A123 20AHr Pouch Cell Battery Build & Info Thread

megacycle said:
Got a confimation of recieved email, hopefully we'll get the info, fingers crossed, never works for me :?
Hmm still waiting for reply, try again today.
 
I need to pull 300A peaks for up to 30sec. Dare I try it with 1p of my AMP20's or should I use 10-15p of my old M1's so I can spread it over plenty of connections?
 
John in CR said:
I need to pull 300A peaks for up to 30sec. Dare I try it with 1p of my AMP20's or should I use 10-15p of my old M1's so I can spread it over plenty of connections?

If you don't have your A123 20Ah cells under the proper pressure range evenly distributed according to specs, then it's better to go with cylindrical cells which contain/maintain the pressure correctly already.

Start here and read following posts afterwords...

A123 20Ah Pressure Specs

Check This Too & Read the Following Post for Sure to See the Correction! :twisted:

.
 
Just sharing / passing along some testing data.

One aspect that seemed interesting to me ... is the minimum sustained / trickle current rate for the A123 20Ah cells.

There are some people who insist that LiFePO4 cells can not sustain a slow trickle charge rate ... but I see a problem with those kind of absolute claims ... the cells do have a self discharge rate... see attached.

The issue with it is that when the OEM A123 tells you that the cells have a known self discharge rate ... about ~3% in 1 year @ 25 degrees C ... up to about ~8% @ 40 degrees C in 1 year... that comes down to a sustained trickle rate in order to counter that effect... the higher ~8% of 20Ah is ~1.6Ah per year ... or ~133mAh per Month ... or ~4mAh per day ... or a sustained trickle rate of ~180 uA .... the Lower ~3% Self Discharge Rate would be a sustained trickle charge rate of about ~67uA.

That is a extremely slow trickle / sustained rate ... so it does make sense why people would think it was nearly zero... and fairly useless ... and not useful in any practical sense ... even the higher ~180uA would take about ~230 days just to put 1 Ah of charge through the cell.

But when looking at the OEM Self Discharge rate guide there is a curve to it ... ie the self discharge rate is faster in month 1 then it is in month 6 ...month 6 is faster than month 12 ... etc ... where things like the ~3% to ~8% are the cumulative loss over the entire 12 months.... that curve to the rate may eventually flatten into a straight line ... but at least in the beginning there is a curve to the self discharge rate.

And this lines up with the Self Discharge rate testing I did for my 55 cells.... for example ... one specific cell over the 1st 2 months had an average self discharge of ~37.87 mWh per day ... but the same cell over a 12 month period had an average self discharge rate of only ~9.8 mWh per day.

So I was curious about how high the self discharge rate would go as someone gets closer and closer to no rest period after charging.... ie a sustained trickle charge.... even if it ended up just being academic.

I also figured I would check / try and quantify the change ... if any ... at different SoC / SoE / Cell Voltages... I suspected the higher SoE would have a higher self discharge rate... maybe even get up to 1mA.

- - - - - - -

So I started testing.

At first it seemed to be going as expected ... CV to 1mA was getting longer and longer ... the curve of the amps seemed to continue to suggest that there was a sustained trickle charge point out there ... it seemed like I just had to go to a higher SoE / terminal voltage to get it up to at least 1mA... eventually the data ended up looking a bit odd ... and not as expected.

CV 3.0v 2A to to 25mA took ~16 Minutes ... went bellow 1mA Graph
CV 3.1v 2A to to 25mA took ~28 Minutes ... went bellow 1mAGraph
CV 3.2v 2A to to 25mA took ~45 Minutes ... went bellow 1mAGraph
CV 3.3v 2A to to 25mA took ~7Hours ... went bellow 1mAGraph
CV 3.31v 2A to to 25mA took ~14 Hours ... went bellow 1mA Graph
CV 3.32v 2A to to 25mA took ~47 Hours ... went bellow 1mA Graph
CV 3.33v 2A to to 25mA took ~24 Hours ... went bellow 1mA Graph
CV 3.34v 2A to to 25mA Took ~18 Hours ... went bellow 1mA Graph
CV 3.35v 2A to 25mA took ~27Hours ... went bellow 1mA Graph
CV 3.36v 2A to 25mA took ~50 Hours ... went bellow 1mA Graph
CV 3.37v 2A to 25mA took ~24 hours ... went bellow 1mA Graph
CV 3.38v 2A to 25mA took ~19 Hours ... went bellow 1mA Graph
CV 3.39v 2A to 25mA took ~23 Hours ... went bellow 1mA Graph
CV 3.40v 2A to 25mA took ~16 Hours ... went bellow 1mA Graph
CV 3.41v 2A to 25mA took ~15 Hours ... went bellow 1mA Graph
CV 3.42v 2A to 25mA took ~11Hours ... went bellow 1mA Graph
CV 3.43v 2A to 25mA took ~8 Hours ... went bellow 1mA Graph

