My excellent adventure through some battery builds

Batteries, Chargers, and Battery Management Systems.
BeachRider2016
100 W
100 W
Posts: 110
Joined: Jul 13, 2016 10:42 am
Location: North Carolinas

Re: My excellent adventure through some battery builds

Post by BeachRider2016 » Oct 24, 2016 9:11 am

Good updates, Ive been trying to figure out how to arrange in a configuration to fit 400 batteries in an ebb frame with your setup.
It seems your setup is limited to 2 rows unless some type of removable connections made to join the packs together/ easier to take apart. Probably too much space consumed for the limited space vs maxium cells in the ebb frame.
This method probably best for attaching to a frame bike- maybe a winter build :mrgreen:

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Oct 24, 2016 3:47 pm

BeachRider2016 wrote:Good updates, Ive been trying to figure out how to arrange in a configuration to fit 400 batteries in an ebb frame with your setup.
It seems your setup is limited to 2 rows unless some type of removable connections made to join the packs together/ easier to take apart. Probably too much space consumed for the limited space vs maxium cells in the ebb frame.
This method probably best for attaching to a frame bike- maybe a winter build :mrgreen:
The benefit for battery holders is that replacing bad cells is easy. Since I'm running used laptop cells, a few failed cells is probable. In fact, in the 240 cells there have been I think 10 cells that passed my initial load testing almost a year ago and have since failed. Popping out a cell and replacing it with another is stupid easy, but it also comes at the cost of not being the most compact way to have 18650 cells arranged. There's 1/16" between each cell and 1/8" where two battery holders join. Then there is the gap between the two halves...another 1/4". It all adds up. So if you have room and can waste some, the battery holder idea isn't so bad, but if you need a compact battery, then just buy new cells and arrange them into a pack that doesn't waste space.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Oct 24, 2016 3:59 pm

ElectricGod wrote:
BeachRider2016 wrote:Good updates, Ive been trying to figure out how to arrange in a configuration to fit 400 batteries in an ebb frame with your setup.
It seems your setup is limited to 2 rows unless some type of removable connections made to join the packs together/ easier to take apart. Probably too much space consumed for the limited space vs maxium cells in the ebb frame.
This method probably best for attaching to a frame bike- maybe a winter build :mrgreen:
The benefit for battery holders is that replacing bad cells is easy. Since I'm running used laptop cells, a few failed cells is probable. In fact, in the 240 cells there have been I think 10 cells that passed my initial load testing almost a year ago and have since failed. Popping out a cell and replacing it with another is stupid easy, but it also comes at the cost of not being the most compact way to have 18650 cells arranged. There's 1/16" between each cell and 1/8" where two battery holders join. Then there is the gap between the two halves...another 1/4". It all adds up. So if you have room and can waste some, the battery holder idea isn't so bad, but if you need a compact battery, then just buy new cells and arrange them into a pack that doesn't waste space.
The 6S6P pack was an experiment to see if I could make it work with internal solder connections and the answer is a definite yes. The next step will be to build packs that are 12S4P and then connect them together into a 24S4P arrangement. I plan to use them in my moped. It's a matter of 4 interconnects on the vertical solder tabs rather than 2. I'm undecided, but probably two will come out the top and two out the bottom for balance wires. Or maybe since it will be spread apart anyway, I can make the folding section shorter and solder the balance wires before squashing it all together. I really only need access to the positive and negative ends of the entire pack for power so accessing all the interconnects is just for balancing.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 06, 2016 12:42 pm

Hi there! Really nice build log you got there for your blue scooter! Very extensive and detailed build i must say!

Non the less.. I was more interested in your laptop battery build..

Did you make some kind of special rig where to charge / discharge so many batteries?

How did you detect when some cells refused to go over 3.9v?

I also plan on some battery testing but all I came up with was to attach the batteries to a ligh bulb, then add a voltage alarm meter, so that the alarm goes off once the cells reach 3.2 or so.. then recharge them in balance charger and see how much capacity is put back into them..

For me, the longest step would be charging the batteries, as I've got only 6s capable charger and I would need to make some sort of box with balance leads included, to be able to charge 6cells at once.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 07, 2016 3:33 pm

Your 6S charger will charge a lot more than 6 cells at a time. I bought a bunch of 18650 battery holders and then soldered them in parallel so each channel on my 6S charger really had 6 cells in parallel. When charging, I maxed out the charge current since at worst they are 2600mah cells x 6 or 15600mah. My charger will only do 10 amps per pack so maxing out the charge current was always safe. That gets me 36 cells charging simultaneously on a 6S charger. I also have a 6S4P battery holder and a couple of 6S1P battery holders. I really should build two more 6S6P battery holders since my main charger has 2 channels and I have a single channel 6S charger as well. That would allow me to charge 108 cells simultaneously.

In my scooter, I have 6 20S2P packs all attached to the same 20S BMS. It makes for a lot of wires in the battery box, but the one BMS manages all 240 cells. This is how it is done in commercially built packs that have multiple parallel cells. A single BMS sees 5 2600mah cells as a single 13000mah cell or how ever the battery pack is built out. My BMS has 12 cells in parallel on each channel by 20 in series or 20S12P. Then I have put together 4 meanwell power supplies in series to get to 82 volts. I just plug is the charger to the XLR port on my scooter and it charges everything at the same time.



