Homemade Battery Packs

Here's a link with info and lots of pics. An owner of an Izip E-bike that has a 5P/10S downtube pack that was worn out, tears it apart and rebuilds it. Went from 2000-mAh cells to 3200-mAh cells, so stock 37V with 50% more range.

http://www.reddit.com/r/ebikes/comments/36wjv7/rebuilding_a_2011_currie_izip_ultra_battery_pack/
DSCF5357.JPG
 
jaunty said:
Allex said:
I would not even use those for my flashlight :)
Not to mention high current ebikes

actually, if you dont me asking. Why not?

Inside are the UR18650F sanyo cells. Are those bad news?

Cheers

It all depends on what currents you will consume.
If we take one 25A controllers, which about is a minimum the guys here use.
The cells you have are only good for 1c cont. and 2c peak. On top of that, they are old and probably have around 2000mah left with high internal resistance - big voltage sag under load and a lot of heat.
So if you have 25a controller this means you need to have at least 10 of them in parallel. And if you want to be nice to them, at least 20. Here is a datasheet for your cells
https://www.neuhold-elektronik.at/datenblatt/N5644.pdf
 
spinningmagnets said:
Here's a link with info and lots of pics. An owner of an Izip E-bike that has a 5P/10S downtube pack that was worn out, tears it apart and rebuilds it. Went from 2000-mAh cells to 3200-mAh cells, so stock 37V with 50% more range.
Thanks, I was just there a few days ago following the 788 welder repair. I very timely blog! I missed this bit. Very encouraging repair to follow. It also made the bike more attractive and worth keeping in mind especially when someone gives up on it due to battery cost. Not the best rides out there but surely serviceable. I still think there is a HUGE market for a rebuildable battery system based on 18650 format. And so I keep looking and dreaming...
 
Allex said:
jaunty said:
Allex said:
I would not even use those for my flashlight :)
Not to mention high current ebikes

actually, if you dont me asking. Why not?

Inside are the UR18650F sanyo cells. Are those bad news?

Cheers

It all depends on what currents you will consume.
If we take one 25A controllers, which about is a minimum the guys here use.
The cells you have are only good for 1c cont. and 2c peak. On top of that, they are old and probably have around 2000mah left with high internal resistance - big voltage sag under load and a lot of heat.
So if you have 25a controller this means you need to have at least 10 of them in parallel. And if you want to be nice to them, at least 20. Here is a datasheet for your cells
https://www.neuhold-elektronik.at/datenblatt/N5644.pdf

Thanks Allex...
I was making an incorrect assumption that 13S batteries would all contribute a small amperage. I'm new to circuits, so I was modeling these as current sources, which isn't the case. They're actually modeled as voltage sources right?

So all the cells in a single 13 series pack, would experience the same current (20A) if using a single pack?
13S-0P? So at 13S-10P each series pack would experience 2A if the motor was pulling 20A? That seems high though. That would make me think that the first cell in series would bear the least burden? :?

Cheers
 
1st
Laptop cells are designed for higher capacity but very modest discharge rates!

13s10p = 48.1V 26Ah battery pack.
When new:
A 26A discharge = a 1C discharge rate
26A discharge ~equal from each cell, if reasonable gauge wire used for parallels
A 1C discharge is reasonable but voltage sag is a definite factor.
≤.5C discharge rate recommended ...
Sanyo Discharge.jpg

However! ... "C rate" IME (In My Experience) is a factor-measure of actual capacity rather than original rated .
If cells at ~50% capacity ... then a 26A discharge pushes the sag voltage to the 2C line of the chart!
Perhaps worse, deterioration is progressive, based on higher C rate discharges.
So, usable capacity-cycles would deteriorate at a greatly accelerated rate.

Also, Charge rate is recommended as ≤ .5C = 13A, when new, but should not exceed .5C of actual capacity = 6.5A for 50% capacity cells. (26Ah battery)
(Determined-confirmed by monitoring-noticing pack heat production from 10A charge into 26Ah pack over the years, as pack capacity deteriorated below 20Ah actual)

Heat is bad!
Actually damaging as well as an indicator of damage-abuse.
Unlike most every other battery chemistry ...
At moderate charge\discharge rates, Li-ion (LiCo) exhibits no notable heat production.
 
DrkAngel said:
It appears you are building parallels first, then running in series using a single connection. ... ?

