Homemade Battery Packs

Am only a newbie and feel "small" with the wee 10s3p still underway...

Almost went bigger, but decided to do two of these instead of big one, no need speed.

BTW I think the cells are Pan... PFs, still have to assemble the mess, clean off bench first, raining anyway and one thing at a time.

.
... Laptop LiPo vs Samsung ICR18650-22p

eZip RMB battery rebuilds-upgrades
laptop LiPo = 25.9V 25.92Ah with 4.20V - 3.70V usable voltage limited to 1C continuous - 2C surge (recommended)
Samsung 18650 = 25.9V 26.4Ah with 4.20V - 3.40V usable voltage limited to 4.5C continuous - 9C surge (recommended)

eZip 24V 450w motor @ 25.9V can output ~480w
24A controller means pack is never tasked at greater than 1C

Weather is warm and my old 2014 eZip laptop LiPo build is shining again - loves 80 degree weather!
Main advantage is the narrower voltage range keeps up speed and performance, (in warm-hot weather).
29.4V - 25.9V (560w - 480w motor outputs : 20.5mph - 18.2mph)




2018 Samsung build is most impressive in cold weather, the low IR offsets the detriments of a cold battery!
Downside is that its deeper-lower usable voltage noticeably degrades speed and power, as discharge progresses.
29.4V - 23.8V (560w - 430w motor outputs : 20.5mph - 16.9mph)




For casual cruising either is just great, but in my typical WOT mode, the Samsung build is a bit annoying.

Next Project
500w Front Geared Hub (eBikeling)

Just ordered one of the cheap ebikeling 500w geared front hub motors for one of my 8spd disc brake mountain bikes. Will upgrade tires and seat and dedicate as my pedal assist transporter. Bike has trigger shifter built onto brake handles, so twist throttle will work nicely, thumb throttle seems imprecise-annoying - imo.

Anyone have better specs on ebikeling 500w geared front hub motor?
Cruising speed?
No load motor speed?

Will graph up speed and assist capability after received and built.

The newer 18650 high capacity cells have an even lower usable voltage range, 4.20V - 3.20V usable voltage.
29.4V - 22.4V (560w - 400w motor outputs : 20.5mph - 15.9mph)



Performance drop is so pronounced as to be ... inadequate-pitiful?

jcool3 said:

4 SPIM08HP 3.7V 8AH LITHIUM POLYMER BATTERIES 25C 200A SUPER CELLS 24V 36V 48V on ebay https://www.ebay.com/itm/4-SPIM08HP-3-7V-8AH-LITHIUM-POLYMER-BATTERIES-25C-200A-SUPER-CELLS-24V-36V-48V-/183036786061?&_trksid=p2056016.l4276
10 of these in series would run my ebike. 37v Question: best way to charge them and do they need to be balance charged?

Click to expand...

Anyone using these packs on ebikes?

.
... Update on recent battery builds

Floureon 7S 5000mah 25.9V 40C Lipo Battery
Lipo output as rated and after a year of occasional use have maintained within 3/100V equalization. Used mainly with 36V 600w WE brush front hub motor as cruise mode, nice range at ~16-18mph.

DrkAngel said:

Now that the ...
Floureon 7S 5000mah 25.9V 40C Lipo Battery <$60 delivered from USA are all sold ...

"Recommended for rebuild of eZip 24V RMB battery packs.
Will fit up to 4 x 5Ah = 25.9V 20Ah vs oem 24V 10Ah (5-6Ah usable) SLA.
Every 24V controller I've ever tested works properly with 7s 25.9V battery, but a bit faster with some more torque ...



eZip RMB pack internal dimensions:
150mm x 65mm x 230mm"

... I am forced to ration my 5 brick inventory.
2x = 25.9V 10Ah testbed looking good, but will likely add BMS and sell with a refurbished eZip. 20-30A $10 BMS - is perfect for 1 eBike with 20A controller..

Click to expand...

Samsung ICR18650-22p have maintained excellently. Near perfect balance after 7 months on fairly continuous use!

