Homemade Battery Packs

Hi everyone,

It's been a long time since I posted in this discussion.
I just wanted to update you guys on my DIY Li-Ion battery. It's working great!




I have a couple of images from the cell logs I am using:

bat_grap1.jpg

bat_graph2.jpg

This was from a 17km trip. My bike has a 36v 500w bafang geared rear hub. Top speed 30-32kph on the flat.
 
A few months ago I "patched" one of my 18650 Li-ion packs with cells of a different manufacture.
All original cells were of same type, capacity and manufacture.
I added a pair of cells to the weak bank of cells.
Non balanced charging still produced good matched voltages at "full" and "empty".

But ... when I did a balance charge, the "patched" bank showed a noticeably higher voltage during charging.
As soon as charging was stopped the voltages were nearly equal.
The problems were that:
1. Charging stopped when the patched bank reached a false "full".
2. If allowed to fully "balance", the patched bank would be noticeably undercharged.

So ... once again ... "best results are obtained using cells of like manufacture, capacity, condition etc."

On the other hand the non balance charging produced excellent results ... no problems!

My "patch" was a bit of an experiment ... with mixed results. 1 more thing learned ...
I will disassemble the weak bank and repair with matched cells.
 
I wanna know where are you sourcing these cells i maybe intrested and making a tinkering with this even though im on the lipo bandwagon. I just got my icharger so im ready to f some s up.... or ressurect the dead... but seriously these laptop cells dont have enough C rating but the way I hypermile and barely sip 1-2A at 20mph I wouldnt mind a small light pack that i could possibly fit in my triangle and have it look stealty.
 
DrkAngel, your explanation of the patched cells, if I understood it correctly, makes perfect sense. The bad cells are ignored during discharge, because they do not deliver much energy, but the patched cells pick up the slack. However, during charge, the bad cells will quickly collect a surface charge. In reality, this should also be distributed to the patched in cells, but it's conceivable that this would slow down the process by adding another leg. In reality, you're probably best dissecting the bad cells, and not simply patching in some supporting cells.

I have considered this patching method before, though, and never ended up patching them in, so I am curious to hear more if you get more info about how your system is working.

Am i understanding correctly that, rather than dissecting out the bad cells and putting in new cells, you just added some cells in parallel to pick up the slack?
 
icecube57 said:
I wanna know where are you sourcing these cells i maybe interested and making a tinkering with this even though I'm on the lipo bandwagon. I just got my icharger so I'm ready to f some s up.... or resurrect the dead... but seriously these laptop cells don't have enough C rating but the way I hypermile and barely sip 1-2A at 20mph I wouldn't mind a small light pack that i could possibly fit in my triangle and have it look stealthy.
Battery acquistion!
eBay, craigslist - search for lot, or batch, laptop, or notebook, batteries.
("Used" search option cuts down the item count, but you might miss some )
Local recycling centers
Large companies - many, swap all notebook batteries for new, ~semi-annually - ideal - same brand, age etc. & most all still good!

Hybrid pack!
One of my next trials will be to mate a RC Lipo pack to a recycled Li-ion (18650) pack.
The object being a good surge capability married to a cheap large capacity.
Any surge would be mostly covered by the RC portion, with the 18650 portion offering a modest addition, followed by a gradual replenishment of the RC.

Alternately, blocking diodes could "segregate" - limit battery outputs to the controller only.
Surge output would come from the RC and "normal" drain would come from the 18650's, until voltage equalizes.

Unfortunately, I have no motors requiring more than a 1C output, during surge.
Until ... I stick my ETEK 8-16HP into something, 300A @ 50-60V surge capable!
 
DrkAngel said:
Hybrid pack!
One of my next trials will be to mate a RC Lipo pack to a recycled Li-ion (18650) pack.
The object being a good surge capability married to a cheap large capacity.
Any surge would be mostly covered by the RC portion, with the 18650 portion offering a modest addition, followed by a gradual replenishment of the RC.

Alternately, blocking diodes could "segregate" - limit battery outputs to the controller only.
Surge output would come from the RC and "normal" drain would come from the 18650's, until voltage equalizes.

Unfortunately, I have no motors requiring more than a 1C output, during surge.
Until ... I stick my ETEK 8-16HP into something, 300A @ 50-60V surge capable!

Hi DrkAngel!

I was thinking about the same thing but,why not use other 18650 cells more "C" capable (and less capacity) such as those used for power tools (Konion-Makita?). It makes the pairing/pack assembly easier. Those cell should also be safer than rc lipo.

8)
 
Konion Makita are actually LiMn batteries, while LiPo and Laptop batteries about both LiCo. I think there are a few threads where people considered mixing LiMn and Lico, but I forget what the results are.

The thing is, that although LiPo and Laptop are both LiCo, the discharge of LiPo is still somewhat different than LiCo in the same way that A123 has a slightly higher nominal than crappier LiPO4. This slightly higher nominal of LiPo is not necessarily bad at all though, especially in this usage.

