Homemade Battery Packs

The numbers reported by your imax charger are correct.

You are mixing up current (A) and discharge rate (C). For your 1800mAh cell, a 0.5C discharge rate is 0.9A (1.8/2). At 0.5A, your discharge rate is only 0.27C.
 
xmasta said:
so It has ran for 76mins, at 0.5A now down to 3.6V (gradually from 4.2) and it only got out 637mAh of energy.. at 0.5A? wtf
Even if it would have been 3.6v all the time then I believe that in 1h it should have outputted 1800 mAh ? (3.6x0.5x1) or what am i getting wrong here..

atm it has ran 127mins, at 0.5A, voltage is down to 3.56v and output is 1067mAh ..

Nothing wrong with these numbers. Ah is capacity unit and equals the current x time. So 0,5A during 76mins (=1,267 hours)=0,5*1,267=0,63Ah. The same goes for 0,5A during 127 mins(= 2,116 hours) = 0,5*2.116hours = 1,06Ah
 
I have been testing a prototype 22.2v 41.6Ah pack, on 24V ezip eBikes.
After deducting road load and wind resistance, there is a certain amount of available torque for acceleration.

Hmmm ... As shown in the following graph, at 16 mph, after you deduct road and wind resistance, my 25.9V pack supplies more than 2x the available torque compared to my 22.2V pack.
("486w" should be replaced with "~418w-450w-486w")

No wonder the 22.2V pack seemed anemic!
Of course, at lower speeds, this difference diminishes, but it is extremely important to consider torque differences at your project speed goals!
Who would have thought that a 12% decrease in voltage would make such a drastic difference at performance.
 
etriker said:
That is interesting. How about distance ? Does the 22.2v pack last longer or make for a longer ride ?
IF! ... and only if ... you were able to restrain throttle position-usage, distance would be the same.
The 22.2V pack vs 25.9V, with equal watt hours of storage capacity, with superhuman throttle and speed restraint, could be of equal speed and range ... but I don't foresee anyone ... ever ... sacrificing the available acceleration and speed for enforced better range.
 
SamTexas said:
The numbers reported by your imax charger are correct.

You are mixing up current (A) and discharge rate (C). For your 1800mAh cell, a 0.5C discharge rate is 0.9A (1.8/2). At 0.5A, your discharge rate is only 0.27C.
incorrect, see the post again (screenshot)

fivari said:
xmasta said:
so It has ran for 76mins, at 0.5A now down to 3.6V (gradually from 4.2) and it only got out 637mAh of energy.. at 0.5A? wtf
Even if it would have been 3.6v all the time then I believe that in 1h it should have outputted 1800 mAh ? (3.6x0.5x1) or what am i getting wrong here..

atm it has ran 127mins, at 0.5A, voltage is down to 3.56v and output is 1067mAh ..

Nothing wrong with these numbers. Ah is capacity unit and equals the current x time. So 0,5A during 76mins (=1,267 hours)=0,5*1,267=0,63Ah. The same goes for 0,5A during 127 mins(= 2,116 hours) = 0,5*2.116hours = 1,06Ah

Thanks, I was calculating wattage instead.. now i got that, duh.. also, damn poor cell!
 
Any chance that there will be a 24V EZIP prototype available for sale this Spring? I'm looking for range and reliability more than raw power.
 
xmasta said:
SamTexas said:
The numbers reported by your imax charger are correct.

You are mixing up current (A) and discharge rate (C). For your 1800mAh cell, a 0.5C discharge rate is 0.9A (1.8/2). At 0.5A, your discharge rate is only 0.27C.
incorrect, see the post again (screenshot)
Really? Which part was incorrect?
 
SamTexas said:
xmasta said:
SamTexas said:
The numbers reported by your imax charger are correct.

You are mixing up current (A) and discharge rate (C). For your 1800mAh cell, a 0.5C discharge rate is 0.9A (1.8/2). At 0.5A, your discharge rate is only 0.27C.
incorrect, see the post again (screenshot)
Really? Which part was incorrect?

All of it, I saw the (A) very clearly and the photo shows it too, i mixed up how to calculate capacity as you can see from previous post. My cells are all above 2000mah and 0.5C would be 1A+. I get that you thought that i thought i'm calculating the rate at wich it charges but no, the C was not in my mind, just messed up.
 
