Building battery from MAKITA Konion PACK

OK,

For the step prior to make your pack, you need to equalize ( balance ) all cells to the same level.

As well, all packs and cells have not the same state of charge so you need to bring all them to the same BEFORE to parallel your cell in the build

What you need is to ensure that all the cells you will parallel to equalize ( balance) need to have max 300mV from the lowest to the highest cells BEFORE TO PARALLEL

So since many cells will have higher voltage difference than 300mV, let say one of the cell have 3.4 and the other have 4.0.. so you can not parallel them from now. just follow the next step and you will be able then to parallel all them

What I suggest is to make few different groups of voltage ranges. Let say you make a group for the cell from 2.5 to 2.8V, another for the cell at 2.8 to 3.1V, another group with cell of 3.1 to 3.4V etc.. up to 4.2V ( this step is not to make your pack but just for the equalization step!)

So instead of charging all cell individually, make these groups with all these range of cells voltage and than you can charge these group instead of charging each cells, that will save you a lot of time! so proceding to a charge of 5 or 6 groups is faster than charging 100 or 200 individual cells as well! :wink:

After all these groups are charged to 4.1V or 4.2V, the final equalizing step would be to parallel all group cells together.

One of the best way I found to parallel cells fast if to use aluminum foil sheet and some smooth packing material. What you do if to place all cells in a sandwich with one top and one bottom layer of aluminum foil sheet. It work perfectly!

But since all cells don’t have a the perfect exact lengh form each other, to ensure that all cell have a good electrical contact to the sheet you can use some soft bubble or foam sheet material to apply pressure on the foil on both top and bottom, than to use rigid material to apply uniform pressure on that sandwich. I will try to post a picture that.
( dont forget to prepare the cell nickel tab to have a good electrical contact for each cells! ( follow the way i suggested in teh thread when i flip the tab nickel sheet of the cell ends with some pliers



Doc
 
Ok thanks guys I really appreciate this but Doc, you're fine explanation puts another little doubt in my mind. You say to group the cells together with voltages that are 300 mv of each other which makes sense. 2.5 to 2.8 and 2.9 to 3.2 and so on. But in your very first post, you say to basically toss any cells below 3V. I hope I can use the cells that show a voltage of less than 3V because after opening 14 packs, I have 12 that were dead (I expected more so that's good) and 58 that are between 2.3 and 2.8 volts. So can I use these cells? My thinking is that if they show a voltage and charge up to 4.1V with no problems than they're good. But of course to know exactly if EVERY cell takes a charge and is good to go, than this brings me back to charging individually :?

Thanks for your patience, I'm old and presently sick with the flu so brain working at an even lower level than usual.
 
That is one smart way to equalize cells, Doc. Fantastique!
otherDoc
 
mistercrash said:
Ok thanks guys I really appreciate this but Doc, you're fine explanation puts another little doubt in my mind. You say to group the cells together with voltages that are 300 mv of each other which makes sense. 2.5 to 2.8 and 2.9 to 3.2 and so on. But in your very first post, you say to basically toss any cells below 3V. I hope I can use the cells that show a voltage of less than 3V because after opening 14 packs, I have 12 that were dead (I expected more so that's good) and 58 that are between 2.3 and 2.8 volts. So can I use these cells? My thinking is that if they show a voltage and charge up to 4.1V with no problems than they're good. But of course to know exactly if EVERY cell takes a charge and is good to go, than this brings me back to charging individually :?

Thanks for your patience, I'm old and presently sick with the flu so brain working at an even lower level than usual.


the cells spec say use between 2.5 to 4.2V max voltage range.

Usually we recommand any cell that charge up to 4.2V to have a low cut off at 3.0V, but 2.5 is still good.

Yes you can use the cells that are 2.5V and above.

One of the thing you could do is to equalize all them ( that are above 2.5v when you receive them), reject all cells under 2.5V.. or keep them for boost pack, or other use.

Once they are equalized ( at least 12h later if you use the sandwich parallel connection method)
it's time to remove them from the sandwich conections (unparallel them), leave them alone for another 12h and then, measure their individual voltages. The one that are out of the average voltage must be rejected, these cells are probably weak and selfdischarge abnormally.

When i mean out of the average i mean if the amost of the cells are at 4.15V after the total 24h hour and that you have some cells that are at 3.8, 3.9 or 4.0V, these cells must be rejected. Than you can be sure that all the rest of the cells are good for your ebike use.