When it comes to load testing cells, I use the same 6S6P rig I charged with or a 20S2P battery holder I built for my scooter. I attach an RC battery monitor to the balance connector or connectors. The monitors alarm when any of the channels dip below the minimum battery voltage limit. I originally used a couple of blow drier elements and blower as my load, but it was really noisy and dangerous. I have since found on ebay a seller of 55 watt, 12 volt halogen bulbs. They were 99 cents each including shipping. I purchased 28 of them so I could set up 4 parallel runs of 7 bulbs to load test at 82 volts. I would post some pictures, but I never took any of it. The individual bulbs have 1/4" blade connectors on them so it's super simple to make up a bunch of jumper wires with that connector. I can reconfigure the whole array any way I want. I have a that spare 20S2P pack I made. This is probably the best way to test cells since I can test 40 of them at a time with only 2 cells in parallel. My first test after charging is to confirm all cells charged to 4.1 volts. They don't always get there so any cell that won't fully charge is obviously defective. I then fill the battery holder with freshly charged cells that all measure 4.1 volts. I then plug the battery pack into my watt meter and confirm I'm pretty close to 82 volts or 20x4.1. The watt meter gets plugged into the bulb array. At 4 parallel arrays of bulbs I should draw something like 18 amps or 55 watts/12volts=4.58 amps. 4.58 times 4 arrays = 18.3 amps. For laptop cells in a 2P setup that's too much current draw since they are typically 1 or 2C discharge rated. I disconnect a couple of bulb arrays to get current draw down to what the cells can handle. I don't need to actually run at the discharge rate of the cells. Well under that is fine too. It just takes longer to load test. The idea is that the battery monitors will alarm when a channel drops below the minimum voltage. I unplug the light bulb load and stop the test. I then pull the cells on the channels that were alarming. Sometimes, the rest of the channels are only slightly better off than the depleted channels, sometimes due to weak cells, that channel will have reached 3 volts and the rest of the cells in the pack are still running at like 3.7 volts. At least one of the two cells on that channel are weak. Those two cells go into a retest container for later. I replace them with other known weak cells that are fully charged. I'm now looking for more cells from the test set to drain too quickly. IE: The next sets of cells to run down too quickly if everything else is still running well go into the retest bin. If the monitors alarm pretty much all at the same time and the watt meter is showing close to the expected Ah capacity for those cells, then that 40 cells go into the "good" pile". The weak set of cells is rarely two weak cells together in parallel. It's almost always just one of the two. As a result, every thing that went into the weak bin, get recharged and retested to find the actually weak cells. These tests are pretty quick since the majority of the cells are too weak for EV use. They run down quickly under a couple of amps load. The cells from that test that show mostly charged while everything else is depleted get set aside for another test run. I recharge them and mark them as possibly good and then take known good fully charged cells to make up the difference in the 20S2P pack and load test all of them. If I did my weak test correctly, then the cells I am testing now are probably good cells and everything in the pack depletes pretty close to the same rate. Wash, rinse and repeat until all cells are tested. By now if I started with a bunch of salvaged laptop packs/cells, I have 4 piles of cells. Ones that are dead, get recycled. They can't be reused so what's the point in keeping them. Ones that didn't charge to 100%. I try to charge them again. If that still fails, they go to recycling too. They are so badly damaged inside to be of minimal use. Cells that ran down too quickly get labeled as weak and I use them for flash lights or whatever. The strong cells have to measure close to factory capacity. Laptop cells typically come in 2600mah if they are from name brand packs. These are used laptop cells, one can't expect perfection. As a result if they are better than 2000mah, then I consider them to be "good enough for EV use". If you are using used laptop cells for EV use, be sure to go with more in parallel to make up for their lower C rates. My scrounged cells have been working reliably for me for a good long while now. In my mind it was a great venture that panned out pretty darn well considering how I have maybe $200 into it. Don't ever expect laptop cells...even brand new ones...to perform like expensive high C rated, high capacity 18650's. These are free/used scrounged from laptop packs. Also, it is possible that after you have tested all the cells and are only using the good ones, that you might have a random cell here and there go bad. I think in the 240 cells I am running that 7-10 have died. With the battery holders it's simple to pop out a bad cell and pop in a good one. I'm always looking for more laptop packs to take apart. As a result, I have accumulated a pretty decent stock pile of used/good enough cells for whatever I choose to put them in.

Making your own charging rig for however many parallel or even 1P is pretty easy. You obviously want to balance charge them. If you look at a 6S balance cable on a lipo pack there are 7 wires in it. Usually one end has a black wire and the other a red wire. Black for minus and red for plus. It doesn't really matter, just use a DMM to determine which end is the positive end vs the negative end in the balance cable. These are the positive and negative ends of the battery pack. Other than being able to only handle an amp or so, they are identical to the positive and negative ends of the pack. The wires in between are the interconnects between cells. Using 18650 battery holders, you are doing the exact same thing as what is in a LIPO pack. Connect the individual holders together in parallel and series to get the results you want. So for my larger charger pack...6S6P...that means I can put 6 cells in parallel on each channel by 6 cells long. I made my pack 3 cells tall by 6 cells long and then made two mirror images so that I could solder them together into a two sided pack for 6S6P. Since each side is a complete 6S1P pack, solder a 6S balance cable onto one side only. Think of the cell closest to the negative end of a battery pack as cell 1 and the one closest to the positive end cell 6 in our 6S pack. Whatever solder connection is the negative end of your pack, gets the negative wire on the balance cable. The next wire goes on the connection between cell1 and cell2. The next wire goes at the connection between cell2 and cell3 and so on. Finally the red wire in the balance cable goes to the positive end of cell 6. Before you load the pack with cells, put your DMM on ohms and confirm that each battery holder position is what you think it should be and in parallel wit what you think it should be in parallel with. A misconnection is a dead short for battery or your charger. Also, label every battery holder position for proper cell orientation. A 6S6P pack with a single cell in place backwards will kill that one cell very quickly. Just lable everything to minimize mistakes. I've done this before and the battery got pretty hot, but there was no fire or explosion...just a hot, dead cell.

On the first page of this thread are a couple of drawings of battery packs with BMS and also balance cables. you can follow them for proper balance wire placement. Also, the 2nd page shows tons of pics of my 6S6P pack.
Last edited by ElectricGod on Aug 30, 2017 5:30 pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 07, 2016 8:45 pm

This is my load made of a bunch of 55 watt halogen bulbs. The bases are steel so I used them to solder all the bulbs together.

Image
Image
Last edited by ElectricGod on Aug 30, 2017 5:37 pm, edited 1 time in total.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
Ianhill
100 kW
100 kW
Posts: 1129
Joined: Sep 25, 2015 5:55 pm
Location: Wales UK

Re: My excellent adventure through some battery builds

Post by Ianhill » Dec 07, 2016 9:51 pm

I like that a lot the h4 halogen come cheap as anything, get a shunt on it I bet there's a fair amount of inrush current till it gets up to temp.
It looks like 4 groups with 28 lamps total so are they in 7 per group ? its hard to see.
From my math I work them out to be 2.6ohm per lamp so seven in series would be around 18ohms then placed in 4 parallel gives a total of 4.5ohms and 1600w rated.
So say we test a 20s load 82v, 1495watts will flow or 18.2amps
10s load 41v, 375watts will flow or 9.1 amps and so on that's if it weird like I guessed at
But these bulbs do have a big inrush current so the initial current will be higher but it's a good rig for higher volts and as your pack splits I think you may have calculated it to hit the sweet spot for what you need good work.

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 08, 2016 1:55 pm

Cool, thanks @ElectricGod for such an extensive reply!

I've built a 6s 4P pack out of powertool battery, which needed to 4 extra cells to make it 6S, as it originally was 5S2P.

I've considered building a laptop battery pack, though for my voltage it looks like I would need almost 20cells in parallel, to comfortably withstand 20-30A current.

Im actually having a mountainboard, the typical amp consumtion seems to be in 10-20A range, that would be around 300-400w. So, each cell should be able to give around 1amp for sure.

max Peak I reached is 60Amps with the pack I have now. Though, the cells are high discharge ones, about 20A per cell.
So I dont expect the laptop batteries to be able to sustain such high load.. but 40A would be great, if they could perform 2A per cell, which would already get them hot probably..