DrkAngel said:

.
... 25.9V 26.4Ah #1 finished

7s12p Samsung cells are capable of 120A continuous 240A surge but controllers are <30A and installed 40A fuse (trying new 58V fuses & holder)
Just finished 1, had to test drive it ... rain stopped for awhile.
No BMS installed! Turned out were only 20A continuous.
Mostly worried about parasitic draw when customers let set for the Winter.

Click to expand...

36V 4.4Ah Hoverboard batteries (Samsung cells)
Used 1st in my Snow Beast and more recently with my WE 36V 600w front hub build, these hoverboard batteries seem perfect. I do charge to a full 42V to insure protection-balance functions. Confirmed via the external points on the BMS. Usually used in pairs, I quick swap pairs and charge at 42V with a modified 48V 150w MeanWell.

DrkAngel said:

While might not ethically be a homemade battery pack ...
... so simple, easy and useful, I had to post it up.

Got a large batch of 36V 4.4Ah (genuine Samsung cells!) hoverboard batteries.
Previously paralleled together 4 for a 36V 17.6Ah battery for testing old brush hub motor.

Anyhow, 3 hoverboard batteries fit nicely in the eZip RMB battery case.

For my Snow Beast, eZip (24V 450w) volted up and geared down for 20mph@36V (675w). The Samsung cells seem very cold tolerant! 13.2Ah might not seem like much but great for my short Winter trips, nice for short spurts near 30mph if used on my cruiser eZips (Dual battery rack and switch allowing 25.9V on one side and 36V on the other). Just flip the switch for TURBO

Click to expand...

Mower upgrade 1 year report
Excellent! Nice balance and durability, after setting since last season, balance was within 2/100V. Review is more about the mower. Being a mulching mower it bogs on heavy or wet grass. On nice dry moderate height grass it does great ... important to keep blade sharp for nest performance.

DrkAngel said:

Mower upgrade
Upgraded my 24V 20Ah SLA mower last year.
Been running with upgraded 22.2V 34.56Ah (6s16p) Old used Laptop Lipo at probably 25Ah++.

Works OK but, since I run as a mulching mower, the extra rpm and torque seems advisable.
So, pushed over the main brick and was able to squeeze in another 1s16p.
Previous 6s were matched in capacity and IR, added bank was just insured to be of better!
Lesser would drag down whole pack - better allows full use, with power to spare.

Yeah, oops ...
6s battery had no fuse installed.
Had to rewire, so dropped in a 40A fuse + holder.
Too wet to try and weather won't dry up for a few days but I expect performance to be noticeably improved!

Home Depot finally has the 20" mulching blade available again so will get a spare, only $9.95 for the Homelite cordless mower.
Straightened and filed the worst nicks out of the old blade but new blade will probably improve even more.

Cells are surrounded and set upon closed cell foam.
Prevents movement and protects from vibration and rough handling.
At probably less than 1C discharge, I am not worried about battery heat.

Click to expand...

33.3V 31.2Ah eZip battery
Not used nearly as much as planned, 36V hoverboard batteries cover all its planned applications. Over 1 kWh, I still plan on trip use or extended cruises. Has good, near rated, capacity and showed excellent balance at last inspection.

DrkAngel said:

... :lol: ...

Finally finished 33.3V 31.2Ah eZip battery!
Naming MaxCell due to the 9s12p being the maximum number of cells that will physically fit in eZip RMB battery shell.
(And to differentiate from my 33.3V 43.2Ah insulated "Cool Cell")

Simple but very snug fit. To secure cells tightly and to protect solder points, top and bottom are covered with 1/8" closed cell foam sheets. Had to put a little weight on case to close properly ... center bows out a bit, center screw removed to allow 9s12p fit.

Side A was nice and straight forward, nice straight runs with heavy tinned copper braid.
Any connection over positive ends of cells were protected with additional HD duct tape!

Side B required a few curves, which was easy if braid curved before flattening.