39 mile report - 22.2V 41.6Ah Rebuild pack[/b]

25.2V Start
23 miles @ ~17mph = 23.7V
+15.67 Miles @ ~16mph = 22.9V


... - 23m ..... 39m
1a = 3.95V ... 3.81V ...
2a = 3.96V ... 3.81V ...
3a = 3.95V ... 3.81V ...
4a = 3.95V ... 3.81V ...
5a = 3.95V ... 3.81V ...
6a = 3.96V ... 3.81V ...
+
+
6b = 3.96V ... 3.81V ...
5b = 3.96V ... 3.81V ...
4b = 3.95V ... 3.81V ...
3b = 3.97V ... 3.81V ...
2b = 3.95V ... 3.81V ...
1b = 3.95V ... 3.81V ...
-
Bank voltage is remaining very close!
Voltages should continue to diverge as pack nears "empty".


Present discharge points between lines.
Present discharge about 60-65% ... estimated by area of graph.

file.php


Amazingly, voltage variance between banks has diminished!
All banks are equalized within 100th V
The real test will be after the 3.7V point is passed.
Happy with the cells!
Unhappy with the lackluster performance of 22.2V. See - 22.2V Battery Woes
Of course the last 16 miles were in sub-freezing weather mostly with a cold battery pack.
If cells continue their stellar performance ... they might become my 20.8Ah 44.4 MAGNUM!

Graph determined using - http://endless-sphere.com/forums/viewtopic.php?f=14&t=33271
 

Attachments

  • 39 Mile Variance.JPG
    39 Mile Variance.JPG
    15.9 KB · Views: 5,658
Sadly, since I started promoting recycled laptop builds, price has gone up and availability has diminished.

Everyone should refrain from all laptop pack purchases until the price drops and I can stock up again!

Just kidding, just got a bonanza of goodness.
I was negotiating with a recycler with Dell connections.
Wanted 100 packs of 11.1V 4320mAh Lipo, either the type with the SONY cells at one price, or the unbranded model at a lesser price.
They agreed to my prices but combined both options and offered in an open eBay auction ... I had to jump on it ...
So just got 200+ packs, 600 genuine SONY LiPo (2160mAh) cells and another 600+ reputable brand - Dell sold them - (2160mAh). 30+ packs guaranteed new in oem packaging.
25.9V 25.92Ah eZip pack rebuilds require 15 packs each. So 2 builds with new LiPo cells.
Giving up on 22.2V packs, much easier to build but pitifully wimpy.

LiPos will be charged in 3 - 4 bay charging stations = 12 at a time. Then will be capacity tested in oem 6 packs with Ah meter, using a 12V DC to 110V AC inverter.
I have components for 2 identical discharge stations.
Should be able to test 16 6 packs per work day (in the background), so might take 2.5 weeks to test all packs.
Now have full 8kWh of sony cells for eMotorcycle.

Most importantly I got a 30 day warranty for any defective packs!
And, last batch, any defective packs were replaced with guaranteed new in OEM packaging!

I might have totally cleaned them out, but will test, report and probably recommend seller ... dependent on quality-condition of received packs and further availability.

No! I will not recommend, or refer, till I receive mine!
 
Dell recycled Lipo.
7s - 25.9V, 12p - 25.92Ah, homemade recycled Lipo pack.
Cells are 12 deep, 6 - 6packs, 12 cells on the side adding the #7s.

file.php


15 Packs required.
12 packs - 2 banks of 3s12p
3 packs - separated, center cells(without tabs) removed, remaining 12 cells as 7th bank.

Big pack 37V 25.92Ah

file.php


A delight for many ... soldering is super simple!
2 solder points per 6 cells!
Until you add balance wires.

Notice - After building these packs, I, at first, relied on charging to 4.20V per cell.
This was way outside the "mapped usable".
"Optimal" appears to be at 4.05 or even lower.
Considering that the construction was from the "defective" packs, was over-charged, but is still going after thousands of miles, I am very optimistic of properly matched and gently charged packs performing excellently!

I highly recommend not exceeding 1C discharge, .3C to .5C or lower preferred.
So build big capacity packs!
Per Ah these cells weigh less than 1/2 of LiFePO4, less than 1/5 of SLA.
 
Here's my first attempt:
photo.JPG

I'm lazy and I suck at soldering, so I went with the 4 x 18650 holsters. Placed three in series and currently have only two in parallel, but plan to place two more on the back side of this plate. It's not yet fully charged in the photo, but comes out to 48v. I soldered the two mains to nickel plates (PCI blanks) that I molded to fit around the plastic plate. The plate is hard to see, because it's clear. Eventually I'll make it into a 'slide-n-plug' pack.

