Homemade Battery Packs

I'm looking for good, reliable info, so could we please stop this unnecessary arguing? Thx

As a beginner, I need some input for testing. I began with 250 cells for my electric bike to test and don't want to spend night and day picking at them with my multimeter and a light bulb. I'm halfway through opening the packs. So far, I'm getting mostly seemingly good cells at a safe voltage.

Because they're used, I have to cut and pull the tabs to test each individual cell. I've had about 3 dead cells total, each in parallel with good cells.

Right now, I have between 30 and 50 cells in parallel (cardboard box + aluminum foil) bulk charging to 4.2V on an imax b6. I think I need to let it sit for atl 2 weeks to measure self discharge. I hope to finish testing by mid-may. If I decide not to test capacity, I'll sell the good ones on ebay for a profit and buy *new* laptop batteries that can be wired with the tabs on.

Since I'll be charging on a balance charger, will I need a bms for balancing and limiting lower voltage during discharge, or am I fine with a voltmeter / battery indicator?

What percentage of good cells should I expect to get out of the self-discharge test?

After putting them in parallel, should I discharge with the light bulb or my imax b6?
 
DrkAngel said:
Several methods and most concise info see - Bulk Capacity Testing

Also check Index under capacity Testing

is this time consuming procedure really necessary if i use factory new cells?
i solderen 198 new cells to a 22s9p pack (no spot welding) and i have <2mV cell drift over the whole pack after about 30-40full cycles and many small recharge cycles^^
 
madin88 said:
DrkAngel said:
Several methods and most concise info see - Bulk Capacity Testing

Also check Index under capacity Testing

is this time consuming procedure really necessary if i use factory new cells?
i soldered 198 new cells to a 22s9p pack (no spot welding) and i have <2mV cell drift over the whole pack after about 30-40full cycles and many small recharge cycles^^
Procedures were developed to insure cells of stable condition, (no self-discharge), and to rate capacity, for building banks of equal capacity.
Essential for used-recycled cells.
Recommended for any cells from manufacturer with less than impeccable reputation.

Using these procedures, testing new cheap Lipo on a cell basis would definitely help build safer more durable packs.

All cells, even new, should be checked for self discharge!
Simply charge all cells to your determined full charge, or slightly higher, then monitor for several days.
Some, slight discharge, is normal.
You must eliminate all, with any noticeably greater self discharge.
They will cause your pack to continually unbalance itself!
Worse?
Self discharging cells are defective with some type-degree of internal short, which is libel to get progressively worse-more dangerous.
You might be able use these cells for a while if you use a quality BMS-balance charger constantly.

But then ... you are using a BMS as a band-aid for a defective pack!

If your pack begins suffering from one bank discharging deeper than the rest, use this method to pinpoint the "bad" cell, for removal and replacement. See - Finding weak cells
 
teslanv said:
2007blueprius said:
once you find something that works guess you stick with it, kindof why I wanted to start off with something common so at a later date it is compatible with whatever else I may acuire

I really like these. http://www.hobbyking.com/hobbyking/...0pcs_set_US_Warehouse_.html?strSearch=4mm hxt

Way easier to solder than a Deans T connector, and plug right into LiPo packs.


that is kindof the info I was looking for, the HK lipos come with these connectors, how about the test leads are those generally jst's?

like this ghismo for example http://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=10328

looks like a jst connector on the side, any experience with it or some of the likes? I saw several kinds and rather affordable,

I managed to cook my hacked pc power supply and now that I got a new one I want to build it proper, I like those pc supplyes as I can charge large ammounts in parallel at once also for those using RC chargers it has a 12 v outpout, I don't have one of those, but I am kind of on the fence as to how to go about it, my understanding is that a 12s works well with a 48v controller , better than a 16s lifepo, something to do with the low voltage cutoff of the controlers beeing more apropriate for the 44.4v 12s, so the ideea was to use one of those battery medics to balance them from time to time, a 6 s is 22.2v?