Please note that this particular method is not followed by the average of person that use the konions and they are still satisfied of the result. But if you wnat to improove the performance of your pack, then, following the most of the mentionned steps will certainly help!

I tested ALOT of Konion cell RI and Capacity and i know that if you follow this method, it will give you the results really similar as if you would measure all capacity and Ri of cells and match them. These cells are robust!.. the only thing they dont like is heat above 50 celsius that can reduce their life, nothing more.

In fact, Ri, Capacity, and age of lithium cells are all in relation. I wrote alot of documentation and test about that in E-S.

Normally a cell that have high RI ( not good) will self discharge faster and have a low capacity.

But sometime it happen rarely that you can have a cell with very high capacity but not so good RI... but it is rare and is more related to the irregularity of chemistry and cell fabrication process.

Also the if you use a method to measure their RI , the result you will get might be different than the one of another E-S member will have. The RI is affected by ALOT of parameters!! so if you wnat to use an absolute RI testing method or device, it must be described cristal clear!!.
Temperature of 2 or 3 degree variatin will make a significant change in RI, and also the time you allow them to stabilize after their last charge, as wel as the method you use. AC impedense DC resistance are two similar but not identical method so if you want to compare your results, IT MUST be done in the same process and using the same calibrtated device.

If you measure the RI to perfectly match the cells, it doesn't matter the method you use , but it must be constant and followed precisely for all cells and under the same condition.

The past results i used for my 432 cell pack was done with my modified RC megapower 960-SR charger discharger that was using the Kelvin 4 wires method.

See teh thread: http://endless-sphere.com/forums/viewtopic.php?f=14&t=2764

I joined two really important file you can use as reference that was made with over 170 Konion cell pair.

Yes you can leave the cells paired if they are in the pack ! Dont forget otherwise you will double your work time! Cells in pair have the same age, same state of charge and are already matched together!

I hope this help! :wink:

Doc
 

Attachments

  • Ri and capacity distribution Konion 2007.JPG
    Ri and capacity distribution Konion 2007.JPG
    140.5 KB · Views: 3,303
  • Data_of_all_C_+_Ri__tests.xls
    49 KB · Views: 88
file.php

What material is that strip? Copper? What is the thickness? What is the width? Where do you buy it? Do you cut it to size? Do you cut it with snips?

If building a large battery for a electric vehicle, I know that we need to know how many Amps will be used to calculate the size of the metal strip used to connect the individual cells. Lets just figure slow electric bicycles and scooters that are mostly just dreams in my mind.

Think mcmaster.com might be expensive? They got a lot of information on copper and different alloys.
Electrical Conductivity

Copper alloys are known for their excellent electrical conductivity. Ratings for electrical conductivity use Alloy 101 as the standard. Note: SAE 841 is not rated.
aboutbox_electrical_conductivity.png
 
marty said:
file.php

What material is that strip? Copper? What is the thickness? What is the width? Where do you buy it? Do you cut it to size? Do you cut it with snips?

If building a large battery for a electric vehicle, I know that we need to know how many Amps will be used to calculate the size of the metal strip used to connect the individual cells. Lets just figure slow electric bicycles and scooters that are mostly just dreams in my mind.

Think mcmaster.com might be expensive? They got a lot of information on copper and different alloys.
Electrical Conductivity

Copper alloys are known for their excellent electrical conductivity. Ratings for electrical conductivity use Alloy 101 as the standard. Note: SAE 841 is not rated.
aboutbox_electrical_conductivity.png


For that i use BCAE1.com car audio website wich have many great java applet to calculate your data.

Yes i used copper. I got from surplus material at work. But i know that some house are made with copper top so try contacting some guy working in the construction domain that are specialized in house top cover.

but you can also get some on ebay as well


http://www.bcae1.com/wire.htm

to find the number of square mils you need with copper,
 
Thank you so much Doc for all the time to explain everything in detail, it's all very clear to me now. For bus bars, I just had a 4 foot left over piece of 1/2" copper plumbing pipe. I flattened it with a hammer, then grinded the edges on the belt sander until the two layers separated. This gave me two 4 foot strips that were a bit more than a half inch wide. I used snips to cut them down the middle to get 4 strips that are a little over a quarter inch wide. This gave me 17 bus bars. I will need to go get another 4 foot of the same pipe at Home Depot to make the rest of the bus bars I need. A little more than 1/4" wide by a little less than 2 mm thick. Should be more than enough.
 