I took it for a spin, and it works (batteries stay cool to the touch), but still needs the extra two packs on the back for the amperage I need. This pack lays flat on my rear basket, but could conceivably fit in a triangle. It's designed such that road bumps won't knock the batteries free from the holsters, but continual testing will prove this design over time. I like the battery holster idea, because I like the ability of swapping out bad cells one by one rather than having to de-solder or build a completely new pack from scratch.

I'm still looking for ways to balance charge this. I only have a two-cell charger that takes overnight to fully charge :(. At this rate I'll be able to ride once a week if I'm lucky.
 
Supersleeper said:
Here's my first attempt:
file.php


I'm lazy and I suck at soldering, so I went with the 4 x 18650 holsters.

I'm still looking for ways to balance charge this. I only have a two-cell charger that takes overnight to fully charge :(. At this rate I'll be able to ride once a week if I'm lucky.

Add a fuse!
1 little short-surge will detension all your springs, making holders worthless.

Get a 48V MeanWell, or clone, After your initial "balance" charge, Meanwell will return pack to same "balance" condition. (Barring any cells hitting "empty", below ~3.5-3.6V)
More basic, get an iMax B6 $20+ and a 12V power supply.

Check each cell voltage at partial discharge. This will pinpoint any weak cells.
Each bank, "s", should consist of cells of equal condition capacity. Weak 48V bank "p" with strong 48V bank.
 
60 mile report - 22.2V 41.6Ah Rebuild pack

25.2V Start
23 miles @ ~17mph = 23.7V
+15.67 Miles @ ~16mph = 22.9V
+21 @ ~15-16mph = 22.2V
~21.6V = optimal DOD

... - 23m ..... 39m ..... 60miles
-
1a = 3.95V ... 3.81V ... 3.71V
2a = 3.96V ... 3.81V ... 3.71V
3a = 3.95V ... 3.81V ... 3.71V
4a = 3.95V ... 3.81V ... 3.70V
5a = 3.95V ... 3.81V ... 3.70V
6a = 3.96V ... 3.81V ... 3.70V
+
+
6b = 3.96V ... 3.81V ... 3.70V
5b = 3.96V ... 3.81V ... 3.71V
4b = 3.95V ... 3.81V ... 3.71V
3b = 3.97V ... 3.81V ... 3.72V
2b = 3.95V ... 3.81V ... 3.70V
1b = 3.95V ... 3.81V ... 3.69V
-
Bank voltage is remaining very close!
Voltages should continue to diverge as pack nears "empty".


Present discharge points between lines.
Present discharge about 80-85% ... estimated by area of graph.

file.php



All banks are equalized within a few 100th V
The real test will be after the 3.7V point is passed.

Concluding test.
On this bike, at the higher gearing, I would rate 22.2V performance as "inadequate".
Happy with the cells!
Unhappy with the lackluster performance of 22.2V. See - 22.2V Battery Woes
Of course the last 36 miles were in sub-freezing weather mostly with a cold battery pack.
Cells show good capacity-condition, minor resorting might make great ... they may become my 20.8Ah 44.4 MAGNUM!

Recharging pack to
4.00V
4.05V
4.10V
4.15V
4.20v
and gauging additional mah gained at each level.

Graph determined using - http://endless-sphere.com/forums/viewtopic.php?f=14&t=33271[/quote]

"Sorting by retained voltage" proved ... fair to good?
After recharging to 4.20V I will remove balance wires and discharge pack.
This should allow a good degree of accuracy in sorting cells.
Low and high cells banked together ... etc.
 
DrkAngel said:
60 mile report - 22.2V 41.6Ah Rebuild pack

25.2V Start
23 miles @ ~17mph = 23.7V
+15.67 Miles @ ~16mph = 22.9V
+21 @ ~15-16mph = 22.2V
~21.6V = optimal DOD

... - 23m ..... 39m ..... 60miles
1a = 3.95V ... 3.81V ... 3.71V
2a = 3.96V ... 3.81V ... 3.71V
3a = 3.95V ... 3.81V ... 3.71V
4a = 3.95V ... 3.81V ... 3.70V
5a = 3.95V ... 3.81V ... 3.70V
6a = 3.96V ... 3.81V ... 3.70V
+
+
6b = 3.96V ... 3.81V ... 3.70V
5b = 3.96V ... 3.81V ... 3.71V
4b = 3.95V ... 3.81V ... 3.71V
3b = 3.97V ... 3.81V ... 3.72V
2b = 3.95V ... 3.81V ... 3.70V
1b = 3.95V ... 3.81V ... 3.69V
-
Bank voltage is remaining very close!
Voltages should continue to diverge as pack nears "empty".