It it would not work out.. I could probably use the pack for e-bike build or such..

--

Thanks for pinpointing out to keep 2cells in paralel, I assume it is easier to keep it that way, as they are usually arranged as 2P.
I know about going parallel when charging, I just wanted to monitor each cell, so that I dont have to test each pair once again, if the cell pair shows weak results.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 08, 2016 9:12 pm

Okami wrote:Cool, thanks @ElectricGod for such an extensive reply!

I've built a 6s 4P pack out of powertool battery, which needed to 4 extra cells to make it 6S, as it originally was 5S2P.

I've considered building a laptop battery pack, though for my voltage it looks like I would need almost 20cells in parallel, to comfortably withstand 20-30A current.

Im actually having a mountainboard, the typical amp consumtion seems to be in 10-20A range, that would be around 300-400w. So, each cell should be able to give around 1amp for sure.

max Peak I reached is 60Amps with the pack I have now. Though, the cells are high discharge ones, about 20A per cell.
So I dont expect the laptop batteries to be able to sustain such high load.. but 40A would be great, if they could perform 2A per cell, which would already get them hot probably..

It it would not work out.. I could probably use the pack for e-bike build or such..

--

Thanks for pinpointing out to keep 2cells in paralel, I assume it is easier to keep it that way, as they are usually arranged as 2P.
I know about going parallel when charging, I just wanted to monitor each cell, so that I dont have to test each pair once again, if the cell pair shows weak results.

I have yet to find any laptop cell that can't withstand 2C. Also, since they are used and free, burn 'em up. LOL! Who cares if they can't take the load. THEY ARE FREE!!! I run 12P at 50 amps or slightly over 2C 100% of the time. It works quite well. On a hot summer day where I live it gets to 100F. The batteries rarely get above ambient. IE: You can run them harder than you think. Don't bother with thinking in terms of 1C just because that's all the more a laptop might draw.

A mountain board needs an efficient, small, low weight, high delivery pack. You just don't have the space for an over sized pack made of laptop cells. I think this is an application where new and high discharge cells is the only choice. I would personally NOT use laptop cells for this application. On a bike or in my case a scooter where weight/size is not overly critical, well that's different. In that application, I can tell you from personal experience that it is a viable option. Let me re-say that...a VERY viable solution. Free is stunning when an equivalent pack made of EV grade cells is going to cost you $1200...which is what I would have paid to get the capacity I have in free/scrounged laptop cells. IE: Choose what you are willing to compromise on and what you need. If you have to be overly idealistic about everything, then that means you are going to spend a lot of money when you don't necessarily need to.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 08, 2016 9:27 pm

Ianhill wrote:I like that a lot the h4 halogen come cheap as anything, get a shunt on it I bet there's a fair amount of inrush current till it gets up to temp.
It looks like 4 groups with 28 lamps total so are they in 7 per group ? its hard to see.
From my math I work them out to be 2.6ohm per lamp so seven in series would be around 18ohms then placed in 4 parallel gives a total of 4.5ohms and 1600w rated.
So say we test a 20s load 82v, 1495watts will flow or 18.2amps
10s load 41v, 375watts will flow or 9.1 amps and so on that's if it weird like I guessed at
But these bulbs do have a big inrush current so the initial current will be higher but it's a good rig for higher volts and as your pack splits I think you may have calculated it to hit the sweet spot for what you need good work.
I use a 50 amp shunt...smallish for an EV that I might ride...but plenty for load testing. I just don't bother if I don't have at least 150 amps in the shunt on an EV. Even the Currie has a 200 amp shunt. It's not a size issue as I have found very compact shunts. Cost between 50 amps and 300 amps is mere dollars and in neither case it isn't more than $15.

The inrush is pretty significant. I have XT60's on everything. I make sure I plug in the battery pack first because whatever gets plugged into the load is going to get scorched pretty good. I assume that I'm going to get a limited life out of whatever connector is on the load. This "load" aka halogen bulbs is more or less infinitely adjustable for whatever I want to test...all the way up to 324 volts or all those bulbs in sereis. To someone just now considering breaking over 100 volts, 324 volts seems ridiculous despite a lot of EV motor bikes running at 800 volts..I'm thinking my load tester is sufficient for the day. Geez...how do you get a 195S/800 volt BMS?
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 09, 2016 5:10 am

@ElectricGod thanks for clarifying the real capability these laptop cells can do..

Well, if you say that they can withstand 2C currents, then I might be okay with just 12P. The pack would already be way over my needs, as I plan on doing a max of 30km, or 15km one way.. as the feet really feels the road when you ride a mountainboard (at least even with pneumatic tires, the deck I have is flexy, so I get to absorb a lot).

So with such a large pack - ~826Wh (at 8s, 12P, each cell around 2.1ah), I might be able to do around 40km, seems to be ''eating up'' about 20wh/km.

Anyways, The largest I've seen is something like 12S8P I think, that is 96cells, so with such count of cells I might go with 8s (should get a new esc) and then the amps should go down a little bit too. I always though laptop cells can go 1C-1.5C comfortably.. but you are really proving otherwise! So yeah, if they can withstand at least 200-300 such harsh cycles.. I think it is somewhat okay to go this route.. especially if the batteries are in sleds, too.

I will check the dimension of what you got with your holders.. These mountainboard decks are usually quite long, so If I got 96 cells total, that is 48 for one ''row''. Which would translate to about 96cm, I believe, if they were almost side to side without sleds. Might do them in two rows, two layers, the enclosure will of course be more bulky but this way it would be possible to cut the lenght shorter.

Here, I can show you a pic from another forum, this guy got 12S4P and he fitted it onto a longboard! Of course, that's only 48 cells, but it is possible to see, there''s still place for about 16 cells in the next 2 rows, where he has places esc (which for mountainboard would go on top of the deck)

Image
It is user's Ackmaniac build. Actually, I might orient the cells like he does, at least it looks like it is way better to do it this way, if you got sleds.

Im not sure what the total weight of these cells are.. but if they are layed out uniform and not in a brick, I think it should be okay.. and should not interfere

with steering or maneuvrability of the board :)

---

Anyways, Should start to consider where to get such high amounts of cells now.. Im not sure I can arrange a deal with pc shops in my country.. as I have an impression they might just give me a weird look, If I ask them.