Installed the new 58V rated fuse holder with 40A fuse. Upgraded charging wire to 18ga, unlikely to charge over 5A, will upgrade to 14ga if I ever want to charge at max recommended 15A.
Entire battery wrapped in clear boxing tape to stabilize and protect. Meter leads poke through easily, for testing.
Sides wedged firmly with sturdy cardboard.

I did add 1/2 cell height wood blocks to support rows 2-4-6-8

MaxCell Specs:
33.3V 31.2Ah = 1038.96mAh - 1kWh+
9s12p = 108 cells
NOS from HP laptop batteries.
72 Sony 2600mAh + 36 Sanyo 2600mAh
Arraigned as 9s8p(Sony)+4p(Sanyo)
Both types effectively "empty" at a static 3.70V
20-21mph (will go 25mph in my upgeared eZips)
Including shell, battery build weighs 12lb 0oz

Compare to eZip OEM SLA battery:
24V 10Ah = 240mAh < 120mAh usable
15-16mph
15lb 9oz

Click to expand...

37V 4400mAh Hover board Batteries (Samsung ICR18650-22p cells) are proving excellent beyond all expectation!
Discharge graph shows cells effectively empty at a static 3.40V.
BMS reportedly protects cells at 2.60V.
2 packs in parallel - Recently discharged deeply, but did not hit BMS LVC, discharged voltage was all cells at 3.33V-3.34V, within 1/100V!!!

With 2 packs, Typical discharge range is 1C (8.8A) -3C (26.4A)
Battery (combined) rated 40A continuous 80A surge.
BMSs (combined) rated 40-60A?

Just putting together 3x 37V 4.4Ah(37V 13.2Ah) "pack" for local eBiker, as auxiliary pack for with his 6 year old 20Ah LiFePO4 battery.
12 hour LiFePO4 2A charge vs 3 hour Li-Co 5A charge!

DrkAngel,

Are these packs for sale? Is it possible to ship to Malaysia? Thanks!

copied post of easy cell testing procedure

Test cells the way I test used cells

1. self discharge
Charge all cells to same exact voltage
Let set several days
Discard any with noticeable voltage loss

2. IR (Internal resistance)
Comparative IR - identical cells under identical discharge will sag in voltage to the same voltage
With cells of identical voltage
Connect discharge device to cells, individually, in turn
check and note cell voltage at a specific time during discharge
Discard any with excessive voltage sag

3. Capacity
From full voltage, run a metered discharge
discard any with substantial capacity loss
or
after marking each cell with capacity, build banks of identical capacity

Bulk testing
After self-discharge test
String batches of 30s cells in series = ~120V
Discharge using 2x 60w light bulbs = 120w
Monitor all cells
Each hour of discharge = 1Ah of capacity
Stop discharge when any cell hits low voltage
3.40V - very safe discharge point, don't recommend below 3.00V
Label each cell with residual voltage
1hr = 1000mAh + 3.95V?
2hr = 2000mAh + 3.75?
3hr = 3000mAh + 3.45?

Bulk building
With all cells identically discharged
Sort all cells from best remaining voltage to worst
Sort into banks with best cells to worst
10s >> 1 2 3 4 5 6 7 8 9 10 10 9 8 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 9 10 etc
This will create banks of nearly identical capacity ... banks will safely balance if within ~.1V ... barring any cells of radically different voltage ...
Combine cells - since all cells are free of self-discharge and were identically discharged, simply bulk charge to full pack voltage
Good Lithium cells, discharged at moderate rate will bulk recharge right back to their starting voltage!
Should have good solid pack of reliable tested cells.
Best news ...
Should be good for bulk charging ... with no need for balance charging!

Oops! I forgot, good for bulk charging if pack of sufficient size for moderate discharge rate.

Thanks for the awesome Wiki!
Quick question... what program do you guys use to determine size/shape/cell count/orientation of 18650 cell pack? I have a full suspension frame and need to get creative to pack as many cells in it as i can

.

dirt huffer said:

Thanks for the awesome Wiki!
QuNick question... what program do you guys use to determine size/shape/cell count/orientation of 18650 cell pack? I have a full suspension frame and need to get creative to pack as many cells in it as i can

.