Present discharge points between lines.
Present discharge about 80-85% ... estimated by area of graph.
This is a very nice display of data. I assume this data is no pedaling. With pedaling I would guess over 100 mi range. Keep up the good work. Do you haul cargo at all? If so have you ever collected battery data under more load (Cargo) using laptop cells? Thanks for sharing your data and putting into a easy to read format.
 
DrkAngel said:
60 mile report - 22.2V 41.6Ah Rebuild pack

Present discharge about 80-85% ... estimated by area of graph.


Recharging pack to
4.00V
4.05V
4.10V
4.15V
4.20v
and gauging additional mah gained at each level.

"Sorting by retained voltage" proved ... fair to good?
After recharging to 4.20V I will remove balance wires and discharge pack.
This should allow a good degree of accuracy in sorting cells.
Low and high cells banked together ... etc.

Initial recharge produced fairly equal voltage in the 12 separate banks.
However ... at each 500th V step of recharge, pack endured prolonged and repeated trickle charging at each stage.
Perhaps due to this, voltage between banks diverged.
Final voltage range 4.16V - 4.23V.

Rather than equalizing back to original 4.20V - 4.21V, some factor allowed-induced differing final voltages.

I tend to suspect the repeated and prolonged trickle charges.
I will re-equalize at 4.20V and retest with a discharge and simple 1 stage recharge.

Alternate suspect would be failing cell - cells? ... no longer retaining 4.20V.

41.6Ah pack (4.20 - 3.71V per cell discharge) accepted 32.86Ah recharge, at an estimated 80-85% DOD, for a ~38-41Ah capacity!
So cells appear to be of excellent capacity condition.

Recharging in 1 step from 3.7V to 4.10V would keep all banks in a safe and productive range, but I must delve into the mysterious recharging behavior!
Will post results!
 
Condition - Capacity Testing

Acquire a batch of identical packs!
Remove cells from packs but leave the connecting tabs-straps.

Stage #1 - determine optimal voltage
"Map" a representative cell ... determine optimal peak charge.
ex - Charge to 4.25V, (3.7V cell)
discharge at .5C for 1min, (shorter time for greater accuracy)
let sit 10sec then measure voltage,
record voltage in 100th V,
repeat.
You are looking for the point where voltage drop "stabilizes", this should indicate the relative capacity at various voltages.
You would want to limit your peak charge to within the "beefier" capacity region of the cells.

DrkAngel said:
3.7V - 4.05V takes advantage of the "beefiest" area of the discharge profile.
It also leaves a healthy "emergency reserve", for "limping" home.

file.php

Be aware! These results are for a specific cell type-formulation and might have no relation to other brands-types-formulations!

Stage #2 - charge all cells to test voltage
I charge a large quantity at less than .5C, in parallel, to somewhat above working voltage (.05V?), 4.25V for 4.20V targeted - 3.7V, 4.15V for 4.10V targeted - 3.6V.
I use a S-150-5V Meanwell modified for 3.5V-5.4V adjustable.
Monitor charging and remove any cells showing heat production.

file.php

file.php


Stage #3 - self-discharge - (bleed down) test
Cells are separated and left to sit, I leave paired cells connected.
I measure daily and segregate by retained voltage, the more days the better.
Any cells with notable drop are removed.
With drop slightly below target peak voltage are retained for other projects?

Stage #4 - quick capacity & IR test!
Apply a ~.5C drain to paired cells.
Note and record immediate "sag" voltage - this gives you a comparative IR (Internal Resistance) measure.
(Any pair with 1 bad cell will show a drastic drop and cells should be separated and bad cell trashed.)
Monitor voltage drop, at 15min intervals, 1 hour test should give a nice "comparative" capacity reading - as indicated by resulting voltage.
Run multiple simultaneously using identical discharge devices, light bulbs?, small fans?, etc?
Most multimeters have a <10A function for determining actual discharge amps.

(Not as accurate but faster-easier.
I rig cells as 3s (11.1V) and test using a 12V discharge.
Initial sag for each cell important in determining bad-weak cells - to be replaced before continuing test!)

Resulting voltage, of each pair, after timed test completed, used for comparison with total batch of identically tested pairs.
 