I, personally prefer to build in series first, so there would be 8 low current series connections, evenly distributed current.
Then I run parallels, creating redundancy and equalization functions ... lighter gauge wires required.
Full current is then combined and collected into the high capacity, heavier gauge, "end" power rails. (Of gauge substantial enough and short enough to minimize voltage drop as a factor )
UL rating is:
14ga = 15A
12ga = 20A
10ga = 30A
8ga = 50A
I tend to double the UL rating for maximum surge-brief throughput and consider line drop as a negligible factor.
14ga = 30A
12ga = 40A
10ga = 60A
8ga = 100A

7s10p 18650 diagrammed - 25.9V 26Ah (w/2600mAh cells)
Li_SP.jpg

The tinned copper braid I use weighs more than 3 strands of 12ga copper and should be good for 100A, so is overkill on my 40A fused pack.
Oh yeah! Don't forget to add a fuse to your pack! Could save you a whole lot of heart ache!!!

You could easily up your equalization factor with dual series connections ... ? Or attach in center of parallel ... ?
Output connections will be 13s apart!



My typical method would have 8 series connections.
As mention in the Wiki article, I tend to build in parallel 1st for large bulk packs only and would use multiple series leads ... providing redundancy, equalization and the necessity for lighter gauge, easier to work and safer to solder to cells, wire.

Going back to this. This is a really nice layout. I didn't really realize that it was even possible to lay the pack out like this, and I still don't know a lot about this, but this seems as "optimal" as possible as far as how things are connected.

But I'm fairly committed to imitating the (old?) Zero battery design, which seems really robust/hardy, even though it is solderless/weldless and relies on tension/compression from rubber tube rings.

IMG_50151-e1328222885868-199x300.jpg


I'm thinking of getter larger-gauge copper braid. Mine is 32A rated, and I feel like it's at least 25% smaller than I'd like it to be. Trying not to be wasteful, but I don't want any major flaws in my battery design. I'm not sure what surge currents to expect. My controller is rated for 25A max, but I'm not sure if that means that's the maximum that the pack will ever see.

And on that topic, which sort of fuse would you recommend, and where in the pack would you put it? Is putting fuses between the parallel groups overkill? (I think cell-level fusing would be pretty neat too, but that seems like a pipedream for this pack)
 
ASK1 said:
But I'm fairly committed to imitating the (old?) Zero battery design, which seems really robust/hardy, even though it is solderless/weldless and relies on tdension/compression from rubber tube rings.

Thanks! New info for me. I'm looking diligently for more information and pictures. If you would be so kind and had any links or photos close at hand, please share. Such designs make total sense to me and seem to be rare. I just can't imagine why such a kit design wouldn't be a massive success.
 
tomjasz said:
ASK1 said:
But I'm fairly committed to imitating the (old?) Zero battery design, which seems really robust/hardy, even though it is solderless/weldless and relies on tdension/compression from rubber tube rings.

Thanks! New info for me. I'm looking diligently for more information and pictures. If you would be so kind and had any links or photos close at hand, please share. Such designs make total sense to me and seem to be rare. I just can't imagine why such a kit design wouldn't be a massive success.

Check out this thread, then, which is where I found the idea: https://www.endless-sphere.com/forums/viewtopic.php?f=14&t=59758&start=25

It's the perfect design for me because it's easy to take apart without destroying the pack (although you'll probably need new shrink tubing), and it allows me to put my growing collection of flat bike inner tubes to use. :lol:

Just make sure that your tube is tight enough on your cells. 35mm tube seems like a good fit for 18650s - you'll need something larger for taller cells.

I'll also probably post pictures of my build, once I'm ready to move forward. Cheers.
 
Hello guys :D This is gonna be my first post on this forum and in this amazing thread. I know it is kinda rude to just all of a sudden burst in asking questions but I have looked all over the internet for some info but it is of no use. This is the first place I have found that has people who are talking about homemade e bike stuff and are still alive. I got a conversion kit (36v, 500w) a while back and finally fixed it (the polarity was switched when I got it so the bike would only go backwards). I also was able to make a battery using the lipo cells from mac books but I lost so many in the process to my lack of understanding of how batteries work that I finished with a 37v battery with what I think is 2Ah. I was able to go about 1 to 2 miles before it got down to 33v and it wasn't even strong enough to pull me. I got home and then realized that I had no way of charging the battery. I tried all kinds of chargers but the only one that worked and didn't suck energy from it was an airsoft trickle charger which is only 12v and 600ma. I have just recently ordered 10 laptop batteries to get the 18650s from them and I estimated that if I had them in series of 10 to get 37v and 12 parallel then I would be able to get 23 AH. I am gonna use my BMS from the other battery which is a 10s but the problem is that: I DONT HAVE A CHARGER. I am so confused as to what charger I should get because I don't like to waste money (who does?). I have tried charging the batteries in small packs of 7.4v with my airsoft balance charger but as I look at the voltage it seems to drop more and more (That is how I lost my first couple of batteries). I figured out how to charge with my BMS but non of my chargers seem to what to charge it but instead they want to discharge it. Do you guys have any suggestions as to what I should do? Feel free to ask me any questions too.
 
