A123 20ah Lithium Install

[retepsnikrep]

The BCM Interceptor does prevent the initial forced charge as the MCM thinks the battery is at 19 bars. In fact the Interceptor prevents any parasitic charging unless your are braking or slowing with foot off the gas. So on my recent journey I had no forced charging of any sort, and the extra battery capacity comes without any onboard penalty. In a normal vehicle you would expect parasitic charging at various times as the car replenishes the nimh pack. So you get an instant mpg gain. Then you can actually start using the extra capacity to overcome hills etc and mpg soars.

The Lifepo4 cells do have a finite calendar life. I hope to get > 6 years out of them, i might get more who knows. Of course it's cooler over here so that may help me.

As far as capacity or battery size determination is reqd, I may add that option to the new BCM gauge thingy so it keeps a running log of an immaginary battery of a size you determine. Then you can see how many AH you would actually be using or require for your commute.

For rough ball park figures.

40ah of capacity allows almost unlimited use of assist upto 50A over a 100 mile trip. Giving 150mpg capability. This can be ekked out over whatever distance you choose of course. I limit my 40ah cells in my original PHEV to 50A discharge.

20ah of capacity allows almost unlimited assist over 50 miles or so again giving very good mpg numbers. I don't limit max assist in this car as it is my sport version and in fact goes better than standard due to the higher battery voltage under load.

The two most positive things about the new conversion are

1) The higher battery performance under load etc

2) The fact it fits in the stock space/case and has no weight penalty

Don't underestimate the time, effort and money involved though, it took me several months and several thousand pounds. I did/do not expect to save any money with the conversion, i did it for my own enjoyment and to save faffing about with stock cells/packs. Others who wish to convert can get some stuff from me like BCM interceptors and BCM Foolers and the BMS if they wished so they have a reasonable head start.

Hope that helps. Peter

[retepsnikrep]

No problems so far. Temps fine, cell capacity seems very well matched and when fully charged all reach max voltage very quickly once the first one is full.

Haven't had time to do a special charger for it yet though as too busy with other stuff.

[retepsnikrep]

Nothing much to report.

I improved by BMS software a bit over the year and added a two stage 1kw charger but that's about it. I haven't looked at the cells and use the car almost every day!

Highest battery temp was about 40C one hot day when they where exhausted. I have no active cooling. They stay in balance very well and give excellent performance as you have all seen on the more power threads. I def recommend the A123 20ah cells No signs of any cell weakness so far.

[3-Wheeler]

Originally Posted by retepsnikrep

....Highest battery temp was about 40C one hot day when they where exhausted. I have no active cooling.....

Hi Peter,

So regarding the 40C temperature, was that after running the battery from full to empty on a short trip, or an artifact of the cells as they approach a discharged state?

Jim.

[Cobb]

Wow..... I need to read up on that. I have the Enginer Phev kit that was recent upgraded to 4 from 2 kilowatts. I notice when I set the voltage for the I2 at 117 volts I get an instant 10 more mpg, but any real regen it throws an ima code p1586. At 115 volts I still get more than stock economy and no codes.

As a side note I had the original blue cells with purple tops the size of vhs tapes, 16 of them from az power solutions. When I upgraded I got 2 metal boxes from real force. Inside the box is the same plastic sleved cells. Both boxes are 24 volts roughly and 2 kilowatts.

Originally Posted by retepsnikrep

MN

If you fake 20 bars then regen is reduced so 19 bars allows full regen and full assist so long as the battery temp is acceptable, and in fact I fake that as well telling the MCM via the interceptor that the cells are always at 20C. Lithium arent really affected by temp like nimh so it doesnt really matter, again my own BMS screen tells me the actual cell temp.

[retepsnikrep]

Originally Posted by 3-Wheeler

Hi Peter,

So regarding the 40C temperature, was that after running the battery from full to empty on a short trip, or an artifact of the cells as they approach a discharged state?

Jim.

The cells heat more when in a low soc say < 30%

Hard use over a very short distance say 100A sustained for several minutes on a steep hill also heats them but not as much as if soc is low.

The 40C temp isn't an issue in fact they were very lively at that temp max reccomended temp is about 60-65C ASFAIK. My BMS is set to give the alarm at 45C.