apreciate all the input
 
2007blueprius said:
that is kindof the info I was looking for, the HK lipos come with these connectors, how about the test leads are those generally jst's?

like this ghismo for example http://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=10328

looks like a jst connector on the side, any experience with it or some of the likes? I saw several kinds and rather affordable,

I managed to cook my hacked pc power supply and now that I got a new one I want to build it proper, I like those pc supplyes as I can charge large ammounts in parallel at once also for those using RC chargers it has a 12 v outpout, I don't have one of those, but I am kind of on the fence as to how to go about it, my understanding is that a 12s works well with a 48v controller , better than a 16s lifepo, something to do with the low voltage cutoff of the controlers beeing more apropriate for the 44.4v 12s, so the ideea was to use one of those battery medics to balance them from time to time, a 6 s is 22.2v?

apreciate all the input

Yes, I have a battery medic. they work pretty well, although at times I question the accuracy. Mine always seems to be ~10-20mV off from my iCharger or Fluke Multimeter. - but close enough to keep me honest. And yes, the LiPo packs's balance leads plug right into the battery medic. I even wired up the same JST-XH balance leads to my homemade 18650 packs, so I can balance those with a battery medic as well.

12S LiPo or most modern 18650 cells have a nominal voltage of 44.4V (3.7V/cell) and a fully charged voltage of 50.4V or 4.2v/cell.
 
Built a replacement 25.9V 25.92Ah Recycled Lipo pack for my eZip.
7s12p 2160mAh cells, paired, in 3s2p 6 packs.
All new packs-cells, with identical date of manufacturer.
(081209 2008 December 09 ... Yes! ... 5 years old+.)
Charged in 4 bay bulk chargers last summer ... all cells were still 4.09 or 4.08V.
Assembled and ran for 20 miles at between 18 and 22mph.

Resultant cell voltages were either 3.82V or 3.81V ... except the 12cell side cells at 3.80V,
Close inspection found a poor solder joint
Fixed ... and charging now.
Will balance at full charge -4.17V per cell as maximum, typical "full" charge will be ~4.05V ... except for trips.
I anticipate many years and many many thousands of miles.

These 6 packs are ideal for 22.2V, 33.3V, 44.4V builds.
7s is awkward ... but does fit efficiently in the eZip RMB packs.

After initial monitored cycles, I intend on bulk charging only ...
No BMS! ... I view them as band-aids for defective packs.
(Of course I also prefer the tech friendly Windows XP ... so ...)
file.php

Additional points:
Removed built in pack fuse, to simplify wiring.
Added in line fuse (40A)
Replaced all internal ≥ 12ga
Except charging wires upgraded with 16ga
Layed sheets of closed cell foam below and above, to stabilize cells in pack.
"Bricks" of cells are supported by "sheets" of Formica.

Old one passed its 3rd year and 7000 miles ... and was built with the rejects, the recycled packs that would not charge properly!
Still 10-12Ah usable ...
Will keep until I need the shell for next build.
 
DrkAngel said:
Built a replacement 25.9V 25.92Ah Recycled Lipo pack for my eZip.
7s12p 2160mAh cells, paired, in 3s2p 6 packs.
All new packs-cells, with identical date of manufacturer.
(081209 2008 December 09 ... Yes! ... 5 years old+.)
Charged in 4 bay bulk chargers last summer ... all cells were still 4.09 or 4.08V.
Assembled and ran for 20 miles at between 18 and 22mph.

Resultant cell voltages were either 3.82V or 3.81V ... except the 12cell side cells at 3.80V,
Close inspection found a poor solder joint
Fixed ... and charging now.
Will balance at full charge -4.17V per cell as maximum, typical "full" charge will be ~4.05V ... except for trips.
I anticipate many years and many many thousands of miles.
Charged cells from 3.82V to 4.16V,
requiring 18Ah.
Precisely balanced all banks at 4.18V.
I tried to run a discharge test at a more moderate speed, ~16mph average, but traffic and adrenaline kept pushing me past 22mph.