You need a dedicated 37V Lion charger See -10s 37V Tenergy charger - 2A charger
or
an adjustable switching power supply and adjust to ≤42V. (41V is safer and helps battery last longer)
Adjusting 48V charger to 42V is preferable - Might provide proper current regulation without need for further mod
MeanWell S-150-48 - 3A charger
or MeanWell S-350-48 - 7A charger
recommended as proper

MeanWell clones might work as well ... but maybe not!
 
DrkAngel said:
You need a dedicated 37V Lion charger See -10s 37V Tenergy charger - 2A charger
or
an adjustable switching power supply and adjust to ≤42V. (41V is safer and helps battery last longer)
Adjusting 48V charger to 42V is preferable - Might provide proper current regulation without need for further mod
MeanWell S-150-48 - 3A charger
or MeanWell S-350-48 - 7A charger
recommended as proper

MeanWell clones might work as well ... but maybe not!

Thank you sooo much :D I was looking at the Tenergy charger but never thought about buying it because it didn't look like what I needed :p That's also a nice page ya got there ;)
 
What kind of frame do you have ethan? Im sure myself or someone on here could help you build a pack. I also have waterproof meanwell power supplies sitting here that would work for charging. Getting started with ebikes is a bit of a task...However this is the best place to be for great info and help :)
 
If by frame you mean what bike I have then I have a specialized hardrock MTB. I was gonna originally use my electra cruiser but I got the rear kit and not only would it probably not fit because it has gears on it but I also realized my coaster brake was on the rear wheel which wouldn't even a good brake for an e-bike. I have ordered the batteries and charger and already have a BMS from my first test of batteries so I should be good right? I'll keep you guys updated on my progress.
 
Please note:
This method is hypothetical and will vary greatly with different components and applications.

Paralleling a regulated PS with an unregulated PS to create an equivalent regulated PS, for charging, of combined amperage.

1. Add blocking-directional diode to unregulated PS (add Amp meter, for initial adjustment)
With battery near empty ... (usable or recommended empty ~37V)
2. Set desired voltage then attach Amp regulated PS and note voltage sag
3. Set unregulated PS voltage to ~same "sagged" voltage as regulated, power on and additionally attach to battery
(When initially attached, will output 0 Amps and diode should prevent any voltage backwash)
4. Adjust voltage, (do not exceed final charge voltage!!!), of unregulated PS upwards until amp output is at its limit, or desired output (leave safety margin!)
5 Temporarily disconnect and confirm output voltage is safely below final charge voltage!
To use
You should always engage regulated PS 1st! - disengage last! (or unregulated PS will draw too many amps and fault)

You now have a much more powerful charger that only adds charge during the previous CC portion of the charge cycle.
Most useful for quick "pump ups" when battery is low.
While this can greatly speed up the CC (Constant Current) portion of the charge cycle ... it will not speed up the "Stage 2" CV (Constant Voltage) portion of the charge cycle.
(Some CC portion of charge is converted to CV, but at a higher Amp rate)

Warnings:
Check with manufacturer for maximum charge current of battery and BMS! (might want to bypass BMS with unregulated PS?)
Ensure no charger will exceed final charge voltage!
Monitor charging to ensure no problems, heat etc.
Rig is calibrated for only the pack it was set up on! Should be recalibrated periodically to accommodate normal battery deterioration
 
Wanted to make sure the idea was clearly presented.
A picture can be worth a thousand words ...

file.php

CC (Constant current) phase is partially altered to a continual Amp diminishment.
There might be a true, possibly brief, CC phase as charging transitions to regulated supply only.

Please remember!
This method is hypothetical and will vary greatly with different components and applications.
 
