Large Group Purchase of A123 20ah cells

[Rush]

Ian,

Great details... thanks for the info.

I asked since I just got a HP3852A and a 20 channel Relay MUX. I plan to use it in conjunction with Mike's Grid charger/Discharger to test 20 sticks to determine which ones are good/bad.

I also think I might be able to get Labview, not sure of what version yet, to log everything. But then again, I might not need it since I think that with a simple program, the HP will log all the data to a spreadsheet.

I've got a lot to assimulate to make everything work together, I'm stepping into territory that is a little beyond my expertise. So hearing what has been done helps me focus on what I need to do to set everything up to achieve some meaningful results.

Again thanks.

[IamIan]

Originally Posted by Rush

Great details... thanks for the info.

Welcome

Originally Posted by Rush

I've got a lot to assimulate to make everything work together, I'm stepping into territory that is a little beyond my expertise. So hearing what has been done helps me focus on what I need to do to set everything up to achieve some meaningful results.

Take your time and you'll do fine.

As for meaningful results ... My recommendations in order of importance would be as follows:

#1> Consistency

#2> PC Data Logging and Analysis

#3> Matching
Having cells and sticks that match each other as closely as possible in as many ways as possible is as important as the strength of the cells and sticks.

#4> Full Usable capacity
( Wh preferred but Ah can be a fine secondary option )

#5> Efficiency
( optionally Internal Resistance as a secondary )

#6> Self Discharge Rate
Ideally over at least 4 to 8 weeks of sit time.

Best of luck.

[IamIan]

One more interesting thing I noticed while crunching data options for trying to pick the over all 'best' 50 cell pack... from my batch of cells.

• If I sort by the 50 Best Wh of capacity cells I get:
  • Weakest cell in the series chain is the limit of the whole pack to avoid cell reversal.
    So a peak 50 cell pack potential capacity of about ~2,874 Wh.
  • An Average cell efficiency of ~93.673%
    Difference from high to low cell is ~4.07%
    This is the potential per cycle out of balancing effect.
  • An average 60 Day Self Discharge Rate of ~42.136 mWh / Day
    Difference from High to low cell is 59.773 mWh/ Day
    This is the potential per day out of balancing effect.
• If sorted by the 50 Best Cycle Efficiency cells I get:
  • Weakest cell in the series chain is the limit of the whole pack to avoid cell reversal.
    So a Peak 50 cell pack potential capacity of about ~2,802 Wh
  • An Average cell efficiency of ~93.827%
    Difference from high to low cell is ~1.53%
    This is the potential per cycle out of balancing effect.
  • An average 60 Day Self Discharge Rate of ~42.171 mWh / Day
    Difference from High to low cell is 58.672 mWh/ Day
    This is the potential per day out of balancing effect.
• If sorted by the 50 lowest Self Discharge Rate Cells I get:
  • Weakest cell in the series chain is the limit of the whole pack to avoid cell reversal.
    So a Peak 50 cell pack potential capacity of about ~2,802 Wh
  • An Average cell efficiency of ~93.745%
    Difference from high to low cell is ~3.45%
    This is the potential per cycle out of balancing effect.
  • An average 60 Day Self Discharge Rate of ~39.677 mWh / Day
    Difference from High to low cell is ~44.54 mWh/ Day
    This is the potential per day out of balancing effect.

I find it interesting how close the results are ... once the worst 5 cells are removed.

Personally I think I'll be going with the 50 best cycle efficiency cells:

• I don't think the extra ~72wh from the peak Wh pack is enough to be meaningful.
• I think the less than 60 mWh/Day of potential Self discharge pack balance issue is so slow that it would take years to be an issue , and any time a properly balanced charge is done it resets that drift clock.
• Although the average pack efficiency is not significantly any better ... The narrower gap in efficiency reducing the potential gap between cells on a per charge basis I think is potentially useful to reduce the burden load on the BMS.

I also like the idea of putting the highest most efficient cells toward the center of the pack and the slightly lower efficient cells toward the outside edges of the pack ... this will just be one more tiny step to try and balance the thermal load in the complete pack.

[retepsnikrep]

Very interesting and detailed data set Ian. When I bought my cells I didn't bother with any of that as i didn't have the equipment to do the testing. I wish now I had at least tested the initial capacity properly though. Anyway my A123 car is actually getting more daily use now than ever before, and is shortly off on another 2000 mile trip around Scotland. The point is even a rough and ready installation like mine is still working fine after two years.

Originally Posted by IamIan

I also like the idea of putting the highest most efficient cells toward the center of the pack and the slightly lower efficient cells toward the outside edges of the pack ... this will just be one more tiny step to try and balance the thermal load in the complete pack.

I also like the idea of putting the most efficient cells in middle of pack and the others towards the edges.

I suspect your pack when built (with adequate cooling) will last for years

[IamIan]

Originally Posted by retepsnikrep

Very interesting and detailed data set Ian. When I bought my cells I didn't bother with any of that as i didn't have the equipment to do the testing. I wish now I had at least tested the initial capacity properly though. Anyway my A123 car is actually getting more daily use now than ever before, and is shortly off on another 2000 mile trip around Scotland. The point is even a rough and ready installation like mine is still working fine after two years.

Thanks.

From what I've seen in my initial testing. I'd say a pack build project can benefit from the kind of testing I've done. To narrow down and quantify things a bit more. Match the best cells together etc.

I also see how many people could probably skip this time consuming step. While they might be taking their chances a bit , and might not get the best potential pack. It seems to me now , that it isn't all that especially likely to be a major problem even if they did skip this step completely. Your pack is a testimonial to that.