At this 3.43v point the discharge was showing I was near the top of the cell ... not wanting to over charge the cell quiet yet ... at least until I do other testing ... I can say ... that it does seem that any potential for a sustainable trickle charge ... is well bellow 1mA ... even on a fully charged cell.

I was not initially expecting the bumps ... or for the CV time to go up and down as it did .... interesting.

Eventually after other testing ... I might try another round of testing by putting all 55 of the cells together in parallel ... in order to try and use that way to amplify the average sustainable trickle charge rate ... at least for detection / verification ... but as it is bellow 1mA ( under C/20,000 ) ... I don't see it being of any use in application... other than just re-enforcing that we can't use a trickle charge to balance the pack.
 
I had a battery seat for 4 months a little flux in cell voltages and very little drain and charged right up to 3.59v and balance easy after 2-3 cycles. Yes 3.45v is full but I balance charge to 3.59 volts. I'm happy but a little heavy.
 
I have several packs that I leave on the charger 24/7. Soon as they are used they go on the charger, sometimes unused for weeks left on the charger.

Having just serviced a 2 year old battery pack that has been used this way, it still gives back the same amount of tested ah's as it did when new. Headway cells were in this pack but I am seeing the same response from various other cells including 20Ah A123's.

The trickle charge coming from the charger and BMS is usually about 1/4 watt in total for the pack when it registers a charge but usually 0 watts whilst the pack is full. This has been through extreme cold winters and hot summers.

As for binding the A123 cells, the only cells I have had that have puffed from those muppets at Victpower were the ones that had been pierced already or had solder blobs between the cells.

I basically stripped all my packs down and cleaned each cell, tested them for punctures and capacity throwing quite a few away that failed this....and rebuilt the packs. Strong tape to give a firm pressure of the pack and pulled no more than 30Amps from a 12 cell pack, still going strong a year later.

Lost a lot of $ on those shitty packs, mostly in my time rebuilding them. Was a quick experiment and whilst I have faith in the actual A123 cell itself, dodgy seconds are not the way forward.
 
Spacey said:
I basically stripped all my packs down and cleaned each cell, tested them for punctures and capacity throwing quite a few away that failed this....and rebuilt the packs. Strong tape to give a firm pressure of the pack and pulled no more than 30Amps from a 12 cell pack, still going strong a year later.

Lost a lot of $ on those shitty packs, mostly in my time rebuilding them. Was a quick experiment and whilst I have faith in the actual A123 cell itself, dodgy seconds are not the way forward.
Thanks for the confirmation of our experience with the grey market. I don't regret the purchase, as I have a working AMP20 pack that costs less than purchasing new. New is possible now, thanks to A123 AMP20 Cells, new, legit, factory direct!. FYI, I'm building a second pack now with the A123 26650 cells, after a full consideration of options.
 
Instead of compressing the cells, how about setting them in lightweight plaster in a rigid plastic box. Each cells would be separated by plaster too.
 
Carlito said:
Instead of compressing the cells, how about setting them in lightweight plaster in a rigid plastic box. Each cells would be separated by plaster too.

I think Rickard tried enclosing them in lucite or some other hardening plastic, and it didn't work out well. Hard to replace a cell, heat dissipation is an issue, and it was heavy.