Since LiPo and Laptop are the same thing (conclusion from my thread asking if the Panasonic laptop batteries are magic LiCo or just differently optimized), just with differently optimized components, perhaps it would make the most sense to simply use laptop batteries that are the optimal C rating you want, rather than mixing a bit of LiPo in. But this whole thread is about using what you have, so it makes sense to me to try LiPo out.

spuzzete said:
Hi DrkAngel!

I was thinking about the same thing but,why not use other 18650 cells more "C" capable (and less capacity) such as those used for power tools (Konion-Makita?). It makes the pairing/pack assembly easier. Those cell should also be safer than rc lipo.

8)



Also, I'm not entirely sure why you would need diodes for mixing LiPo and LiCo. If I'm not mistaken, the discharge curves are similar enough within their normal operating parameters, such that you're not going to have any cliffs of voltage difference that could allow the batteries to dump a lot of current into one or the other.
 
I thought I would show my finished battery pack.

Test%2520Fit%2520of%2520Batteries.jpg


Out of twenty four laptop batteries I purchased off of e-bay; I ended up using twenty of them to create this pack. You can see my previous posts on reviving what were at one time new batteries that had basically suffocated on the shelf.

My original thought was that I was going to be able to build two separate packs; but in reviving the laptop batteries that had been slowly drained/suffocated by their internal battery management boards I was only able to get back less than 2/3rds of the original capacity, so I wound up making one big pack.

My charger is a Thunder T6 Multi charger that has four output zones that will charge 1S to 6S, however the available charging amperage drops above a 3S setting.

Thunder%2520T6%2520Charger.jpg


Given this I designed the pack to be easily split up into four zones to match the chargers capability of balance charging four zones at once. This was necessary as the charging zones out of the charger share a common ground.

Wired%2520up%2520as%2520a%252048v%2520pack.jpg


As you can see there are three black banna plug shunts that are easily removed to slit the pack up into the four seperate zones. I used color coded banna jacks; green is ground for each zone, yellow is 1S/V1, blue is 2S/V2 and red if 3S/V3 for a 12.6 zone.

Here's the pack being charged after a ride:

Final%2520Charge%2520configuration.jpg


For a battery case I used a $8.89 plastic tool box that I got at the local walmart.

8%2520Dollar%2520Battery%2520Case.jpg


Here it is; attached to the cargo carrier on the bike.

First%2520mount-lasted%25202%2520miles.jpg


After riding for 20 miles I had to remove the battery pack from the toolbox to do some minor repairs. Added silicone separator sheets between the zones after a layer of batteries shifted and had a minor short.

WP_000256.jpg


I think using banna plugs and jacks inside of the toolbox gives me some operational flexblity as I can easily take out a zone to revert to 36v from 48v.

More pictures with captions are available at:

https://picasaweb.google.com/fredsReprap/EBike?authuser=0&feat=directlink
 
fredsparkle said:
I thought I would show my finished battery pack.

Test%2520Fit%2520of%2520Batteries.jpg



For a battery case I used a $8.89 plastic tool box that I got at the local walmart.

8%2520Dollar%2520Battery%2520Case.jpg


Here it is; attached to the cargo carrier on the bike.

First%2520mount-lasted%25202%2520miles.jpg


After riding for 20 miles I had to remove the battery pack from the toolbox to do some minor repairs. Added silicone separator sheets between the zones after a layer of batteries shifted and had a minor short.

WP_000256.jpg


I think using banna plugs and jacks inside of the toolbox gives me some operational flexblity as I can easily take out a zone to revert to 36v from 48v.

More pictures with captions are available at:

https://picasaweb.google.com/fredsReprap/EBike?authuser=0&feat=directlink

Hi!

Nice job on the battery pack!

However I would suggest to move the battery into the triangle of the bike frame like I did if you have enough room. It's not hard, and the results are amazing, your bike will feel much more safe and balanced to ride. I was having problems with the toolbox in the rear rack, especially on bumpy roads (even small bumps).
You can see what I did on my bike here http://endless-sphere.com/forums/viewtopic.php?f=3&t=35289&start=30#p589979 or on my website in the signature.

8)
 
spuzzete said:
Hi!

Nice job on the battery pack!

However I would suggest to move the battery into the triangle of the bike frame like I did if you have enough room. It's not hard, and the results are amazing, your bike will feel much more safe and balanced to ride. I was having problems with the toolbox in the rear rack, especially on bumpy roads (even small bumps).
You can see what I did on my bike here http://endless-sphere.com/forums/viewtopic.php?f=3&t=35289&start=30#p589979 or on my website in the signature.

8)

Thank you, I looked at your post and it looks great. I will shortly have three battery packs, so being able to swap is good. Also the frame space on my bike is not a neat triange like it is on yours, so really don't have the room.

I am also waiting to see what final capacity my pack levels out at; as it still seems to be climbing as it gets more use.
 
DrkAngel said:
A few months ago I "patched" one of my 18650 Li-ion packs with cells of a different manufacture.
All original cells were of same type, capacity and manufacture.
I added a pair of cells to the weak bank of cells.
Non balanced charging still produced good matched voltages at "full" and "empty".