Tomorrow is predicted to be sunny, in the mid 50's, so hopefully, I will have opportunity for a monitored deep discharge at a more reasonable speed.
This will provide a range estimate and an accurate capacity rating for pack and a confirmation of all banks being of equal capacity.

Previously performed capacity mapping confirms these cells of a >3.75V to 4.17V optimal usage range. (2004 and 2006 dated cells)
file.php

Being 2 - 4 years newer than previous test samples, I will perform a repeat test of this dated pack.
 
DrkAngel said:
Previously performed capacity mapping confirms these cells of a >3.75V to 4.17V optimal usage range. (2004 and 2006 dated cells)
file.php

Being 2 - 4 years newer than previous test samples, I will perform a repeat test of this dated pack.
I have a quantity of "new" REV:4 - 2008
and
a larger quantity of "new" REV:5 - 2010
so I will run 2 tests concurrently, I do have a matched pair of V - A - w - mAh meters.

Capacity mapping procedure

For these cells, I will test sample in the oem 3s2p configuration.
From a full 12.60V, I will 1st, perform a metered full discharge to a static 10.80V.
(4.20V - 3.60V per cell)

From 10.80V, I will set MeanWell PS at 10.86V and charge till mA "charge" drops to negligible (~.01A?)
mAh required will be noted,
then charge voltage increased in .06 increments,
this equals .02V (2/100th V) increments per cell, matching previous maps.
 
Regarding salvaging laptop 18650 cells.
How about determining the condition by when you do the first charge they do not accept a charge currant of 1C?
All of the cells I have charged so far: The ones that did not accept the 2A charge on 2 parallel cells (thats roughly less than 0.5C) show very poor capacity.

Sounds like a good way to start of the salvaging process?
 
Skalabala said:
Regarding salvaging laptop 18650 cells.
How about determining the condition by when you do the first charge they do not accept a charge currant of 1C?
All of the cells I have charged so far: The ones that did not accept the 2A charge on 2 parallel cells (thats roughly less than 0.5C) show very poor capacity.

Sounds like a good way to start of the salvaging process?
You can try it ...
Defiantly not something I would rely on as a definitive test.

But could yield some interesting results!
 
"Mapped" building simple ~cheap 500w & 1000w chargers for 63 - 84V (15s to 20s) Li-ion.

Cheap-easy (470-500w) ≤84V charger
Requires Genuine MeanWell S-150-24 with current mod (R33 mod) & 1 generic 350-48 power supplies

Easy (1000w) ≤84V charger
Requires Genuine MeanWell S-350-24 with current mod (R33 mod) & 2 generic 350-48 power supplies
(Oops! :cry: will try to finish S-350-24 current mod values.)

Both are adaptable for down to 37.6V (9s) chargers.
Requires simple voltage mods on all units.
Will reduce charging watts tho ...

I have only mapped current mods for the MeanWell S-150 series and the S-350 series. ES Wiki - MeanWell Mods
The NES series and the SP series etc. have no documented current adjustment ... unless someone else knows of such?
 
This battery recovery sucks without "good" equipment :(
I am using a 6s RC charger like a "B6" And chinese single cell 18650 charger.

Does anyone know why a cell will not accept a high currant charge? 0.5-1C+- End result so far is low capacity.
Cells that do not want to charge like they should goes to my chinese charger and gets charged to 4.2V

A lot of the cells have great potential :) I have very bad success with Sanyo :( I think its because the Sanyo cells did not die in the laptop and was cycled a LOT.

And my success with cells at 0V so far is 100% and have great capacity.
 
Skalabala said:
This battery recovery sucks without "good" equipment :(
I am using a 6s RC charger like a "B6" And chinese single cell 18650 charger.