-JD
 

This is on my Durtledge build A123 20ah 24s battery. From the midwest group buy from J.R.Holmes. I have got 25 cells and replaced one. And one is of a little lower after discharge like 3.27 and the rest 3.28v but not worried. I can get 17 ah and twrice 18ah. i don't do that often. More like 9-14 ah discharge. The 1420 is working fine after exchange of balance wires to 3.59v balance every time. I have put the 36v 3a. @ 43 volt charge voltage and some cells ran to 3.8 volt one cell 3.9v opps, discharge soon after.
Just wondering how everybody esle is doing with there's.
 
acuteaero said:
Also there is JD's spot-welding approach http://endless-sphere.com/forums/viewtopic.php?f=14&t=29959 which I think is pretty interesting and appears to work well enough in his experience (an important qualifier!) but definitely requires full and absolute commitment! Given the current low price of these A123 cells I would be more willing to consider trying a non-reversible and high-impact approach like that!
Sir,
I am a student and making an electric car for Formula SAE competition. We are using Li-Poly cells of 31Ah. The C rating is 8 cont. and 12 peak. Which means the cont. current is 248 Amp and peak is 372 Amp. My continuous motor current is around 175-200 Amp. I am planning to put a tin or silver coated copper bar(the yellow object in the picture) of 5mm in between the +ve and -ve terminals of the cell for series connection. I will be making holes in both the terminals and also in the copper bar to hold them together.

But I don't think the cell 0.3mm terminals can hold the weight of 5mm copper bar in that position and will be affected by vibration.

Can anyone help me to determine the width of copper required to withstand the temperature rise due to high current. Is there any better way to build the battery pack.
 

Attachments

  • 11.png
    11.png
    436.5 KB · Views: 2,988
  • 22.png
    22.png
    86.8 KB · Views: 986
You might check with agniusm (group member) for a connection board. He makes the best kits out there but might have to be custom for your battery depending on tab sizes.
otherDoc
 
On my 211A short, couple min run i can confirm that there is no heat build up on series terminations. your concern is end terminals.
 
kalyandash2010 said:
acuteaero said:
Also there is JD's spot-welding approach http://endless-sphere.com/forums/viewtopic.php?f=14&t=29959 which I think is pretty interesting and appears to work well enough in his experience (an important qualifier!) but definitely requires full and absolute commitment! Given the current low price of these A123 cells I would be more willing to consider trying a non-reversible and high-impact approach like that!
Sir,
I am a student and making an electric car for Formula SAE competition. We are using Li-Poly cells of 31Ah. The C rating is 8 cont. and 12 peak. Which means the cont. current is 248 Amp and peak is 372 Amp. My continuous motor current is around 175-200 Amp. I am planning to put a tin or silver coated copper bar(the yellow object in the picture) of 5mm in between the +ve and -ve terminals of the cell for series connection. I will be making holes in both the terminals and also in the copper bar to hold them together.

But I don't think the cell 0.3mm terminals can hold the weight of 5mm copper bar in that position and will be affected by vibration.

Can anyone help me to determine the width of copper required to withstand the temperature rise due to high current. Is there any better way to build the battery pack.

Hi,
hello, and nice to meat you in this great Forum!
(sorry for my bad English!)

-one short question: Why you go with these cells, why you not go with the new high current/low Ri 18650 cells from Sanyo, Sony, Panasonic, Samsung like Tesla, BMW and more?

-In the past I must create also a little project with this type of cells but this was my first and last project with this cells, because no helping hand from this issue from the manufacturer and a lot of more cons.
 
999zip999 said:
It's easier to connect one 20ah 20c cell then 8 round cells in parelle for same ah.

hi.

your math are not correct and your answer are also not correct, do you have read the post from the student, they have probs with the big 31Ah cells
and need help....
 
ecross said:
999zip999 said:
It's easier to connect one 20ah 20c cell then 8 round cells in parelle for same ah.

hi.

your math are not correct and your answer are also not correct, do you have read the post from the student, they have probs with the big 31Ah cells
and need help....