Doctorbass said:
marty said:
file.php

What material is that strip? Copper? What is the thickness? What is the width? Where do you buy it? Do you cut it to size? Do you cut it with snips?

If building a large battery for a electric vehicle, I know that we need to know how many Amps will be used to calculate the size of the metal strip used to connect the individual cells. Lets just figure slow electric bicycles and scooters that are mostly just dreams in my mind.

Think mcmaster.com might be expensive? They got a lot of information on copper and different alloys.
Electrical Conductivity

Copper alloys are known for their excellent electrical conductivity. Ratings for electrical conductivity use Alloy 101 as the standard. Note: SAE 841 is not rated.
aboutbox_electrical_conductivity.png


For that i use BCAE1.com car audio website wich have many great java applet to calculate your data.

Yes i used copper. I got from surplus material at work. But i know that some house are made with copper top so try contacting some guy working in the construction domain that are specialized in house top cover.

but you can also get some on ebay as well


http://www.bcae1.com/wire.htm

to find the number of square mils you need with copper,
Called a roofing supply place. They got 14 inch wide copper flashing. Cost is $12.91 per foot. They call it 16 once. Thickness is .0216 inch. Also called 22 Mils (a Mil is 1/1000 of an inch)

If I cut battery cell connector strips 3/8 inch wide. 3/8 inch is .375 inch or 375 Mils

22 Mil X 375 Mil Copper strip is = 8250 Mils

I HAVE STOPPED UNDERSTANDING WHAT I AM TYPING!!!!!!

As per calculator at http://www.bcae1.com/wire.htm
8250 square mils is 10504 circular mils.

As per chart at http://www.bcae1.com/wire.htm
10 Gauge wire is 10384 mils which is pretty close to what I think my copper strip might be, 10504 mils.

If I am figuring this correctly a .0216 inch X .375 inch copper strip is about equal to a 10 Gauge wire which I think is more then big enough.

IS MY THINKING CORRECT? I AM NOT SO GOOD WITH MATH AND ELECTRICITY.

Marty goes down to the basement to pound on a 10G solid wire with a hammer and anvil to see what size it is when it is all flattened out.
 
Doctorbass said:
OK,

What I suggest is to make few different groups of voltage ranges. Let say you make a group for the cell from 2.5 to 2.8V, another for the cell at 2.8 to 3.1V, another group with cell of 3.1 to 3.4V etc.. up to 4.2V ( this step is not to make your pack but just for the equalization step!)

So instead of charging all cell individually, make these groups with all these range of cells voltage and than you can charge these group instead of charging each cells, that will save you a lot of time! so proceding to a charge of 5 or 6 groups is faster than charging 100 or 200 individual cells as well! :wink:

After all these groups are charged to 4.1V or 4.2V, the final equalizing step would be to parallel all group cells together.

One of the best way I found to parallel cells fast if to use aluminum foil sheet and some smooth packing material. What you do if to place all cells in a sandwich with one top and one bottom layer of aluminum foil sheet. It work perfectly!




Doc

Doc what do you think about bulk charging these batteries in series before creating parallel strings? I was think about purchasing some cheap 18650 battery holders and stringing them together in 16S (4 x 4 bank 18650 battery holders). They are available on fleebay for a couple bucks each. I have a 48 volt 500 watt Meanwell that is variable up to 60 volts. 16 x 3.7 is 59.2 volts, this would allow 16 batteries to charge simultaneously. Would this work or is there something I'm missing?
 


Your calculations look good and 10 gauge equivalent is ok.
 
Degull said:
Doc what do you think about bulk charging these batteries in series before creating parallel strings? I was think about purchasing some cheap 18650 battery holders and stringing them together in 16S (4 x 4 bank 18650 battery holders). They are available on fleebay for a couple bucks each. I have a 48 volt 500 watt Meanwell that is variable up to 60 volts. 16 x 3.7 is 59.2 volts, this would allow 16 batteries to charge simultaneously. Would this work or is there something I'm missing?

No !!

Absolutly NOT :?

serie connections DONT balance them.. it's PARALLEL connections that balance them :) and you need to balance them BEFORE connecting them in parallel groups with a tolerance under 300mV max between the low and the high cell. Once they are paralleld for 12 hours, than you can be sure they are balanced perfectly. THAN you can make your parallel groups and then connect them in serie.


Konion autobalance : YES
Konion autobalance whatever the voltage difference: NO so you can't connect them in serie if they are not equalized before !


bulk charging if your pack have not beed equalized with in a certain level will make the high cell to overcharge.