Does anyone know why a cell will not accept a high currant charge? 0.5-1C+- End result so far is low capacity.
Cells that do not want to charge like they should goes to my chinese charger and gets charged to 4.2V

A lot of the cells have great potential :) I have very bad success with Sanyo :( I think its because the Sanyo cells did not die in the laptop and was cycled a LOT.

And my success with cells at 0V so far is 100% and have great capacity.
Laptop cells are not designed for, and recommended against, charging at greater than .5C.

I have also noted that C rate should be rated by actual capacity rather than original capacity!
As it aged-deteriorated, my 25.9V 25.92 recycled laptop pack began displaying increased heat during typical charges and discharges.
10A charging displayed noticeable warmth during charging at nearly exactly the point when usable capacity dropped to ~18Ah. I dropped half my modular charger ... going to a 5A charger, which resulted in a nice cool pack during charging.
This seemed to mirror an increase in IR, voltage sag under full throttle increased relative to percentage of actual capacity.

Most Sanyo cells are 3.6V rated and thus seem designed to be best charged to 4.10V?
I did begin checking packs before disassembly, many are rated as 10.8V - 14.4 (3.6V) rather than 11.1V - 14.8V (3.7V).

iMax and other chargers have 2 charge settings, 4.20V for Lipo and 4.10V for Li-ion.
These type chargers are designed to charge to 4.20V, not above, in fact, due to line drop etc. actual finished charge usually results in ~4.18V rather than a "full" 4.20V.
Also to note ... the fast charge setting estimates full with an extra safety margin and, likely, will charge faster but to an even lower resulting voltage.
Also, cheap multi-meters often vary from actual voltage by several 1/100th V. (I lucked into possession of a Fluke 115 multi-meter for $5 yesterday ... Retails ~$159!)

If a cell starts at 0V, and recharges in an acceptable manner there is still a good chance that it suffers from some degree of self discharge.
If indeed self discharging, and you decide to use, it should be segregated into a separate pack with cells of same condition, and continually monitored.
 
Thanks DrkAngel.
So if the cell does not accept the charge currant according to original capacity using 0.5C then it should/might have a low capacity?
 
Skalabala said:
Thanks DrkAngel.
So if the cell does not accept the charge currant according to original capacity using 0.5C then it should/might have a low capacity?
There are too many factors to consider!

EG
2600mAh cell @ 3.50V should accept 1.3A charge.
2600mAh cell @ 4.10V should not accept 1.3A charge.

2200mAh cell might accept 1.1A charge till near full ... due to large percentage of anode-cathode, however ...
high capacity cells 2600-2900mAh + sacrifice C rate capability to increase electrolyte-capacity and might not accept .5C charge over much of its capacity range.

C rate might work as a rule of thumb for capacity, but only for comparing cells of identical manufacture.
Same brand
Same batch
Same beginning state of charge
Same age
Same ... etc.
 
And what if the RC charger will only charge it to 4.08/4.12+- ?
My batch of 0V cells for today had 3 out of 12 that took good capacity :mrgreen: Rest will not take charge at all :p
How do you capacity test individual cells? It seems to me that one can do a voltage sag test: All the cells I do a voltage sag test on has a low capacity when it shows a big amount of sag, and when it does not sag a lot it has a good capacity. Obviously C rating will also play a role on the actual capacity but might work.

All the cells taking only low charge currant and they also sag badly has done poor with the RC charger capacity test.
 
I'm having trouble touching all the contacts in my bulk charging foil box. Should I solder all the connections, or is there a more effective method that I haven't seen yet
 
lxgoldsmith said:
I'm having trouble touching all the contacts in my bulk charging foil box. Should I solder all the connections, or is there a more effective method that I haven't seen yet
Finished pack will have all cells soldered. ... ?
I recommend all cells be pre-tinned ... abraided, fluxed and small solder bead applied, before wiring together.
Pre-tinning cells now, should make foil box contacts better.