Sorry for going in detail but im just very interested in trying to make a laptop cell battery build.. Im not sure but perhaps it is even possible to mix them with better, higher discharge cells.. as it seems you are also mixing lipo and li ion, so li ion and li ion should go together even better..
Last edited by Okami on Dec 14, 2016 5:00 pm, edited 1 time in total.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 09, 2016 5:34 pm

I work for an IT company. My office is about 250 people and we all have at least one laptop. As a result, people have batteries go bad from time to time. We have battery recycling buckets in several places around the office. I check those buckets from time to time and our in house computer guys all know I want every laptop pack they come across that is getting replaced. A while ago, we had 40-50 old laptops that had been retired. The laptops needed to be sent back to our corporate office, but the batteries did not so I got all of them. I have no idea about local computer stores. They have to recycle the old laptop packs by law where I live, but I think it costs them to recycle them. I probably could go to the local computer places around me and get more laptop packs than I could ever deal with since I would take them off their hands for free. LOL...maybe I should start a battery business. I could take the packs apart and collect the good cells for sale. I could probably get $2 per cell.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
spinningmagnets
100 GW
100 GW
Posts: 10749
Joined: Dec 21, 2007 10:27 pm
Location: Ft Riley, NE Kansas

Re: My excellent adventure through some battery builds

Post by spinningmagnets » Dec 09, 2016 8:03 pm

Among "serious" battery pack builders (who might own a spot-welder) the plastic holders get a lot of distain, but...with the cells being free? I can see the appeal, especially for a stand-up scooter that can be folded and then carried on a train or bus.

For a pack that is higher voltage, but low amps? I think this isn't a bad solution. The big sellers like Luna and em3ev don't have anything to worry about, nobody will sell these due to liability concerns, but there are many plans posted in several public forums (makezine, instructables, this thread, Drkangel, etc). 98% of battery pack customers just want to buy a finished pack, no garage building.

You are using higher pack-amps from this than I would, but other than that?...if I had access to free laptop cells, I could see me doing this (ie, beef up the series connections to keep them as cool as possible). Great pics, thanks for posting...

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 09, 2016 9:03 pm

spinningmagnets wrote:Among "serious" battery pack builders (who might own a spot-welder) the plastic holders get a lot of distain, but...with the cells being free? I can see the appeal, especially for a stand-up scooter that can be folded and then carried on a train or bus.

For a pack that is higher voltage, but low amps? I think this isn't a bad solution. The big sellers like Luna and em3ev don't have anything to worry about, nobody will sell these due to liability concerns, but there are many plans posted in several public forums (makezine, instructables, this thread, Drkangel, etc). 98% of battery pack customers just want to buy a finished pack, no garage building.

You are using higher pack-amps from this than I would, but other than that?...if I had access to free laptop cells, I could see me doing this (ie, beef up the series connections to keep them as cool as possible). Great pics, thanks for posting...
The only issues with battery holders that I have run into are...in order of most common problem first to least common problem last.
1. Keeping the cells from coming out of the holders
2. Populated with cells, a 20S2P battery holder pack is fairly heavy and glue joints get stressed. IE: reinforce glue joints as needed.
3. Cells that originally tested good going bad
4. Too much current draw per cell making them warm up.
5. insufficient current handling capacity in the battery holder.
6. accidentally popping a cell into a holder backwards

So let me explain a little...
1. My battery holders are packed to gether so there are no gaps between them. As a result I have never found a battery completely dislodged from it's holder, but I have a found quite a few that had shifted the 1/16" gap between cells in opposing holders. IE: they never actually come out of the holders, just shift a tiny bit while still maintaining full electrical contact. I usually push all the cells into each holder all the way down. If I pull out a pack after it has been in use for a month or so, just about all the cells have shifted a tiny bit. I wouldn't use battery holders without a way to keep them in the holders or the cells will slowly come out from bumps and vibrations.

2. I have had just about all the packs have a glue joint failure of one degree or other. Originally I used super glue between each 4S battery holder as the only thing holding the assembly together. That worked for several months, but eventually even though the packs can't move around or shift inside the battery box, I noticed that a few glue joints had failed. I reglued the joints back together with super glue and then added strips of 1/8" plywood to the tops and bottoms of all the packs that are glued down with flooring glue designed for tile or wood attached to concrete. It stays a little flexible, but when cured, stays stuck in place super well too. Gluing to concrete is not easy! This stuff is pretty much overkill. I have had no further glue joint failures. I think since my packs are long and narrow that they could flex down their lengths a little and that caused the glue failures between 4S modules. Now with the added plywood strips, they can no longer flex. IT has been a very long time since I had a joint failure.

3. Every used cell that originally went into my scooter was tested by me to make sure it met a minimum criteria for "good enough". However, over time, several cells have died and no longer take a charge. Out of 240 cells that number is less than 12. I randomly open the battery box and connect up a celllog to each balance connector. I quickly find which cells are going weak/died, pop them out and replace with other used/good cells. I forget exactly when I started using the used cells on my scooter, but I'm pretty close to a year now and that's all that have gone bad. I'd say not bad at all and I ride 10 miles to and from work every day on my scooter. If the weather is remotely decent, my car stays home.

4. I've never had a significant problem here, but then at most I'm running 2C continuous per cell. In the summer, I've ridden home in the early evening with it in the high 90's and just poured on the throttle every chance I got. Then when I got home I immediately checked the battery temperatures and the laptop cells were air ambient temp or 5-10F warmer. If a cell dies, that one cell gets a good bit warmer and enough to darken the color of my labels in each battery slot. I use a brother labler. The text on the label strip is created via heat application. That's why the labels get darker if they are exposed to heat. Still...that's 10 or so that are darker, but still readable. If I was really pulling hard (5C) on these cells or the battery holders, I would think I would see evidence of it everywhere. IE: black labels all over the place, distorted/melted plastic from heat due to over loading the solder legs on each holder or hot cells. The truth is, none of that has happened except darker labels where a cell died.

5. The spring clip touching the battery poles could handle 15 amps no problem, but the solder leg underneath is much smaller and can handle 6 or 7 amps at most. I overlapped them from each half of the pack and then added some solder to "beef them up more". It's never been a problem, but then I'm not exceeding the amperage of those solder legs either. I considered adding a small piece of copper wire to the legs, but it just hasn't been necessary at 2C per cell. If I were to use high discharge cells, first off they wouldn't be scrounged laptop cells and I would have built a properly welded pack, not used battery holders so it's a mute point, but yeah these battery holders wont hold up to 5C continuous discharge closed up inside a battery box without them showing evidence of over heating.

6. Early on, I labelled every battery slot for correct cell orientation. That was a lot of little tiny bits of label...40 per battery holder times 7 holders. It took hours to do. Despite my correctly labeling everything, somehow, I have randomly stuck a cell in backwards. I think it has happened 4 or 5 times. I usually notice immediately and so that cell has had a dead short with it's opposite side neighbor for a couple of seconds...no harm done. I check the voltage and everything is fine and the battery never warms up even a tiny bit. However, once and thank god...only once...I put a cell in backwards and didn't notice. I ran with that cell in backwards for several months before I did a random pack check-up. There was a single cell that was dead flat and the label behind it was black. The plastic battery holder showed no evidence of melting or distortion from heat. I'm guessing that cell got to around 150F and ran out of juice. The voltage of it's neighbor on the other side of the pack was much lower than the rest of the cells in the pack since it was carrying the load that two parallel cells should do, but was undamaged. I noted the position of that cell just in case it might fail later, but it never has. LOL! I guess that's proof of being able to run an LG laptop cell at 4C continuous.