Click to expand...

No known "programs"

Mapping Cell Placement

Measure and draw available space, to scale, on drawing paper.
Break open some rolls of pennies. (19mm penny is close to 18mm cell size)
Arrange into and tightly fill "available space".
Count pennies 120?, divide by desired voltage 48V = 13.
120/13=9.23 9x13=117 pennies-cells
Arrange into 13 piles of 9 pennies.
Mark each pile with a number (underline 9 to avoid confusion with 6)
Arrange into "space".
You should now have a plan-map.

Skalabala said:

stealthhack, buddy be careful with what you are doing.
I see you have 3.6V mixed with 3.7V cells. So the max voltage should not go higher than 4.05V and lowest 3.75V otherwise you will be looking for trouble with old cells!
Did you do a sag test on each bank while putting a 1C load on the whole pack?

Read a bit more in this thread As you have a really BIG bomb there! :lol:

Click to expand...

Hey there,
I am following this thread for sometime. the point you are mentioning, mixing 3.6 cells with 3.7V cells, seems interesting. Why do youthink it's a problem? why the voltage limit of 4.05 to 3.75??

Thanks

DrkAngel said:

tomjasz said:

I've got some new salvaged 18650 batteries from laptop packs the read 3.7v, 24 of them, the packs say they were 14.4 volts but that doesn't add up. 8X3.7=29.6. Huh? Also I have 18 from a 3M projector that rated the battery pack at 10.85 or 11.1v. They read 3.6v. That math pans out. Please explain. 18650 batts that are 3.6v not 4.2? How do I utilize them? I was hoping for 7s for replacement in my little Schwann scooters for the 24v SLA. How do I charge them? Build them into a 7s6p and use a Tenergy 2a 25.9 smart charger? Can a balance charger be used? Wouldn't that be better since you seem to say these should be charged a lower input? Confused but having fun at collecting the goods. Soon to have a new salvage source that may be very prolific. So I need to figure this charging thing out. Is there a BMS out there that will work? Is there a balance charger that can be used as well?

Click to expand...

4s2p 4 x 3.6V = 14.4V

I would recommend against using 3.6V and 3.7V in the same pack ... unless you use an equal assortment in each bank and limit usable voltage between 3.7V and 4.10V?

Using the iMax B8 (will balance charge 7s) as an example, there are 2 LiCo charge settings.
Lion @ 3.6V for charging to ≤4.10V and Lipo @ 3.7V for charging to ≤4.20V

After removing any cells with self-discharge and capacity matching each bank, balance charging at every cycle becomes unnecessary.
I do recommend the cell level voltage meter-alarm as a pack monitor-safety device. see - eBike Toolbox

Click to expand...

Hey DrKAngle, I have followed your post for a long time. They are very informative. I am building my pack. I have Li-ion cells from laptop batteries, 10.8V (3.6 nominal), 11.1V (3.7 nominal). as per your comment, i should not mix them? are the chemistry of these cells so different?

3.6V vs 3.7V rated cells are rated as the average discharge voltage over the entire discharge cycle at a specified discharge rate.
Factored are Internal Resistance (IR) and rated DOD
Not as the discharged "empty" voltage, more a coincidence than a "rule", with older LiCo cells.

Previously, not so important if equally distributed among banks.
Mixing old tech with new tech has become a much more important factor.
(Typically not a factor with Laptop cells.)

Older LiCo cells were effectively "empty" near 3.6-3.75V.
Newer tech cells drop this "empty" point to 3.4V or even 3.2V.
Mixing will either waste capacity on the newer cells or dangerously over discharge the older cells.

4.10V was recommended as optimal charged voltage due to observed optimal capacity vs enhanced cell life.
"Optimal" varies by cell "model" and requires mapping to determine ... but can often be estimated by looking at manufacturers discharge graph.

Just don't forget to test cells!
self discharge
IR
capacity
etc.