Foil box ... ?
Box, with foil on bottom, attach neg 4.20V.
Stand cells with neg side down ... Solder bead on bottom making reasonable contact.
Place foil on top of cells.
Place layer of closed cell foam on top of foil.
Place solid flat board(?) on top, add weight to insure firm contact.
Hook Pos 4.20V to top foil. ... ?

Personally, I solder all cells in parallel then charge.
All cells should be of similar voltage.
Laptop cells between 3.00V and 3.50V have such little energy that they can be ganged together with minimal transfer current between cells.

Pics - Bulk Capacity Testing
 
Yes. When I open up the box and remove the foil after some time charging, a few cells have not changed in voltage. I'll try to apply a second foil layer and more even pressure before I move to soldering.
 
DrkAngel said:
Built a replacement 25.9V 25.92Ah Recycled Lipo pack for my eZip.
7s12p 2160mAh cells, paired, in 3s2p 6 packs.
All new packs-cells, with identical date of manufacturer.
(081209 2008 December 09 ... Yes! ... 5 years old+.)

file.php
Side bank demonstrated excessive discharge.
Disassembly and careful inspection revealed 1 torn lipo ribbon.
So bank was only supplying 11/12th the capacity.

Replaced damaged pair and ready for retesting.
All banks balanced within 1/1000thV of 4.159V ... got me a Fluke multi-meter w/1000th V readings.

Running tests at 4.15V+ before switching to 4.05V as typical charge level.

2008 Trailz w/noisy MY1018z motor and 16T mod (5900 miles)

25.9V 25.92Ah pack motivates at ~22mph.
Same battery pushes my 2009 Trailz LS Comfort Cruiser w/XYD-16 motor at ~20mph.
Now I like the "legal" monicker, and 22mph (Spring fever - can't slow down!) sucks the power from my battery at an accelerated rate so ...

I believe I will attempt to build another 22.2V pack.
Previous 22.2V pack consisted of 40Ah+ of old re-re-recycled Sanyo cells and I rated as ... lack-luster - too wimpy! in performance.

Proposed build will consist of the same type Lipo cells used in 25.9v 22mph pack.
Instead of using the 12 side cells as an additional 7th bank they will add 2 cells to each of the 6s "bricks".
Same total capacity
7s12p = 25.9V 25.92Ah = .671 kWh
will become
6s14p = 22.2V 30.24Ah = .671 kWh
22.2V pack should be an enforced economical and "legal" alternative for my 2008 Trailz w/16T mod and my 2013 Trailz LS w/13T mod.

22.2V Lipo should outperform, or at least compare favorably with, 24V SLA.
SLA 24V controller w/~21V LVC should be perfect for the 22.2V Lipo.
Balance charge with cheap iMax B6 or equivalent ~$20
6A CC/CV charging with MeanWell S-150-24 ~$20 + shipping
And should provide 600% the range of the oem SLAbatts! ... at 1/2 the weight!!!

Oh ... and ...
5x the range of the $400 Currie Lithium alternative (6.14Ah)
3x the range of the $600 Currie Lithium alternative (9.6Ah)
 
15.9V 25.92 eZip Lipo build finished!

After cell replacement and balancing all banks within 1/1000V of 4.159V,
running bike at an average 18+mph for 24 miles,
through 40-50º F streets and
nasty headwinds ...
All banks are between 3.804V and 3.808V.
I am well satisfied!

Previous capacity mapping indicated 3.75V as effectively empty.

I intend on maintaining voltages between 3.80V and 4.05V as optimal for best life and usability.
(Charging to ~4.15V for extended runs?)

Am presently doing a metered recharge to determine actual capacity at 3.80V - ~4.06V (iMax Lion charge).
Then I will further charge to iMax Lipo ~4.16V setting to determine additional capacity.

Although I will schedule bank voltage monitoring, I foresee no reason for balance or equalization charging!
All cells being new with identical date of manufacture I anticipate an extended lifespan with an equistatic and delayed deterioration.
Last pack, with used cells, and higher charge - lower discharge voltages performed admirably for 3 years 7000 miles ...
And still has better capacity-range than the $600 oem lithium replacement!
 