On a side note about laptop cells. My old load tester could draw 24 amps. I would hook it up to my spare 20S2P battery holder and load test 40 cells at once. That works out to a little less than 5C per cell. The batteries would get pretty warm, but would still run the load for long enough to show around 2000mah or better of discharge assuming all "good enough" cells. This discharge rate was way over what I have ever run them at in my scooter, but they took it and kept going. My new halogen bulb load is about 4.6 amps per bulb and at 82 volts that means 4 bulbs in parallel or 18.3 amps continuous or 9 amps per cell. That's better than before so I'm not pushing the cells so hard, but they take it just fine with a little less heating up for the load test where I am weeding out the weak cells. I don't think I would want to run laptop cells at effectively 3C continuously, but for a load test it's OK.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 10, 2016 8:41 am

@ElectricGod Well, I think this is sort of ''battery business'' is not a bad idea.. After all, im actually one of these folks who hate to see how much ''good stuff'' is dumped into the dumpster and never seen again and probably buried under a huge pile of earth along with all sorts of other stuff for no possible way to recycle all of that later on..

So yeah, like with laptops, I think these cells could be called ''refurbished or used'' or whatever.. So some extra safety measures could be taken.. like in extreme case to hook up a bms to each cell or something (at least when charging)..

Though, Im not using a bms on my 6s pack as I have balance charger.. but If I went with bulk charging I would probably want to make sure that none of the cells are dead or something before charging them..

--

About that safety measures..

Can you explain what type of heat labeling are you using? I did not really catch the idea how exactly are you doing that.. though it seems like a really clever idea to see which cells are heating up the most..

Ok will update this post later more ;)

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 13, 2016 8:51 am

Okami wrote:@ElectricGod Well, I think this is sort of ''battery business'' is not a bad idea.. After all, im actually one of these folks who hate to see how much ''good stuff'' is dumped into the dumpster and never seen again and probably buried under a huge pile of earth along with all sorts of other stuff for no possible way to recycle all of that later on..

So yeah, like with laptops, I think these cells could be called ''refurbished or used'' or whatever.. So some extra safety measures could be taken.. like in extreme case to hook up a bms to each cell or something (at least when charging)..

Though, Im not using a bms on my 6s pack as I have balance charger.. but If I went with bulk charging I would probably want to make sure that none of the cells are dead or something before charging them..

--

About that safety measures..

Can you explain what type of heat labeling are you using? I did not really catch the idea how exactly are you doing that.. though it seems like a really clever idea to see which cells are heating up the most..

Ok will update this post later more ;)
I ran for several months without a BMS on these batteries. I just checked them every couple of times I charged to see if any cells were getting out of balance. There were small variations like 3.5-3.7 volts when mostly discharged. Any cell that was truly weak would be way off from the average voltage of all the rest of the cells. That happened a couple of times, so I popped out the weak cells and replaced them, but my testing for bad cells was pretty thorough anyway so the few weak cells I found were probably just early detection's of cells that were going to die anyway.

If I had access to hundreds of packs on a regular basis, it could be worth it to salvage and sell cells, but for the few packs I see at random...not so much. I have a couple hundred cells right now, but that's not enough for more than a one time sale or maybe two. I collected them a few here and there and then a couple of times I hit a boon of 10-20 packs. Laptop cells don't have the discharge rates of real EV grade cells so you need more in parallel just for that reason...even though I know for a fact that they handle 2-3C pretty well. Then there is the fact that they are going to have limited capacity of around 2000mah to be "good enough" which means more parallel cells to get decent capacity. None of this makes them a bad idea, just that you don't buy 50 cells and expect them to perform like EV cells.

Heat labels...lol...Any of your typical labelers use cartridges like below. The material on the plastic label is heat sensitive. It's how the labeler puts text on the label tape. Look at a typical receipt from the grocery store or gas pump.. You can run your fingernail across the paper and mark it. It's the same stuff on the label tape...just more resilient.

https://www.amazon.com/Brother-P-touch- ... label+tape
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 14, 2016 5:20 pm

@ElectricGod mh I think I will have to look something similar (labels) in our local shops..

''In the early days'' of my first battery build, I considered putting one of these in my battery pack or close to a motor:

http://www.ebay.com/itm/10-Pcs-Bimetal- ... SwBPNXR4d5

The one shown is normally closed but I imagine I could have used normally open and if temp is reached, the switch turns on a led or alarms a buzzer, so that you can know while riding that the temp is high..

--

Though, I somehow scrapped the idea as for the motor I got a fan and batteries seemed to be holding quite okay but that is another story.
----

About this ''business perspective''. I see it as viable, if there is a regular source of batteries (like from a pc repair shop) and the process is quite automatized. The main clients would probably be people who are willing to wait and are not entirely ready to buy ''brand new quality cell packs'' but are willing to take the risk and use what is available.

So yeah, the steps would probably be:
1) get the batteries
2) take them apart
3) test / bin the cells
4) test again the ''good'' ones
5) make the pack
6) some more real life tests
7) ready to sell / use

--
The charging part is probably the most painful one, as it involves more investment in charger power or such. Though, have not started this operation yet, so hard to speculate.. another cool option would be to repair or ''restore'' the laptop packs, I've tried but I think the bms went blank already with the original cells and also it was hard to make the plastic look as it originally was, as I had to pry it apart and some cells did not fit ideally into their places.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 15, 2016 8:57 pm

Okami wrote:@ElectricGod mh I think I will have to look something similar (labels) in our local shops..

''In the early days'' of my first battery build, I considered putting one of these in my battery pack or close to a motor:

http://www.ebay.com/itm/10-Pcs-Bimetal- ... SwBPNXR4d5

The one shown is normally closed but I imagine I could have used normally open and if temp is reached, the switch turns on a led or alarms a buzzer, so that you can know while riding that the temp is high..

--

Though, I somehow scrapped the idea as for the motor I got a fan and batteries seemed to be holding quite okay but that is another story.
----

About this ''business perspective''. I see it as viable, if there is a regular source of batteries (like from a pc repair shop) and the process is quite automatized. The main clients would probably be people who are willing to wait and are not entirely ready to buy ''brand new quality cell packs'' but are willing to take the risk and use what is available.