But ... when I did a balance charge, the "patched" bank showed a noticeably higher voltage during charging.
As soon as charging was stopped the voltages were nearly equal.
The problems were that:
1. Charging stopped when the patched bank reached a false "full".
2. If allowed to fully "balance", the patched bank would be noticeably undercharged.

So ... once again ... "best results are obtained using cells of like manufacture, capacity, condition etc."

On the other hand the non balance charging produced excellent results ... no problems!

My "patch" was a bit of an experiment ... with mixed results. 1 more thing learned ...
I will disassemble the weak bank and repair with matched cells.
Update !
Opened up my pack, for proper repair and removed my "patch job" (additional cells tacked onto a weak bank).
Got ready to separate the bad bank for testing and matched cell replacement.
Noticed a cracked solder connection, re-soldered it.
Did a balance charge, then a full cycle.
Result ... back in the pink! Good as my first tests.
Matched within 1/100th of 4.18V (29.2V) & within couple hundredths at 3.7V (26V).
 
I did mention previously ... somewhere, that my iMax B8 was unreliable when charging a single cell, or bank.
It allows too high a voltage!

Well, I should have heeded my own warning.
After rebuilding a pack, I decided to "tune" 1 bank and forgot to keep an eye on it.
Voltage measured 4.36V on the single bank!, and it was still charging!
One pair of cells was hot (120F w/optical IR meter) another warm (+100F).
I removed charge and cut hot cell from "bank", and separated other cells.
Next morning, cell voltages ranged between 4.12 and 4.16V, down from their typical 4.20V post charge voltage.

Yup! Cooked um!

Important!
Monitor all functions of a new charger! (Until confirmed accurate)
 
Here's mine. Still work in progress, 14s12p, at 12p most of them gives around 20ah when discharged to 3.0v. Batteries are from 4 years old Toshiba laptops. Those are panasonic cells, I also have a bunch of sanyo.

dQhxj.jpg
 
I would highly recommend that you not discharge below a resting voltage of 3.6V!
Below that voltage there is minimal mAh and every deep discharge is damaging, to capacity, and cycle life.
 
Hi guys,

I made a video with my method for testing the batteries.

[youtube]jHV8n8CVEZ0[/youtube]


Let me know what you think.

8)


P.S. Thanks DrkAngel for your help.
 
I picked up a few Genuine MeanWell S-150-24 switching power supplies.
Decided to play and came up with a 29.4V 5A battery charger for my 7S 25.9V Li-ion packs.
Charger is set to 29.1V = 4.15V per cell-bank.
(Externally, fully adjustable as 4.10V - 4.20V + or -)

MeanWell (Mean Well) S-150-24

29.4V 5A.JPG
"Modded" up from 24V and down from 6.5A.

It was necessary to downgrade and limit amperage, to prevent overheating and self destruction.

Still working on a "full" indicator.
Ampmeter would work ... looking for an affordable solution.

See - MeanWell Mods - S-150-24
 
I like how you made use of the LED hole :D

Those little extension coverings are nice, nothing I've seen before on the clone meanwells. An extra touch that makes things cleaner.
 
I think this is An appropriate place to advertise since there are a number of 18650 builds, but feel free to suggest I remove this post. I have 200-300 new 18650 cells in boxes of 30. They are sanyos. http://endless-sphere.com/forums/viewtopic.php?f=31&t=42800
 
Kin said:
I like how you made use of the LED hole :D

Those little extension coverings are nice, nothing I've seen before on the clone meanwells. An extra touch that makes things cleaner.
These were all NEW! Genuine! MeanWells!
Yeah, I know ... An unbelievable bargain!
Less than $3.50 each ... delivered!
100+ sold in a few hours.

All Gone!
Don't ask!


Power cord socket has an externally accessible fuse.

I used a 2 wire voltage gauge connected to the power rails.
Plug into battery - gives pack voltage.
Then turn power switch on to charge.

Much less of a connection spark if charger is powered on 1st ... tho.
 
Hi all,

First, thanks for drkangel for running this show and everyone else who contributed with some really good information, I too now r tearing down lappy packs and putting together something..
well, configuration will depend on how much good cells i'm going to get :)

Anyway, i'm posting because of a question that i didnt see coming up in the thread, I mostly have lenovo packs and some of them have confusing markings.

file.php


5.2 would make them cells to be 2600mah, but 4.53 comes up with some funky 2265mah .. ?

I'm confuse! Not all of them are marked in this confusing way now i wonder, do i have a lot of 2200 in my hand or 2600 ..

SE US18650GR is written on them, thats the only part that gives results in google, suggesting sony 2200? (second line is "T 8A1230I24T")

thanks in advance for your input, also if anyone knows, are those cells good? in the world of 18650 cells kindof.
 

Attachments

  • Photo 10.09.12 20 41.45x.jpg
    Photo 10.09.12 20 41.45x.jpg
    144.5 KB · Views: 3,336
I guess it's better to ignore the confusing stuff on the battery if there's stuff written on cells and believe information that google provides using cell markings?
Seeing many different lenovo packs with two different capacity on them, one nominal one .. PSE? (circular logo) PSE being lower.
 
Back
Top