Hey guys, so I received my batteries yesterday as well as my charger. The charger worked like a charm to charge my first prototype battery but I was no longer gonna use it so I took off the BMS/PCB and put it on my new pack. each laptop battery had 9 cells, 3s3p. I used 3 laptops battery plus 1/3 of a left over one in series to get 37v. So far I have made 3 packs of 37v and 7.8AH with one being able to be used. I tried it out and it was so much fun. I was able to go about 15mph :D I want to combine to packs in parallel but I need to know if I should connect each "pack" just - to - and + to + or if I should connect each cell to cell in parallel. The pack to pack method would be much easier but seems too simple. The each cell to cell method would be much harder to do but I feel it is the right way in order to balance the batteries.
 
So I have a 1000W 48V motor and I'm building a 13S10P pack...

Im thinking about what size fuse to use near the battery. something that won't pop even when full throttle from a stop, up a hill.

70A? Anyone used a flippable breaker or something similar? I have found several 48V ~80A breakers for solar systems,,but I'm not sure how clunky they are.

once I get it all hooked up and try and get a peak Amp reading.

Cheers
 
Fuse size?
A fuse is typically to protect your system during some fault, short or other abnormal discharge.
With a 1000w motor a 48v controller might be limited to 40A? (check label or query for manufacturer specs)
If so, fuse should be notably higher to prevent blowing fuse during "normal" operation.
With a 40A controller and reasonably matched components, a 50A fuse might be recommended.
If 50A fuse "blows" during normal operation then progressing to a 60A fuse might be advisable.

60A controller should be used with wiring of 10ga copper, at minimum!
40A controller should be used with wiring of 12ga copper, at minimum!
30A controller should be used with wiring of 14ga copper, at minimum!
Based on 2x the UL safety rating for residential wiring.
 
DrkAngel said:
Fuse size?
A fuse is typically to protect your system during some fault, short or other abnormal discharge.
With a 1000w motor a 48v controller might be limited to 40A? (check label or query for manufacturer specs)
If so, fuse should be notably higher to prevent blowing fuse during "normal" operation.
With a 40A controller and reasonably matched components, a 50A fuse might be recommended.
If 50A fuse "blows" during normal operation then progressing to a 60A fuse might be advisable.


Upping the fuse rating like that is not such a good idea. You should use thicker cables with stronger fuse, or if your controller has slow current limiting, you should use slower fuse. In fact, that's what you should do in first case, use slow fuse on EV, but with proper rating according to cable area.
 
You up the fuse rating till normal operation does not cause it to "blow".
Slow blow fuse recommended.

Well ... of course you make sure your wiring will handle the amps you put through it!

But then, you fuse slightly higher than your maximum operating current.
This will cause the fuse to protect the motor, controller and battery in the case of any accident or major malfunction ... much more important than protecting the wiring which should be chosen of a gauge to handle the amp fuse you use.
 
riba2233 said:
You must fuse according to your cables ratings. Fuse is there only to prevent fire, so it pops before cables catch on fire, during short circuit events.

riba2233 said:
Fuse will not protect the controller, transistors and IC's blow in us, while fuses blow in ms-s range... That's not the point of fuse. It's to prevent fire in case of short circuits, to blow before the cables catch on fire.
¿Isn't it more important to base current protection (fuse) on the lighter gauge wiring and circuits of the motor and controller than the much heavier supply wires?

It sounds like you are willing to sacrifice all the expensive eBike components:
motor = $400
controller = $150
battery = $500
to protect ... $10 worth of wire! ... :?:

I recommend ...
Heavier than "necessary" supply wires (to limit IR, line drop and thermal waste) and a marginally higher than actual maximum current use fuse! - (much lower than required for wire)
This method has saved more than just the supply wires ... many times(!) ... in many differing applications!
 
DrkAngel said:
¿Isn't it more important to base current protection (fuse) on the lighter gauge wiring and circuits of the motor and controller than the much heavier supply wires?
No, if you mean the phase wires, because it does not see battery current, it usually sees higher currents than the battery provides to the controller, in spikes as commutation occurs, dependent on settings, inductance, etc.

So a fuse rated for what the motor might need to see in pulses or startup could let the constant currents in the battery allow a fire, depending on how great your motor phase currents are.

If it is a brushed motor, then the currents are different, but still higher than battery currents, in the PWM pulses.