So yeah, the steps would probably be:
1) get the batteries
2) take them apart
3) test / bin the cells
4) test again the ''good'' ones
5) make the pack
6) some more real life tests
7) ready to sell / use

--
The charging part is probably the most painful one, as it involves more investment in charger power or such. Though, have not started this operation yet, so hard to speculate.. another cool option would be to repair or ''restore'' the laptop packs, I've tried but I think the bms went blank already with the original cells and also it was hard to make the plastic look as it originally was, as I had to pry it apart and some cells did not fit ideally into their places.
I had similar idea with repairing laptop packs. Name brand packs are made of 2600mah cells. WHY WHY WHY? At 1 or 2C there are several high capacity cell options. 3600mah is doable at those low discharge rates. An extra 1000mah is a huge increase for a laptop battery pack, but no one uses them. Anyway, that was my idea...when my laptop pack gets weak and it's run time is getting down in the 20 minute range, get whatever is the maximum capacity cells that exist and replace the worn out 2600 mah cells. The BMS in the laptop pack works just like it does in your EV. Disconnect the positive end first and work your way down to the negative end of the pack. Isolate and cover every bare wire end as you go. When reconnecting everything, start at the negative end of the battery pack and work your way to the positive end. Going backwards or connecting a balance wire in the middle first will possibly fry the BMS. I can't speak for all BMS in every laptop battery pack, but some of them know what capacity the cells are supposed to be and then measure against that baseline to determine current charge level and remaining capacity. It's possible, that replacing the 2600mah cells with 3600 mah cells will confuse the BMS controller chip. I would think that at worst, the remaining capacity meter in the laptop software would be wrong. BUT...batteries get weaker as they age so my suspicion is that the BMS is pretty dumb and the laptop just monitors pack output voltage to determine remaining capacity on most packs.

Those little temp sensors are uber cheap. If you wired them strategically so they were evenly spaced in the pack, 4 or 5 of them would cover battery temp really well. Wire them all into OR gates and then the outputs of all the OR gates to the buzzer. You don't really care that sensor A or sensor F detected too much heat, you just want to know that it happened and to be warned about it. You could also add an LED to each NO sensor that turns on when the sensor is closed. As long as that sensor is hot, the LED is lit so you know approximately where the overloaded cell is. When the sensor is too cool to close, the LED is off and drawing no current. However, running the cells within their capacity limits just means planning for more cells in parallel in which case they don't get hot. Mine never get hot.

Charging tons of cells is uber easy. You just need a 6S charger and then parallel the cells as much as you can. I have a 6S6P battery holder I built just for this purpose. I can charge up to 36 cells simultaneously. As long as I'm charging 6S, the charger doesn't care that it has 2 or 200 cells in parallel as long as each channel has at least one cell in it. With multiple cells in parallel, the more charged cells in a channel will be drawn down by the less charged cells. The overall effect is the voltage of any particular channel will be the average of all voltages of all the cells on that channel. When charging any balance charger is going to try to get each channel to the same voltage. I have found that if I have a channel that is taking much longer than another channel to reach fully charged, that it probably has a few cells that are more deeply discharged than the rest. AS a result it takes longer to get that channel up to fully charged. Since I have multiple cells in parallel, I pop a cell out of a lower voltage channel and swap it with a channel that is at a higher voltage. I keep doing this until all channels show similar voltages, then charging proceeds much more quickly. I have a spare 20S2P battery holder, if I had a spare 20S BMS, I could use it to balance charge 40 cells in that battery holder simultaneously and use my 82 volt scooter charger to run it. I have 3 cheap 12S BMS just sitting there collecting dust. I could build a 12S12P battery holder and balance charge 144 cells simultaneously. I have several 12S chargers that could serve this purpose. IE: Charging tons of batteries at the same time is pretty easy to do. The harder part is load testing them. At most, you want 2 cells in parallel so that you only have two cells to restest on a weak channel. More cells in parallel on a load test means more possible variables for the weak channel.

If I had the regular stream of battery packs coming in all the time, building packs from used cells or just selling cells would be doable. As an option, selling battery holders for them would supplement the business. The sales page for this little operation would need to be very realistic and brutally honest.

IE:
Warning: You are buying used laptop cells. Do NOT use these cells for more than 2-3C discharge rates!!!
Warning: You are buying used laptop cells. Do NOT expect them to provide the capacity of brand new laptop cells.
Warning: You are buying used laptop cells. Do NOT expect them to perform like EV grade cells.
Warming: You are buying used laptop cells!!! Did you get the message yet? THESE ARE USED LAPTOP CELLS THAT HAVE BEEN TESTED TO BE REASONABLY GOOD. Use with those expectations in mind!!!
Warning: All sales are final. We have tested the cells you have received to confirm they have at least 2000mah of capacity at 3C discharge rates. Some cells may have more capacity than others.
Warning: All sales are final. There are no warranties. These ARE USED LAPTOP CELLS!!!! Expect the possibility that a cell may fail later on so buy some spares.
Warning: You are buying used laptop cells. PLEASE, PLEASE, PLEASE, PLEASE, don't weld them into a pack, use battery holders so that if/when a cell goes bad you can replace it easily.

I can see people bitching that the cells don't have the capacity they want or that they can't pull 20 amps from a 2C cell or that a cell went bad. Hence the above warnings. LOL!
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
Ianhill
100 kW
100 kW
Posts: 1129
Joined: Sep 25, 2015 5:55 pm
Location: Wales UK

Re: My excellent adventure through some battery builds

Post by Ianhill » Dec 15, 2016 9:39 pm

I see alot of auctions in the UK on used laptop cells they seem to be getting around £100 per 50units in unknown state and they are non returnable, I like the warnings its like tom hanks in turner and hooch this is not your room.

I like the idea of a led visual indicator that tells of high temps on motor, battery or controller, I Think the battery area is where the improvement can be made in discharge temps by maintaining the 70F with a closed vent system below this then vents open and becomes an air cooled pack when temps rise, but that what I need in cooler weather in the summer I need flat out cooling

Maybe 3 dual colour leds next to name tags motor battery and controller, the led is green when turned on and in temp range then red when theres a fault in circuit or temp out of range I'm intreged.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 15, 2016 10:57 pm

Ianhill wrote:I see alot of auctions in the UK on used laptop cells they seem to be getting around £100 per 50units in unknown state and they are non returnable, I like the warnings its like tom hanks in turner and hooch this is not your room.

I like the idea of a led visual indicator that tells of high temps on motor, battery or controller, I Think the battery area is where the improvement can be made in discharge temps by maintaining the 70F with a closed vent system below this then vents open and becomes an air cooled pack when temps rise, but that what I need in cooler weather in the summer I need flat out cooling

Maybe 3 dual colour leds next to name tags motor battery and controller, the led is green when turned on and in temp range then red when theres a fault in circuit or temp out of range I'm intreged.
I assume that's for complete laptop battery packs? If so, then you are buying cells in an unknown state. I would test the individual cells and eliminate the dead/weak cells so that wouldn't be an issue. If I had to buy used laptop packs, they would have to be uber cheap so that when I recovered whatever good cells I could find in them, I would still make a profit on the remaining good ones. It is quite common that any given pack is some good cells and some weak/dead ones. It's already a crap shoot that there are "good enough" cells in the used packs so paying for them isn't a great idea.