Please re read what i suggested with the sandwich suggestion :wink:

Doc
 
Here is an exemple of when i had alot of A123 cells to parallel .

It's an easy way to paralell all your cells fast at low cost. ( use this method for all cells grouped with in 300mV difference between the highest and the lowest voltage cell) only.. otherwise too much charge current will flow to the lowest ! and ytou will stress that low cell.

I call this my Doc's perfect cell equalizer ! :mrgreen:

You can add some weight with a good distribution over the surface to ensure you get a good pressure on all cell tabs. ( not shown in the following pictures)

It also have many advantages like if you reverse one of the cells in the lot to be paralleled, the overcurrent generated thru this cell will just blow the aluminum in this area.. just like a fuse!.. the thickness of this aluminum seem just perfect for that! :wink:

Doc
 

Attachments

  • P1080686_800x600.jpg
    P1080686_800x600.jpg
    108.7 KB · Views: 3,560
  • P1080688_800x600.jpg
    P1080688_800x600.jpg
    100.9 KB · Views: 3,560
  • P1080690_800x600.jpg
    P1080690_800x600.jpg
    89.6 KB · Views: 3,560
  • P1080691_800x600.jpg
    P1080691_800x600.jpg
    89.9 KB · Views: 3,560
  • P1080692_800x600.jpg
    P1080692_800x600.jpg
    85.5 KB · Views: 3,560
Marty got copper! Beautiful shinny metal!
Bought from a roofing supply place. 14 inch wide copper flashing. Cost is $12.91 per foot. They call it 16 once. Thickness is .0216 inch. Also called 22 Mils (a Mil is 1/1000 of an inch)

How should I clean the copper before soldering? Tried water to clean off fingerprints. Water don't work. I am thinking to clean it with water and dish soap using a green scouring pad.

Picture does not show how shinny this copper is. Told the wife that it was gold :wink:
copper600w.jpg
 
marty said:
Marty got copper! Beautiful shinny metal!
Bought from a roofing supply place. 14 inch wide copper flashing. Cost is $12.91 per foot. They call it 16 once. Thickness is .0216 inch. Also called 22 Mils (a Mil is 1/1000 of an inch)

How should I clean the copper before soldering? Tried water to clean off fingerprints. Water don't work. I am thinking to clean it with water and dish soap using a green scouring pad.

Picture does not show how shinny this copper is. Told the wife that it was gold :wink:

The fingerprint on your copper is due to the grease on the finger that trasnform into acid that atack the copper. It's normal !

just using a scotchbright will do the job and clean everything.

Doc
 
that stuff is G10 or bakeware or fiberglass board .. un-printed circuit board.. pretty tough stuff and almost fire proof.. but you can substitute with all sorts of stuff.. pvc, lexan, acrylic, my personal cheap solution is " Crazy carpet " . . thick plastic kids slide down snow hills on.. 1$ for a big piece of it..

http://www.google.ca/search?hl=en&sugexp=frgbld&gs_nf=1&cp=10&gs_id=t&xhr=t&q=crazy+carpet&bav=on.2,or.r_gc.r_pw.r_qf.,cf.osb&biw=1280&bih=619&wrapid=tljp1331818544543018&um=1&ie=UTF-8&tbm=isch&source=og&sa=N&tab=wi&ei=MvBhT4L6Nqby0gGm8YTLCA
 
Ypedal said:
that stuff is G10 or bakeware or fiberglass board .. un-printed circuit board.. pretty tough stuff and almost fire proof.. but you can substitute with all sorts of stuff.. pvc, lexan, acrylic, my personal cheap solution is " Crazy carpet " . . thick plastic kids slide down snow hills on.. 1$ for a big piece of it..

http://www.google.ca/search?hl=en&sugexp=frgbld&gs_nf=1&cp=10&gs_id=t&xhr=t&q=crazy+carpet&bav=on.2,or.r_gc.r_pw.r_qf.,cf.osb&biw=1280&bih=619&wrapid=tljp1331818544543018&um=1&ie=UTF-8&tbm=isch&source=og&sa=N&tab=wi&ei=MvBhT4L6Nqby0gGm8YTLCA


Great idea for teh crazy carpet!

Doc
 
Doc, are you selling these packs? did I miss something about that or is there another thread somewhere about buying /selling them?
 
http://endless-sphere.com/forums/viewtopic.php?f=9&t=38961
 
Back
Top