Even my testing shows, that even if someone put the weakest ( but not DOA ) cells together into the weakest pack. My data set from my lot of cells suggests the worst they might have seen would be about ~5.6% efficiency balance separation per cycle , a ~2,600 Wh pack capacity , and ~677mWh/Day Self Discharge balance separation. That is only ~7.2% less usable capacity. Your BMS will easily keep pace with the ~677mWh/Day. I think it would be pretty difficult to have some kind of driving conditions that the ~5.6% potential per Cycle balance separation would be able to get ahead of your BMS either. It's only ~3 Wh per full cycle.

Originally Posted by retepsnikrep

I also like the idea of putting the most efficient cells in middle of pack and the others towards the edges.

I suspect your pack when built (with adequate cooling) will last for years

Thanks.
I hope so too.
My ideal would be 10+ years out of it ... but I do recognize that , that will be a long hall ... and allot can happen in 10 years.

[Complexity upon Cimarron]

Anyone see the news this week? A123 declared bankruptcy and announced that Johnson Controls will buy them out, I think. If you really want to make a bulk order, better to do it while the cells are still available!

[Hugh-Falls]

Originally Posted by Complexity upon Cimarron

Anyone see the news this week? A123 declared bankruptcy and announced that Johnson Controls will buy them out, I think. If you really want to make a bulk order, better to do it while the cells are still available!

Another way to look at this situation is to avoid them entirely at this time because they have nothing to lose by cleaning out all the bins. They won't be around so who warantees the purchase? You could be left holding a bunch of culls or worse.

[IamIan]

Originally Posted by Hugh-Falls

Another way to look at this situation is to avoid them entirely at this time because they have nothing to lose by cleaning out all the bins. They won't be around so who warantees the purchase? You could be left holding a bunch of culls or worse.

At the very least ... I'd recommend testing what you get and buying enough to be able to weed out the worst to get a reasonably good matched pack of cells.

- - - - - - - -

And another way to look at it ... is ... wait ... the intellectual property rights end up being owned by a company better equipped to bring them fully to market ... if they choose to do so.

[Complexity upon Cimarron]

Originally Posted by IamIan

... the intellectual property rights end up being owned by a company better equipped to bring them fully to market ... if they choose to do so.

Yeah, and that's a big if. At least they weren't bought by an oil company, like the battery technology used in the General Motors EV.

[IamIan]

Originally Posted by Complexity upon Cimarron

Yeah, and that's a big if. At least they weren't bought by an oil company, like the battery technology used in the General Motors EV.

The if is more like ... a if the specific battery technology is actually profitably cost effective worth it ... not as much if they choose to ... at least for my 2 bits.

For Profit Companies want a profit ... the global battery market last I checked in 2011 was over $89 Billion dollars a year ... that is allot of money ... And given the increasing global usage of batteries ... for portable electronics like cells phones ... our vehicles , HEVs, PHEVs, BEVs ... and with Utility grids looking to add more themselves ... this already large market is very likely to continue to grow for many years to come.

If you are a company that has a product you think can get you a significant piece of this many $Billion a year and growing market ... either you do it ... or your competition does it for you ... and someone else makes those Billions ... your loss.

Even if there is some suborn and foolish person / company that just sits on some useful patent ... that is why they expire ... eventually the patent expires and any body can use it.

I think the GM EV oil company reference you were referring to the NiMH patents for the EV1 Batteries ... there is allot of false information about that floating around ... There were never actually blocked ... It followed supply an demand just like anything else ... in order to leverage the technology like any patent holder in the world be in Intel , Sony, GM, Exon, General Mills , etc ... you need an investor who has deep enough pockets to commit to a contract for ___ thousands or millions of units per year for ___ Years ... in order to make it financially viable to risk the up front capital to build a factory , establish contracts with material supply vendors, establish contracts with shipping companies to get raw materials to the factory and finished good from factory , and establish contracts with distribution outlets in order to try and sell the product ... those parts of the market that had enough demand to bring in such an investor , did it ... NiMH batteries were sold all over the place AAAA , AAA, AA , C , D , 7series AAAA packaged as a 9v , power tools, Forklifts , Telecommunications , etc.

Also ... the NiMH battery was invented back ( before 1970 ) ... before the Ovshinsky work and patents about a specific type of variation / improvement to the batteries electrochemical design ... long before GM itself purchased a controlling interest in 1994 ... Not only were those specific patents leased out to 50 different companies to market ... and they did ... but anyone was perfectly able to produce NiMH batteries without those specific Patent methods and not be in any violation at all ... there have been several battery companies that have worked to improve the NiMH battery chemistry seperately without using the Ovshinsky patent methods ... Bipolar NiMH batteries and Nilar NiMH batteries are variation on the NiMH chemistry that do not infringe or use the specific Ovshinsky patent methods , and are thus immune to any such issues.

I don't think there was a conspiracy ... I think there was just not enough consumer / market demand ... if there had been enough demand for what the technology of that time was capable of , it would have sold ... but there was just not enough demand.

Even today ... the HEV , PHEV , BEV market demand ... is still a tiny % piece of the total vehicle market demand ... and that is with modern technology with more than 3x more energy / range capacity , from smaller , lighter, and more efficient motors and control electronics ... even with the completely superior technology of today ... the market demand is still a tiny piece of the total global market ... it is growing ... but it is still a tiny %.

Lastly ... GM did actually sell a few of the EV1s ... not just lease them ... sold them outright ... about ~6 Years ago , as an adult student at Pennsylvania State University ... I had a class that saw , and did some minor work on one that they owned ... it was part of an engineering class I had there ... they were converting it to run on a type of fuel cell that a graduate student had built.