Stage 3 Comparative IR
Test cell pairs for a comparative internal resistance.
While monitoring cell voltage, apply modest drain (2A?) and note voltage sag.
Remove any with notably greater voltage sag. These are likely very poor capacity cells or one of pair is bad. Separate and test for alternate use, flashlight etc.


how do you define, notably greater voltage sag?

When testing same brand-model cells:
If 1 pair sags .10V and another sags .20V ... likely, one cell is bad.
Separate and test individually, either 1 cell is bad or both are poor.

Use of "comparative IR" is to compare cells against same brand-model.

"Cheap" IR meters seem inaccurate to ridiculously non-functional.
Proper IR meters, typically, test at 1C. ... IIRC

DrkAngel said:

3.6V vs 3.7V rated cells are rated as the average discharge voltage over the entire discharge cycle at a specified discharge rate.
Factored are Internal Resistance (IR) and rated DOD
Not as the discharged "empty" voltage, more a coincidence than a "rule", with older LiCo cells.

Previously, not so important if equally distributed among banks.
Mixing old tech with new tech has become a much more important factor.
(Typically not a factor with Laptop cells.)

Older LiCo cells were effectively "empty" near 3.6-3.75V.
Newer tech cells drop this "empty" point to 3.4V or even 3.2V.
Mixing will either waste capacity on the newer cells or dangerously over discharge the older cells.

4.10V was recommended as optimal charged voltage due to observed optimal capacity vs enhanced cell life.
"Optimal" varies by cell "model" and requires mapping to determine ... but can often be estimated by looking at manufacturers discharge graph.

Just don't forget to test cells!
self discharge
IR
capacity
etc.

Click to expand...

Thank you so much for the valuable comments. I am doing the IR, SD and apacity tests. I just want to understand, when you say Older LiCo cells and new ones, which year of manufacturing you are talking about? (i mean for old and for new).

Thanks

Only in the past 2 years have I noticed any of the lower voltage cells in any laptop packs. Most are still "standard " LiCo. Mixing together, a few, has no determent except not taking full advantage of full capacity of, the "few".
Just maintain a universally safe ≥3.70V static discharged voltage, measured after throttle released.

.
... Cheap "new" 36V 4.4Ah hoverboard batteries.

Just had offer accepted for "Random 36V 4.4Ah hoverboard battery packs (new) Samsung or LG" @ 4 for $100, $25 each.
"Guaranteed delivery" by Monday.
Will post up info as soon as delivered, link and recommendation as soon as preliminarily tested.

Update - Tracking shows Saturday delivery.

DrkAngel said:

Just had offer accepted for "Random 36V 4.4Ah hoverboard battery packs (new) Samsung or LG" @ 4 for $100, $25 each.
"Guaranteed delivery" by Monday.
Will post up info as soon as delivered, link and recommendation as soon as preliminarily tested.

Update - Tracking shows Saturday delivery.

Click to expand...

Received the 4 packs.
All labeled 36V 4.4Ah
2 labeled with Samsung, other 2 of no known manufacturer.
Only 1 charges to near 4.20V!
Discharged all below 3.40V and doing metered charge on each. Will have results tomorrow.
Have not broken open double shrink wrapped packs yet, will after charge results tomorrow.

Not charging to full voltage is typically the result of a low capacity cell, or cells, which will charge higher triggering the high cell voltage safety-disconnect.

DrkAngel said:

DrkAngel said:

Just had offer accepted for "Random 36V 4.4Ah hoverboard battery packs (new) Samsung or LG" @ 4 for $100, $25 each.
"Guaranteed delivery" by Monday.
Will post up info as soon as delivered, link and recommendation as soon as preliminarily tested.

Update - Tracking shows Saturday delivery.

Click to expand...

Received the 4 packs.
All labeled 36V 4.4Ah
2 labeled with Samsung, other 2 of no known manufacturer.
Only 1 charges to near 4.20V!
Discharged all below 3.40V and doing metered charge on each. Will have results tomorrow.
Have not broken open double shrink wrapped packs yet, will after charge results tomorrow.