The LED indicator idea could be two LED colors since logic gates are digital. IE: green for under temp and red for over temp. The temperature sensor has a switch on temp that is defined when the sensor is made. IF you had sensors that turned on at 40C and another at 50C and another at 60C, then you could detect 3 different temps and activate or deactivate something based on those temps. Those sensors are quite small so several of them together would take up very little room.

Another idea, would be to use a real resistive temperature sensor and some op-amps. The voltage drop across the temp sensor would trigger the op-amp based on a reference voltage on its other input. Several op-amps in parallel to the temp sensor, but with different reference voltages would create multiple cross over points using a single temp sensor and each one would be adjustable. 3 LEDs or low, medium and high on a fan are all options.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

User avatar
Ianhill
100 kW
100 kW
Posts: 1129
Joined: Sep 25, 2015 5:55 pm
Location: Wales UK

Re: My excellent adventure through some battery builds

Post by Ianhill » Dec 15, 2016 11:30 pm

I bought an I charger so I could balance charge 10 at a time then do an IR test and discharge singularly at 2a to 3v and see just how much capacity they had left so I had two values on the cell then IR and mah then I grouped them and made a pack after a few months of riding my mach 1 razor and three times I had to repair pack I was 27cells worse off and I was out of cells to replace i did not fancy paying out for more so I went the lipo route because it was very time consuming it would take a week or two to strip down the cells test them all arrange and build a pack and when they were going to cost its not so fun as I was going through 18650 like fun.
My biggest mistake was a naked pack and grossly overloaded laptop cells around 4c so I needed more like 18p but I suppose i was and still am learning, I went the cell fuse route made up a busbar and fused each cell to the busbar with soldered fuse wire but the problem I was getting was at 3.2v or so being so overloaded the 9p string I had would drop a cell to dead short making my 10s9p into a 9s9p effectively with a massive hot spot dead shorting, that happened three times so I vowed to keep my parallels as little as possible that's why I'm doing the 20s1p now so if a cell drops low I will catch it on bms no problems.
_IMG_000000_000000.jpg
10s 9p razor e300 my1020 motor

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 16, 2016 2:54 am

Ianhill wrote:I bought an I charger so I could balance charge 10 at a time then do an IR test and discharge singularly at 2a to 3v and see just how much capacity they had left so I had two values on the cell then IR and mah then I grouped them and made a pack after a few months of riding my mach 1 razor and three times I had to repair pack I was 27cells worse off and I was out of cells to replace i did not fancy paying out for more so I went the lipo route because it was very time consuming it would take a week or two to strip down the cells test them all arrange and build a pack and when they were going to cost its not so fun as I was going through 18650 like fun.
My biggest mistake was a naked pack and grossly overloaded laptop cells around 4c so I needed more like 18p but I suppose i was and still am learning, I went the cell fuse route made up a busbar and fused each cell to the busbar with soldered fuse wire but the problem I was getting was at 3.2v or so being so overloaded the 9p string I had would drop a cell to dead short making my 10s9p into a 9s9p effectively with a massive hot spot dead shorting, that happened three times so I vowed to keep my parallels as little as possible that's why I'm doing the 20s1p now so if a cell drops low I will catch it on bms no problems.
_IMG_000000_000000.jpg
Yes parallel cells runs the risk of a single dead cell killing all the rest of the cells it is in parallel with. That's what happened to the 20S LIPO packs I built. As a result I have a dead cell in all 3 packs. Of course if you were running laptop cells at 4C, well that is a for sure way to kill them. I load test them at 3C and many of them get pretty warm, but at most I do 2C continuous. I think your pack would have survived if the discharge rate per cell was more reasonable.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 17, 2016 4:11 am

Ianhill wrote:I see alot of auctions in the UK on used laptop cells they seem to be getting around per 50units in unknown state and they are non returnable, I like the warnings its like tom hanks in turner and hooch this is not your room
I think these ebay listings for such price is not that great! All you might get ''extra'' is the work to tear apart the casing and have tabs and probably balance wires already on the cells.

Other than that, at least for EU, folks, there is a site https://eu.nkon.nl

There you can get cells for close to 2 eur a piece (new and also possibly with higher discharge!) The only ''con'' might be the fact that there's still a shipping cost and that tabbing each cell (by their company) costs extra 0.4 eur, so okay, 3 eur a piece or so, depending on pack size for already tabbed, new, known cells.

These are 2 examples:

https://eu.nkon.nl/rechargeable/18650-s ... 50-ch.html (+10 piecies for €2.18,
+50 piecies for €2.05) - Though, brand is unknown to me and im not that ready to search up any tests

https://eu.nkon.nl/rechargeable/18650-s ... 0-26f.html (+10 piecies for €2.65,
+50 piecies for €2.55) - about the same specs but Samsung cell.

With shipping included for these samsung, I got a figure of 3.19 eur a piece (with tabs already spot-welded)=50 cells,
if it goes without a tabs (like for a sleds), then the final price is 2.79 eur a piece (when buying 50)
-----
So yeah, paying close to 2 eur is a bit risky.. I think, especially considering someone has used these cells for a while..+ all the extra work tearing them apart, especially, if you plan to use sleds anyways)

------
-----

@ElectricGod

Woah lots of great info to process and to reply to, I will try to do this shortly now, as Im not having enough time now to craft each of the reply to each of the segment

I've also thought / got the idea that it is somewhat stupid laptop battery manufacturers have not decided to use 3500mah cells instead of 2600mah ones.

As for the calibration, I think some laptops have built in this function in the bios or even on the OS level, to calibrate the battery pack.. you just have to fully charge it and then completely deplete it..

----

The charging / discharging part of used laptop batteries

@ElectricGod So in which part do you actually measure the capacity the pack can output?

Somehow in my mind it seems ideal, if it were possible to determine each cell's capacity in the early run,. Though, from what you said, it now looks like a better option (more practical) is just to make the pack with the ''strong'' cells and then you measure the capacity of the whole pack the way it is, when you have ''rejected'' the weak cells.

It seems, that with ''fiddling with'' and replacing parallel pack batteries, till they reach optimium charge for all groups, also sorts them out and makes the pack of consistent cells in each group. After that it is possible to measure the capacity of the whole pack (when charging it), althrough you still dont know the capacity of each of the group's pack, so I assume they just need to balance each other out.

So okay, I think I actually answered to my own question. That the capacity can be tested at the end and not at the beginning of the ''trials''.
----

About the temp sensor thing. Yes, the arduino way + thermal probe is probably the most sophisticated way. This way you can select different temp range and decide what action is done at each ''step''.

