Battery selection question

[Earl]

I recently bought a 500 watt, 36 volt rear wheel drive (26" wheel, w/7 speed sprocket) kit that I expect to be delivered in the next week.
500 watts @ 36 volts is 13.88 amps load. I'm considering assembling a battery consisting of 10 series banks of 3 cells parallel of 18650 lithium ion cells. (a total of 30 cells)

My question is: I see cells rated for constant discharge rates of 5, 10, 15, 20 etc. amps however...
It appears to me (absolute newbie here) that if I have 30 cells and the total draw/load is13.88 amps, then the sustainable discharge rate required per cell would be 13.88/30, or .46 amps per cell. If that is correct, then there is no point in buying high rated discharge (expensive) cells.

maybe i'm missing something.

 

[Earl]

clarification: I meant 3 cell banks in parallel and 10 banks of those connected in series for 36v.

 

[Nealh]

Not quite !!!
Your cell rating will depend on the max controller current and a bit of headway (30% or more, 50% would be better) so as not to stress the cells any where near there CDR, this helps to extend cycle life without knackering the cells.
A cell operating at near it's Continuous Discharge Rate (CDR) will get very warm or hot, this is bad and will induce higher internal resistance which in turn will cause bad or severe voltage sag. Both of these together means a cell is past it's best.
As you rightly mention your 36v battery will have 10 parallel strings of 3 cells each string connected in series.
Your current draw is from each cell string so you have to divide your max controller rating by 3, this is the max current each cell will need to deliver but as mentioned you will need a cell rated a bit higher. A cell with a higher current draw will run cooler then a cell with low current draw.

Example;
Your controller max current supply is 17a so each cell in the 3 string parallel group needs to be able to handle 5.66a each. In this case you would buy the best current cell you can afford so ideally a 10a cell or better. The more cells in parallel the easier life the cells have.




You need to do your math again and start with your max controller current.
Show us a link to your kit.

 

[Nealh]

Continued......
Each cell string of 3 x 10 gives you the 36v nominal rating.
String #1 is 3.6v nominal and needs to be able to deliver 17a,
String #2 is connected in series for 7.2v 3 nominal and needs to be able to deliver 17a,
String 3 again in series for 10.8v and needs to be able to deliver 17a and so on and so on.
The voltage is accumulative to get the nominal 10s/36v but the cell string amp/current rating isn't , this is static as the current just flows from one end to the other.

 

[Earl]

Thanks! Ahaaa, I thought it might not be a simple as I am. LOL So then, it appears a cell with a CDR of 15 amps would operate cooler than a cell with a CDR 0f 10? Brings to mind another question, which would run cooler then, a bank of 3 cells with a CDR of 15 or a bank of 4 cells with a CDR of 10 assuming capacity of the cells is the same.

The control that comes with the kit will handle up to 48 volts, but I don't know the amperage it will sustain. I'll know that when I receive it though.

I've been piecing electrical information from various sites, but I haven't found it to integrated to my purpose, so it's been a mix and match proposition, which makes it a bit difficult to get a comprehensive take on the ebike application with lion cells.

 

[Earl]

Continued......
Each cell string of 3 x 10 gives you the 36v nominal rating.
String #1 is 3.6v nominal and needs to be able to deliver 17a,
String #2 is connected in series for 7.2v 3 nominal and needs to be able to deliver 17a,
String 3 again in series for 10.8v and needs to be able to deliver 17a and so on and so on.
The voltage is accumulative to get the nominal 10s/36v but the cell string amp/current rating isn't , this is static as the current just flows from one end to the other.

OK, the light is starting to come on. heh So, a reasonable minimum would be cells with CDR20 for brief full throttle and half throttle would result in half that rate?

 

[Deleted member 25121]

There seems to be some confusion here with terminologies, series, parallel, bank, string...

I think you're planning to make up blocks of cells with each block containing 3 cells wired in parallel. If each cell could provide say 3.6V at up to 5A then each block could provide 3.6V at up to 15A.

You are then joining 10 of these blocks up in series so the total assembly could provide 36V at up to 15A.

Regarding your 500W 36V motor, in very simplistic terms the 36V is the maximum voltage you should apply to the motor and the 500W is the average power the motor is designed to consume but it could take much higher power for a short time. It's the job of your controller to control the power applied to the motor and it does this by regulating the current being supplied. A "17A" controller would provide a maximum of 17A to the motor and the actual voltage applied would depend on the resistance of the motor windings and the load on the motor. It could be significantly less than the battery voltage.

That's a very simplistic explanation but I hope you get the picture.

 

[Nealh]

Until you know the controller max current then you won't know which cells to consider. Other considerations are the miles/range you expecting to get out of the bike, a 3P 10s battery can be 8.8a/316wh - 10.6ah/381wh with 18650 or quite a bit more if you use 21700 cells.
Range calcs will vary widely too many variables to factor in, generally you can allow 6-10 wh per mile for a regular fit person, an average fit person up to 15wh per mile and really unfit 20wh+ per mile.

 

[vfr400]

The 500w kits normally have 22 amp controllers, and if it's a direct drive motor, it'll be drawing the maximum current a lot of the time. If you use Samsung 30Q cells and a 30 amp BMS, you should be OK with a 3P battery, but it won't give you much range. It's true capacity will be around 7.5Ah, which would give about 12 miles of normal use.

As a general rule, whatever the maximum current stated for the cells that you want to use, divide it by two, so run 10 amp cells at no more than 5 amps. Make sure that your BMS has at least 5 amps of headroom too. I'd be looking at a 40 amp BMS for your kit.

It's not worth the hassle and expense to make a single battery unless you need a custom shape. It's nearly always cheaper in the long run to buy a ready-made one.

 

[Nealh]

The control that comes with the kit will handle up to 48 volts, but I don't know the amperage it will sustain. I'll know that when I receive it though

The controller might be a dual voltage model, if it is only a 48v one then you will have to use a 13s/48v battery otherwise you will be hitting the LVC of 40v very early on if you use a 36v battery.