This thread is exclusively about a123 prisimatic LifePO4 cells. The round a123 LifePO4 cells are 2.3ah, so (8) cells x 2.3ah = 20.4ah. Zip's math is correct, and his assessment of it being harder to parallel (8) cells than use one cell straight out of the box is correct, not only from an assembly standpoint, but from a management standpoint, and it is even harder with LiCo. That is why tesla has such a complicated LiCo battery pack with a separate receptacle for each Individual cell instead of simply soldering them together.

That is also why this thread is about LifePO4 that does not experience thermal runaway, and not LiCo such as you and the student are talking about, which is why your comments are totally out of place here. If you want to find out the hazards of paralleled LiCo, please go start a thread about it. The student asked for advice about joining the tabs on large format cells, which this thread addresses, so their posts are relevant, and they received the help that was available.

-JD
 
Oatnet, it is useless to explain to someone who thinks to be the best out of us here on ES. He does not understand thermal runaway, proper assembly to begin with.
 
I like large format cells 15ah or 20ah 5c min. So the A123 is great for a commuter bike. I wish there was so much problems from resellers. As Victpower and short tabs that was making a bad name. A123 cells@ 70.00 ea. is a harder sell so goingcdrdf with a different A1123 format or different cell in parelle. Or looking for a large format cell around 30.00 usd ?
 
I'm not on that much lately so am not current in this area. Is there anyone on the forum that solders Lipo pouch cells together including a balance tap wire for a fee? I need ten, 10 Ah Lipo pouches soldered together but not final assembled into a pack. The pouches are a standard 10 Ah, 5C item on Battery space and I can buy them or buy them from the person who solders them. I have to insert micro sized temp probes between each 2 cells along with spacing materials so that the pack will interface with my special BMS. I have very detailed pics of how it needs to be configured. Basically two bundles of 5 cells each connected with a piece of 10 awg cable.

EDIT: Got brave and did it myself. Cells have copper tab extensions and it wasn't difficult.
 
Has anyone found a vendor that sells plastic spacers for mechanically joining the cell tabs?

I'm specifically looking for the sort in this video:
http://youtu.be/O3VtYwJz78g

They appear to scale with dove tails connecting each bar.
 

Attachments

  • Screen Shot 2014-07-08 at 1.48.31 AM.png
    Screen Shot 2014-07-08 at 1.48.31 AM.png
    81.6 KB · Views: 2,555
I believe OSN Power makes those kits for AMP 20 A123 cells. Agniusm also makes kits for the same cells. He resides in Lithuania.
otherDoc
 
Thought I'd toss a few pics of my pack build up.
I used the compression method. 12S per pack and two packs on the bike.

I've only had one puffer... My own fault; I got distracted by a co-worker while charging and accidentally shorted it. It lasted for a while after that but with reduced capacity. Otherwise, the packs have been in use for about a year and a half now.

2012-07-20 22.16.10.jpg

2012-07-20 22.17.15.jpg
2012-08-22 12.40.30.jpg
2012-08-22 12.21.00.jpg
2012-08-22 13.18.04.jpg

Each cell body is separated from it's neighboring cell by a thin sheet of lexan - this prevents shorting of the cell bodies together.

The tabs are compressed separately from the cell bodies. The two lexan plates compress the cell bodies and the tabs are compressed by a quasi-floating end plate. The cell tabs are connected with Alum bars and isolated from one another with nylon plates. There are screws in the top of each of the aluminum bars so that I can charge the cells individually - each cell has it's own dedicated charger...

2014-06-20 20.19.56.jpg
 
EVil said:
Thought I'd toss a few pics of my pack build up.
I used the compression method. 12S per pack and two packs on the bike.... The tabs are compressed separately from the cell bodies. The two lexan plates compress the cell bodies and the tabs are compressed by a quasi-floating end plate. The cell tabs are connected with Alum bars and isolated from one another with nylon plates. There are screws in the top of each of the aluminum bars so that I can charge the cells individually - each cell has it's own dedicated charger...
Superb design & build. Kudos. What are using for dedicated charger? Thanks.
 
These is a good repete of a drutledge build which I followed two years ago.
It's been working fine for two years. Two 12s for 24s at 30amps to 40 amps. Its a commuter and works fine I don't know what wattage I can deman from this pack.
Always balance charge to 3.59v.
 
Back
Top