Op-amps might also work though I would personally go with arduino as for it might be bit easier than to build an ''old-school'' circuit out of simple components.

User avatar
ElectricGod
100 kW
100 kW
Posts: 1444
Joined: Nov 01, 2015 5:24 pm

Re: My excellent adventure through some battery builds

Post by ElectricGod » Dec 19, 2016 3:27 am

Okami wrote:
Ianhill wrote:I see alot of auctions in the UK on used laptop cells they seem to be getting around per 50units in unknown state and they are non returnable, I like the warnings its like tom hanks in turner and hooch this is not your room
I think these ebay listings for such price is not that great! All you might get ''extra'' is the work to tear apart the casing and have tabs and probably balance wires already on the cells.

Other than that, at least for EU, folks, there is a site https://eu.nkon.nl

There you can get cells for close to 2 eur a piece (new and also possibly with higher discharge!) The only ''con'' might be the fact that there's still a shipping cost and that tabbing each cell (by their company) costs extra 0.4 eur, so okay, 3 eur a piece or so, depending on pack size for already tabbed, new, known cells.

These are 2 examples:

https://eu.nkon.nl/rechargeable/18650-s ... 50-ch.html (+10 piecies for €2.18,
+50 piecies for €2.05) - Though, brand is unknown to me and im not that ready to search up any tests

https://eu.nkon.nl/rechargeable/18650-s ... 0-26f.html (+10 piecies for €2.65,
+50 piecies for €2.55) - about the same specs but Samsung cell.

With shipping included for these samsung, I got a figure of 3.19 eur a piece (with tabs already spot-welded)=50 cells,
if it goes without a tabs (like for a sleds), then the final price is 2.79 eur a piece (when buying 50)
-----
So yeah, paying close to 2 eur is a bit risky.. I think, especially considering someone has used these cells for a while..+ all the extra work tearing them apart, especially, if you plan to use sleds anyways)

------
-----

@ElectricGod

Woah lots of great info to process and to reply to, I will try to do this shortly now, as Im not having enough time now to craft each of the reply to each of the segment

I've also thought / got the idea that it is somewhat stupid laptop battery manufacturers have not decided to use 3500mah cells instead of 2600mah ones.

As for the calibration, I think some laptops have built in this function in the bios or even on the OS level, to calibrate the battery pack.. you just have to fully charge it and then completely deplete it..

----

The charging / discharging part of used laptop batteries

@ElectricGod So in which part do you actually measure the capacity the pack can output?

Somehow in my mind it seems ideal, if it were possible to determine each cell's capacity in the early run,. Though, from what you said, it now looks like a better option (more practical) is just to make the pack with the ''strong'' cells and then you measure the capacity of the whole pack the way it is, when you have ''rejected'' the weak cells.

It seems, that with ''fiddling with'' and replacing parallel pack batteries, till they reach optimium charge for all groups, also sorts them out and makes the pack of consistent cells in each group. After that it is possible to measure the capacity of the whole pack (when charging it), althrough you still dont know the capacity of each of the group's pack, so I assume they just need to balance each other out.

So okay, I think I actually answered to my own question. That the capacity can be tested at the end and not at the beginning of the ''trials''.
----

About the temp sensor thing. Yes, the arduino way + thermal probe is probably the most sophisticated way. This way you can select different temp range and decide what action is done at each ''step''.

Op-amps might also work though I would personally go with arduino as for it might be bit easier than to build an ''old-school'' circuit out of simple components.
If all the cells were discharged to 3 volts and then I fully charged them to 4.1 volts, well then I could test charge capacity. What's the point? The capacity you really care about is under load so that's what I
measure.


As far as "optimal" is concerned...LOL...that's what the load testing is about...to find the weak cells. I eliminate the less than "good enough" cells in load testing, the rest go in battery holders to power EV stuff. Yes...it's true there are some cells that are stronger than others, but you are not going to get away with 100% of the cells being 2600 mah. If I load test and weed out anything less than 2000mah, then as long as my expectations are 2000mah, then they are already optimal. Since I'm running 20S to get 30% of the used cells at 2600mah would be fairly miraculous. So lets be more realistic and say 15% are nearly new cells and the rest are something less than that. It wont really matter if all the best cells are in a single pack or distributed among all the packs. After all they ARE used laptop cells. You make the best of what you have.
ElectricGod...it's a joke...don't take things so seriously folks!

XB-502 Moped:
https://endless-sphere.com/forums/viewt ... 0#p1222730

Currie scooter:
https://endless-sphere.com/forums/viewt ... 7#p1227407

Okami
100 mW
100 mW
Posts: 45
Joined: Jun 02, 2016 3:00 am

Re: My excellent adventure through some battery builds

Post by Okami » Dec 19, 2016 3:37 am

Ok - so I got a quick reply for you, perhaps you will get the chance to reply soon!

A very clever way to measure the cells! Indeed. I think you have changed my understanding now on how to check cells for high load use..

Ok, some conclusions I got when I tried to discharge my Panasonic CGR-CG pack 4s2P pack (~16v, ~2250mah per cell, when new and 2-3c discharge - datasheet data)

I got to about 3.1 -3.2 volts under load per cell (~25w) and the cells became considerably weaker, if they were pulling the mentioned 25w of load before, then now they pulled only 21-22w of power)

When I disconnected the load.. the charge went up to even 13.43v
I forgot to monitor the total voltage when under load but it was around 12v..

Anyways - I concluded that the performance is quite bad when the cell is below 3.3 under load.. (same conclusions can be found on the net, of course)
And the load voltage is about 3.5v when the cell should be stopped discharging probably

The capacity I managed to sqeeze is out was about 2.650 mah, out of 60-70% full pack when I started monitoring the discharge.

---

So - anyways when I connected to the balance charger, the different between highest and lowest cell was about 0.08v I believe, so almost 0.1v (~3.44 and 3.36)

---

@ElectricGod, though, Im writing all of this to ask you - so what exactly happens when you have different capacity parallel groups? This was the thing why I wanted to test each cell, so that I can arrange them in a manner, the capacity levels are somewhat equal..

How will the pack act when one group has 2ah, another one has 1.8ah and the last one, for example, 2.2ah..

--

I actually wanted to make new topic about mixing high discharge cells with low discharge cells (like laptop batteries).. and what would be the effects of doing so, it looks like there are ppl who have even mixed lipo and li-ion, li-ion and lifepo, and even Pb and lifepo.

So yes, another thing im looking for is, what I should be beware of besides monitoring the cell voltage when connecting two different packs in parallel (same voltage of course but perhaps a little weaker capacity). I do believe that the weaker pack will probably drain faster and will have to be disconnected, while the bigger one will probably have more juise and otherwise.. just supplement the weaker pack, when it has reached its low point.

Post Reply