Greetings from a (nearly) complete noob. I am currently starting a project with my two sons (ages 9 and 11) of refurbishing and converting two small gokarts from 49cc two stroke gas engines to electric. I plan to put two 600W 36V brushed motors I got cheap from Craigslist on each cart (one motor powering each rear wheel), and would like to run them at 48V. SLA batteries with sufficient duration to allow a "real race” would make these carts WAY too heavy, so I would like to figure out an affordable way to use Li-ion batteries. I also like the idea of teaching my boys how to make Au out of OPG (other people’s garbage), so I am LOVING this thread.

My question is as follows: I have found a source of Li-ion batteries that seems inexpensive, but being unfamiliar with this, I’m afraid I might be buying someone else’s garbage only to make it my own. Any feedback/advice would be GREATLY appreciated, whether specific recommendations about accessories or techniques, insight telling me that these really are garbage, or a kindly “Don’t be an idiot, just pay the money and buy a safe battery online” kick in the head.

Here are the specifics: Panasonic PA2446U battery packs which were intended for Tecra 750/780 laptops. H says each pack contains 12 Panasonic CGR17670 cells, each rated 1.6A. According to the seller, these packs are unused and still in their factory boxes, but they have been stored for around 4-5 years in his garage. He is offering the packs for $13 each if I buy 30 packs, or just a little more than $1 per cell. I figure from reading this thread that should be enough to make two pretty decent 12s12p packs for the gokarts.

Google tells me that the older 17670 cells typically are 3.6V, 1400mAh. The seller couldn't answer my question about if 1.6A really meant 1.6Ah, and he says that the operating voltage from each pack ranges from 10.8-12.6V, in which case I am not sure how 12 cells might be configured in each pack to get these numbers if my cursory research is anywhere near correct. I don't think he's an electrical engineer, and when I talked to him there was a baby crying in the background, so I recognize he might have been distracted.

He is willing to charge the packs prior to sale, so I would at least be able bring a multi-meter to see if they can hold some charge. My guess is that these are older, lower capacity cells than might be available today, but if these things don’t go bad just sitting in the box, it might be a good find.

Thoughts?
Below are pictures:
Packs.jpgCells.jpg
 
hackerboysf said:
My question is as follows: I have found a source of Li-ion batteries that seems inexpensive, but being unfamiliar with this, I’m afraid I might be buying someone else’s garbage only to make it my own. Any feedback/advice would be GREATLY appreciated, whether specific recommendations about accessories or techniques, insight telling me that these really are garbage, or a kindly “Don’t be an idiot, just pay the money and buy a safe battery online” kick in the head.

Here are the specifics: Panasonic PA2446U battery packs which were intended for Tecra 750/780 laptops. H says each pack contains 12 Panasonic CGR17670 cells, each rated 1.6A. According to the seller, these packs are unused and still in their factory boxes, but they have been stored for around 4-5 years in his garage. He is offering the packs for $13 each if I buy 30 packs, or just a little more than $1 per cell. I figure from reading this thread that should be enough to make two pretty decent 12s12p packs for the gokarts.

Google tells me that the older 17670 cells typically are 3.6V, 1400mAh. The seller couldn't answer my question about if 1.6A really meant 1.6Ah, and he says that the operating voltage from each pack ranges from 10.8-12.6V, in which case I am not sure how 12 cells might be configured in each pack to get these numbers if my cursory research is anywhere near correct. I don't think he's an electrical engineer, and when I talked to him there was a baby crying in the background, so I recognize he might have been distracted.

He is willing to charge the packs prior to sale, so I would at least be able bring a multi-meter to see if they can hold some charge. My guess is that these are older, lower capacity cells than might be available today, but if these things don’t go bad just sitting in the box, it might be a good find.

Thoughts?
If you think about it who else would buy them if you don't buy them? The laptop is way out of date so a user of that laptop is long gone. A dollar a cell, not me, more like 25-50 cents a cell. Your tearing down the cells and testing them. the cells are ok but you may want to look for newer cells because they run 2-2.6AH it would make a lighter weight pack if you need that. I guess what I am saying it's a buyers market and if they don't sell the packs to someone they have to trash them.
 
Tecra 750 was a Win 98 machine.
So ... batteries likely to be 10 - 15 years old.
20th century technology.
Likely, of poor-crap condition ...
 
Thanks for the responses, DrkAngel and emiyata. Will take a pass on this pile of OPG and keep looking and thinking about options while we continue tearing the karts apart and figuring out how to put them back together as Ekarts.

Many thanks, too, to everyone who participates in this awesome forum. Man, do I love the internet!
 
Back
Top