Lithium Battery Reports & Tests' 9/08

[BMI]

But there is so much more to it than just price..

Want to power an X5 on a 10ah pack ? can't do that with 2C stuff..

40ah for the same price is one thing, but on a bicycle this is nearly impossible to mount reasonably ( not impossible lol.. but bulky.. ).

I like my PSI cells.. solid flat voltage at as many amps i can use on a hub motor.

I recently swapped my clyte 409 for a X5302 on my Norco, my 72v 20ah LiMn pack is not happy with 50 amps....

One smart cookie!!

 

[swbluto]

But there is so much more to it than just price..

Want to power an X5 on a 10ah pack ? can't do that with 2C stuff..

40ah for the same price is one thing, but on a bicycle this is nearly impossible to mount reasonably ( not impossible lol.. but bulky.. ).

I like my PSI cells.. solid flat voltage at as many amps i can use on a hub motor.

I recently swapped my clyte 409 for a X5302 on my Norco, my 72v 20ah LiMn pack is not happy with 50 amps....

One smart cookie!!

I agree with this guy. You just can't directly compare 2C technology with what seems to be 20-30C technology. For X amount of needed amps, you could theoretically have 10 times less the capacity although you may not always want 10 times less the capacity and, also, I've heard quite a few horror stories about consistency and quality of the Thunder Sky cells. If the 2C technology provides the amps you're seeking with the capacity you're seeking, it might make economic sense, but it wouldn't so much for shorter-range high-performance applications like racing or "true mountain biking" or just a "toy performance bike" like the PK ripper and so on. But, if you're only using the application only occasionally, it might more sense just to skip to LiPo since the calendar life would be comparable and the cycle life of LiFePO4 wouldn't be that advantageous.

 

[bobmcree]

i can easily produce a graph showing the cell delivering 10 Ah by charging it up to 4.0v instead of 3.65 as i did. I wanted to provide data showing how the cell would perform when charged to the recommended voltage. My experience with the Lifebatt green cells, which after all were made in the same factory by PSI, was that they increased in capacity over the first 40 or 50 cycles, and you must of course remember that my one BMI test cell was selected at random.

I have read from a couple of sources that to have the best chance to achieve the hoped-for 3000-5000 cycles, one must discharge the cells to only 40% for the first 40-50 cycles, which after all is only 1% of the expected lifetime. This is what i am suggesting to people i am building lifepo4 packs for at present.

My measurements are only good to 2% so i could easily report 9.8 Ah when the true value is 10.0. Anybody who claims to have better accuracy is probably wrong. I use a 6-1/2 digit meter calibrated to a NIST traceable 10 ppm standard, but i use it to measure the voltage across a 1% resistor. This means my measurements can never be better than 1.001% and there are many sources of error that contribute to an honest measurement accuracy claim of 2%.

I expect that over the next week while the bmi cell is cycling on my automated system that it will increase in capacity. If it remains at the same low impedance, as i expect, the Watt-hours delivered to the load will be much greater than the green lifebatt cells which have now been cycled about 20x.

I showed photos of my high current test setup, but anybody can duplicate it. All you need is a decent op amp that does not have a lot of DC offset or drift, a precision resistor, and a current metering device (fet or transistor). I chose a .010 ohm 100w resistor so that it would limit current through my fet array to 300A even when the battery is first connected. This will give me a volt at 100A so for .1A resolution i only need to measure a millivolt, which is pretty easy. I chose the biggest heat sink weirdstuff.com had in stock, and i used the IRFB4110s because they have good thermal conductivity from the device to the case for the couple of hundred watts i need it to dissipate when the single lifepo4 cell is fresh. I then used an lcd i have from the power supply kits, from MPJA.com, a transformer i have lots of, and a switch to select the set current, which is adjusted by a pot sourced with a zener reference, or the actual voltage across the resistor. I used the 199.9v range on the meter and divided the voltage at its input by 10x which reduces the dc offset and noise and lets me read 199.9A, plenty so it does not increase the basic error for measurements above 10A. The fets can only dissipate a few hundred watts, but if i want to measure higher voltages like a whole pack, i can just put some resistance in series.

I really was planning on producing more data today, but i was using the CBA2 as a metering device when a test lead melted and shorted, causing a chip in the CBA to blow a crater. Does anybody have the part number of the so-8 chip? I have 4 of the HP 6-1/2 digit meters and they are all HPIB but the interface is ISA so i have not had them logging data for awhile. Any advice on interfacing from newer computers to HPIB? I have a couple of other logging meters, but i have to go down to california tomorrow and help my dad out with some personal stuff. I will cycle the bmi and lifebatt cells another 10x over the next week, and when i return i will post more data that i expect will show both cells to increase a bit in capacity.

 

[BatteryMooch]

Hi Bob,
That SO-8 device is a TI TLC272 op-amp. Double-check the surrounding passive components too. I've seen them damaged more than once when a CBA blew. I also recommend Infineon's IPP048N06L MOSFET as a replacement for the IRL2910. Not only is it rated for use as a load (DC SOA plot line) but it runs significantly cooler than the IRL2910. The 2910 max's out at 85W or so in the CBA before the rated 175C junction temp is reached. And that's only for lower voltage levels too!

The IPP048N06l can give you over 100W continuous at 145C junction temps if you use a better fan on top and an additional fan on the bottom of the CBA. More info in my thermal analysis of the CBA II at http://www.camlight.com/techinfo/CamLight_Systems_CBA_II_Thermal_Analysis.pdf

But, if you're only using the CBA as a voltage logger, there's no need to update the MOSFET.