If the motor fails in a way (overheats and two phases shorted together from melted wires in the axle, etc) that draws damaging levels of high current from the controller, then if the fuses are in the phase wires, it might prevent damage to the wires themselves--but it probably won't blow fast enough to save the electronics (FETs, gate drivers), if it's really high enough to instantly damage things anyway. And if the FETs fail, they'll probably blow up and open the circuit anyway.

A battery-wire fuse of any size, small or large, wouldn't stop the motor failure itself--it would just blow once the system drew enough current long enough, and protect the battery (and other) wire insulation from melting and potentially causing a fire if they then allowed the conductors inside to short to each other.


If the controller fails in a way that draws high current from the battery, then a battery-wire fuse sized to save the wires would also blow then, too, and again protect the rest of the system (whatever's not already damaged) from the heat generated from the failure.


FWIW, a motor probably wouldn't be damaged by anything that a fuse could prevent, anyhow. It could *cause* damage that a fuse could prevent, under some circumstances.


You *can* size the fuses to protect the motor wires, or the controller, but the motor itself doesn't need protecting via fuse--generally anything the motor does that would cause a fuse to blow means the motor is *already* damaged.

If the motor is overheating to cause the damage, then there there is something *else* wrong; probably it's overloaded, which you can prevent with limiters in your system elsewhere, throttle, cotnroller, etc., or you could add thermal cutoff or rollback to prevent overheating to the point of damage.

But a fuse won't do anything to stop the overheating, unless you simply size it to blow when the current is high enough to cause such overheating. That also means it'll blow on startup from a stop with "instant start" controllers that don't ramp up current to the motor, because from a stop they'll draw whatever current the controller can put out...whcih means that if that's enough to blow a fuse to prevent overheating, it's also enough to blow a fuse *anytime* the controller draws that much current.


Controllers also won't blow a fuse unless something is wrong (already failed) if everything is setup right and the fuse is already sized high enough to allow normal operation, and the cotnroller is sized right for the motor, and has limitations in place to prevent overcurrents or overheating. Unfortunately, most controllers are *not* well-designed like that, whcih is why so many blow up so easily...they don't have internal thermal rollback or cutoff, they don't have anything beyond basic *battery current* limiting, etc.

So either you have to add that limiting or rollback or cutoff to the controller (or system), such as with thermal sensors and a Cycle Analyst v3 or similar, or you could fuse it so that there is no way you can pull enough battery current long enough to overheat anything...which again, also means that unless it is a soft-start controller it'll also blow the fuse at startup from a stop. And it also means you probably can't climb much of a slope, if doing so draws current high enough to overheat the motor, cuz that would also blow the fuse, if it's sized to prevent overheating.


So...the fuse is there to prevent a number of things...but in our application on ebikes/etc it isn't going to be able to protect the motor or controller from a number of types of damage--unfortunately they are the most common types. It's just going to protect the battery and it's wiring, and hopefully prevent an on-bike fire or other lesser damage should the worst happen and short across the supply wiring to the cotnroller either inside the cotnroller or between it and the battery.
 
Just built a 25.9V 25.92Ah eZip RMB upgrade for a local eZip owner.
Used recycled, some used, Dell Media Bay battery cells of ~10 year age.
Careful evaluation produced a metered capacity of 19Ah+ from 3.71V to 4.16V per cell with all banks equal within <1/100th V at 3.71V

Strange! ... Everyone seems to say that Lipo "ages" badly!

2013 builds with 8 year old NOS cells output near 100% of rated.

LiPo Hybrid Pack

Anyhow, just converted another eZip to 24-36V capability.
So, decided to build a couple 33.3V LiPo RMB eZip packs.
Tested and rated 70+ 6 packs of LiPo cells.

Plan on 9s8p of 2160mAh cells for 33.3V 17.28Ah.
There is extra space ... enough for 3 3s 2000 mAh RC LiPo for 19.2Ah with much improved surge capability.

Turbo Pack - 3s?
Also decided to build a 3s battery pack.
Mufti-purpose!
1. For 12V devices
2. 12V for AC inverter - may integrate small inverter - 150w or dual lighter plugs ... usb too? (120V AC lights, radio, computer, tools etc)
3. For Turbo function - in conjunction with 7s 25.9V pack
The 2 pack - on\off\on switch allows running from primary pack ... the "switching" to "in series" (just attach 3s neg to 7s pos for 10s Turbo.)
 
Back
Top