DrkAngel said:
25.9V 25.92 eZip Lipo build finished!
Am presently doing a metered recharge to determine actual capacity at 3.80V - ~4.06V (iMax Lion charge).
Then I will further charge to iMax Lipo ~4.16V setting to determine additional capacity.
iMax B8 "Lion charge" from 26.60V (3.80V) to 28.49V (4.07V) required 13.54Ah
Resetting iMax B8 to "Lipo" and charging to 29.19V (4.17V) required + 4.12Ah

To determine optimal voltage range I am capacity mapping samples of these cells
3s2p 2008 Rev 4 Lipo ... for present build and
3s2p 2010 Rev 5 Lipo ... for future builds

Step 1
Charged all cells precisely to 4.18V
Then metered discharged and equalized to 3.52V
Rev 4 (2008) metered 4128mAh
Rev 5 (2010) metered 4100mAh + (accidentally discharged below 3.52V then recharged - so approximated)
4128 / 4320 = 95.5% of oem rated
(Might supply 100% of rated with 4.20V - 3.50V discharge!)
2008 = ~6 years old cells ... Makes me question "age deterioration"! ... ???

Both 6 packs discharged evenly - all 3 banks within 1/100th V at 3.52V!

Step 2
Rigged 2 identical mAh meters to MeanWell S-150-12 and
will charge 2 - 3s2p in 6/100th V increments (2/100th V per cell)
(35 - mAh readings per 6 pack)

Step 3
Will graph data and present

If weather is poor ... might run tests and have data ready by tomorrow ... ?

Oh! "Accidentally" got a Fluke multi-meter ... of impeccable reputation ... so I used it to calibrate my other multimeters.
2 were within 100th V - 3 were not!
 
DrkAngel said:
DrkAngel said:
25.9V 25.92 eZip Lipo build finished!
Am presently doing a metered recharge to determine actual capacity at 3.80V - ~4.06V (iMax Lion charge).
Then I will further charge to iMax Lipo ~4.16V setting to determine additional capacity.
iMax B8 "Lion charge" from 26.60V (3.80V) to 28.49V (4.07V) required 13.54Ah
Resetting iMax B8 to "Lipo" and charging to 29.19V (4.17V) required + 4.12Ah

To determine optimal voltage range I am capacity mapping samples of these cells
3s2p 2008 Rev 4 Lipo ... for present build and
3s2p 2010 Rev 5 Lipo ... for future builds

Step 1
Charged all cells precisely to 4.18V
Then metered discharged and equalized to 3.52V
Rev 4 (2008) metered 4128mAh
Rev 5 (2010) metered 4100mAh + (accidentally discharged below 3.52V then recharged - so approximated)
4128 / 4320 = 95.5% of oem rated
(Might supply 100% of rated with 4.20V - 3.50V discharge!)
2008 = ~6 years old cells ... Makes me question "age deterioration"! ... ???

Both 6 packs discharged evenly - all 3 banks within 1/100th V at 3.52V!

Step 2
Rigged 2 identical mAh meters to MeanWell S-150-12 and
will charge 2 - 3s2p in 6/100th V increments (2/100th V per cell)
(35 - mAh readings per 6 pack)

Step 3
Will graph data and present ...
Recycled Laptop Lipo
Capacity mapped, 2008 and 2010, the 2160mAh "Media bay" secondary battery cells.

Lipo 4320 (Dell) 2008-2010.jpg

Total mAh charge, for 4320mAh rated cells, 3.50V to 4.30V was 4450mAh.
3.7V to 4.20V looks to be almost precisely 100% of oem rated capacity!
(4 year old cells)

Note - 2008 (Red) series mAh meter exhibited a slight "idle charge" ... so graphed line is slightly higher than actual.

See also - Capacity Mapping (Optimal Charge-Discharge Voltages)
 
Back
Top