced back when the load was removed. Now, as soon as I see them in a battery, I don't even test them. They go straight in the bin.

 

[flippy]

i had come to the same conclusion, these cells are utter shit. they are the same size as 7 18650's but only deliver the same power as 3 of them. why they are used at all is beyond me considering the alternatives.

 

[flippy]

finally got the spot welder and some other toys like the bigger 4x5 cell holders.

call me enthusiastic but i already built 1 pack and is charging now. it does take considerably longer with the 500% increase in capacity

decided to make pics of the second pack wich is now halfway done for your entertainment and scrutiny:

yes, i am anal about being able to see cell coding from the outside.

cells are all perfectly balanced and from the same batch and most of them (about ten excluded) even came from the same machine according to the coding on the cells.

also found out the balancing boards are utter shit. some fire at 4.23 and others on 4.21 and some even had a problem as they had 0.006 amps of constant draw. tossed those out and replaced those with a ebay balance board i orderd months ago for a project box that never worked.