Not charging to full voltage is typically the result of a low capacity cell, or cells, which will charge higher triggering the high cell voltage safety-disconnect.

Click to expand...

3 of 4 packs defective!
Discharged below 34V then metered charge with 3 of a metered capacity below 2.0Ah and will not accept 42V charge.
Indicates old(?) cells of differing voltage that hit cell low and/or high BMS voltage cutoffs.
Disassembled 1 pack and manually balanced cell voltages and will cycle again to note high and low voltage balances and capacity.
Already complained to seller and awaiting reply.
Not willing to recommend until response received.

DrkAngel said:

DrkAngel said:

DrkAngel said:

Just had offer accepted for "Random 36V 4.4Ah hoverboard battery packs (new) Samsung or LG" @ 4 for $100, $25 each.
"Guaranteed delivery" by Monday.
Will post up info as soon as delivered, link and recommendation as soon as preliminarily tested.

Update - Tracking shows Saturday delivery.

Click to expand...

Received the 4 packs.
All labeled 36V 4.4Ah
2 labeled with Samsung, other 2 of no known manufacturer.
Only 1 charges to near 4.20V!
Discharged all below 3.40V and doing metered charge on each. Will have results tomorrow.
Have not broken open double shrink wrapped packs yet, will after charge results tomorrow.

Not charging to full voltage is typically the result of a low capacity cell, or cells, which will charge higher triggering the high cell voltage safety-disconnect.

Click to expand...

3 of 4 packs defective!
Discharged below 34V then metered charge with 3 of a metered capacity below 2.0Ah and will not accept 42V charge.
Indicates old(?) cells of differing voltage that hit cell low and/or high BMS voltage cutoffs.
Disassembled 1 pack and manually balanced cell voltages and will cycle again to note high and low voltage balances and capacity.
Already complained to seller and awaiting reply.
Not willing to recommend until response received.

Click to expand...

Packs appear to be bulk rejects ... seller sending 3 replacement packs.
"Repaired" 2 packs:
replaced 1 low capacity pair,
Replaced 2 low capacity pairs ... then noticed both removed pairs were never banded together ... cheap-unskilled labor.
3rd poor pack has no repair possible. Well, it does have "Samsung" cells, some dead "Samsung" cells ...
10 no brand generic blue cells and 10 "Samsung" cells none of which (Samsungs) were connected, with 0V ... for weight or to claim "Samsung" ???
Will post up pictures.

Can't really recommend this auction!
But will post up link ... Only for the DIYer!
Cell salvage?
Oops, sale ended anyhow...

.
... "36V 4.4Ah Hoverboard Samsung Battery"

Wow! Samsung cells!





Whaa?





What the hell!!!





All "Samsung" cells dead with 0V, and yes, not even connected.

.

Battery repair success.
3 of 4 defective batteries now metered at 4Ah+.
Got 3 more replacements, likely defective too, scheduled for delivery next week.
Could be as many as 6 good batteries for $100 ... and some work. 36V 26.4Ah .733kWh

Hoverboard batteries, typically, do not have the balance function with their BMS.
Opening the pack and manually balancing recommended, if usable capacity drops, or fails to fully charge to 42.0V.

DrkAngel said:

Battery repair success.
3 of 4 defective batteries now metered at 4Ah+.
Got 3 more replacements, likely defective too, scheduled for delivery next week.
Could be as many as 6 good batteries for $100 ... and some work. 36V 26.4Ah .733kWh

Hoverboard batteries, typically, do not have the balance function with their BMS.
Opening the pack and manually balancing recommended, if usable capacity drops, or fails to fully charge to 4.20V.

Click to expand...

Saw that listing originally. Glad I didn't bite on it. Thanks for posting.

3 replacement packs arrived ... Just as bad as the 1st batch, except ... no pretense of being Samsung - no branding on cells!
Labeled 2015, one with a leaking cell pair.
Opened up and balancing to same voltages before 1st capacity test.
Replacing leaking pair with genuine Samsung.
No need for further reporting! ... ?