[Eli]

On a rather unrelated note, my 1yr old stock pack hit 108ºF(42ºC) today while giving her a good top off. Got to 81.8% before the BCM stopped regen.

Ambient temp was 77ºF(25ºC), and cabin temp was 87ºF(31ºC). Took about 15 miles of low to moderate regen(~10-25A).

[IamIan]

Any 2 year updates on the Prototype A123 20Ah pouch pack? ... Interesting things you've noticed? ... Looking back on it you would have done anything differently if done a 2nd time?

I was reading some people using a sandwich of sorts of aluminum to clamp connect the pouch tabs instead of trying to solder them ... Link... Thoughts?

Originally Posted by retepsnikrep

At around 50% soc at 20C (Temp) I would estimate at this early stage that cell voltage rises to no more than 3.50v under 50A charge. If we say the nominal voltage of the cells is 3.35V in this example that gives a V dif of 150mv x 50A = 7.5W of heat being generated. Or 375W for the entire pack. If we assume the pack is a block of copper weighing 22kg someone clever with maths and specific heat capacity can work out how long it will take to heat up? etc etc.

Copper has a molar heat capacity of about ~24.44 joules per Mol per degree K ... 29 grams of copper per mol ... 22kg of copper would be ~758.6 Mols of Copper ... ~758.6 * ~24.44 = ~18,540 Joules per K ... 1 Wh = 3600 Joules ... 20 Ah cells I would guess would only be able to sustain ~50A for about ~20 Minutes straight ... ~20 Minutes of 375W of heat = ~125 wh of heat energy ... ~125 wh * 3,600 = ~450,000 joules / ~18,540 = up to ~24.3 Kelvin / Celsius change in temperature from whatever the initial temperature was.

That does not take into account any heat leaving the battery system ( conduction or convection ) ... and it does not account for any difference in specific heat between the A120 pouch 20Ah cells and copper.

Also ... AFAIK A123 Pouch 20AH cells Internal resistance changes as their temperature changes ... so there will also be a compounding effect from that changing the ~375W of heat.

If we have some initial at rest temperature readings from the center of the pack and some final temperature readings from the same location after a full discharge run , we could try to correct for some of those differences ... Center should be worst case because it will be the most concentrated heat and most insulated.

[retepsnikrep]

Originally Posted by IamIan

~24.3 Kelvin / Celsius change in temperature from whatever the initial temperature was.

Thanks for that analysis, my seat of the pants thoughts suggest that's pretty close and fits in with my experience over the last two years. Certainly a good basis to work from.

As for two years use. No significant issues and more importantly no cell failures even when using upto 75A regen and 125A of assist. Looking at my older posts and records I estimate I have done ~20,000 miles with the pack fitted.

Useful capacity is probably about 16ah but bear in mind my cells were not capacity tested prior to instalation, and the worst cell in the pack limits the overall capacity. I could easily have had an 18ah cell amongst the pack when I installed it. I now have a very accurate electronic load (bought with the proceeds of the OBDIIC&C) and am just testing some spare (new to me) A123 cells for my EBike upgrade, at the moment they are varying between 18.4 and 19.5ah. So the weakest cell will always be a limiting factor. Best to test each one and match capacity as near as possible.

I intend to remove the pack from my 40ah phev at some stage and test the 50 highpower cells individually to see if any of those has deteriorated more than the others. I need more spare time.

There are only a few small things I would change about the A123 install as i don't have unlimited resources.

I would not bother with the insulating sheet of plastic between each cell, nor would I use adhesive to stick them together. (I'll never be able to get them apart !! LOL) if a cell fails I will have to bypass it.

I like the soldering as i'm pretty good at it with 40 years of tinkering and IMHO it gives better connections with less heating and losses. However for ease of assembly and connection the folding copper 1mm plate may be a better option. Lots of discussion on assembly on the endlesspehere forum.

[RedJellyBean]

Would it be feasible for you to sell a kit that provides the interface needed between the A123 cells and the stock BCM?
The cells can now be bought for about $1500-1600, if I'm not mistaken - so that part is less cost prohibitive than it was, and it's been proven pretty decently well in your setup. Hobbyists can put the cells together into the stock battery pack just fine, but most don't have the means or knowledge to make the 165v Lithium pack to work with the stock BCM.