 

[BMI]

i can easily produce a graph showing the cell delivering 10 Ah by charging it up to 4.0v instead of 3.65 as i did. I wanted to provide data showing how the cell would perform when charged to the recommended voltage. My experience with the Lifebatt green cells, which after all were made in the same factory by PSI, was that they increased in capacity over the first 40 or 50 cycles, and you must of course remember that my one BMI test cell was selected at random.

I have read from a couple of sources that to have the best chance to achieve the hoped-for 3000-5000 cycles, one must discharge the cells to only 40% for the first 40-50 cycles, which after all is only 1% of the expected lifetime. This is what i am suggesting to people i am building lifepo4 packs for at present.

My measurements are only good to 2% so i could easily report 9.8 Ah when the true value is 10.0. Anybody who claims to have better accuracy is probably wrong. I use a 6-1/2 digit meter calibrated to a NIST traceable 10 ppm standard, but i use it to measure the voltage across a 1% resistor. This means my measurements can never be better than 1.001% and there are many sources of error that contribute to an honest measurement accuracy claim of 2%.

I expect that over the next week while the bmi cell is cycling on my automated system that it will increase in capacity. If it remains at the same low impedance, as i expect, the Watt-hours delivered to the load will be much greater than the green lifebatt cells which have now been cycled about 20x.

I showed photos of my high current test setup, but anybody can duplicate it. All you need is a decent op amp that does not have a lot of DC offset or drift, a precision resistor, and a current metering device (fet or transistor). I chose a .010 ohm 100w resistor so that it would limit current through my fet array to 300A even when the battery is first connected. This will give me a volt at 100A so for .1A resolution i only need to measure a millivolt, which is pretty easy. I chose the biggest heat sink weirdstuff.com had in stock, and i used the IRFB4110s because they have good thermal conductivity from the device to the case for the couple of hundred watts i need it to dissipate when the single lifepo4 cell is fresh. I then used an lcd i have from the power supply kits, from MPJA.com, a transformer i have lots of, and a switch to select the set current, which is adjusted by a pot sourced with a zener reference, or the actual voltage across the resistor. I used the 199.9v range on the meter and divided the voltage at its input by 10x which reduces the dc offset and noise and lets me read 199.9A, plenty so it does not increase the basic error for measurements above 10A. The fets can only dissipate a few hundred watts, but if i want to measure higher voltages like a whole pack, i can just put some resistance in series.

I really was planning on producing more data today, but i was using the CBA2 as a metering device when a test lead melted and shorted, causing a chip in the CBA to blow a crater. Does anybody have the part number of the so-8 chip? I have 4 of the HP 6-1/2 digit meters and they are all HPIB but the interface is ISA so i have not had them logging data for awhile. Any advice on interfacing from newer computers to HPIB? I have a couple of other logging meters, but i have to go down to california tomorrow and help my dad out with some personal stuff. I will cycle the bmi and lifebatt cells another 10x over the next week, and when i return i will post more data that i expect will show both cells to increase a bit in capacity.

Fantastic work Bob. I am sure everyone greatly appreciates all the effort you are going to with your thorough testing of these cells.
Although as a later generation and improved cell I expect the grey BMI cell to have slightly better overall performance compared to the first generation green Lifebatt cell since both cells are of similar origin with most of the same cell design team being involved in the development of both cells.

It would be really interesting to see how Headway/Ping etc, cells would perform in the same tests since these are not related to BMI or Lifebatt cells in any way. I am not sure how Doc is going with his testing but he has two of the BMI cells so uniformity could be determined between all three BMI cells which are from entirely different production runs.

Anyway Bob, I for one will be closely watching your results and I sincerely hope all the best works out on your trip to CA to visit your dad.

 

[TylerDurden]

BMI cells use LiFePO4 powder manufactured under licence in Germany from the patent holder of LiFePO4 battery technology (Phostech).

Really?...

Prove it.

 

[swbluto]

BMI cells use LiFePO4 powder manufactured under licence in Germany from the patent holder of LiFePO4 battery technology (Phostech).

Really?...

Prove it.

I was checking out the phostech lithium site at

http://www.phostechlithium.com/prd_LiFePO4P2_e.php

and they apparently list their customers' information in the links of the two LiFePO4 versions. They mention K2, Saft and ATL and no one else, so that does seem to be pretty damning evidence for LiFeBatt's claims.

WAIT!

I found this company page.

http://www.phostechlithium.com/clients_list_e.php

And they DO mention LiFeBatt! So what was the controversy with lifebatt? The amount of powder that's actually used from Phostech?

Damningly, I see no mention of BMI but yet there are inconclusive stars dotting the map indicative of unnamed customers.

Also, interestingly, LiFeBatt also claims use of Dumont Zytel material for their battery case corroborating some similar origins of BMI and LiFeBatt.

Also, this page, http://www.alibaba.com/product-free/103234400/High_Discharge_20C_Quality_LiFePO4_10Ah.html
claims that this battery is the brands for LifeBatt, BMI and EnviroSafe by Mr.Armin. Who is Mr.Armin?

And I've found a page with an interesting discussion with the relationship between BMI, PSI and LifeBatt. http://www.evdl.org/archive/index.html#nabble-p23498313

 

[BMI]

BMI cells use LiFePO4 powder manufactured under licence in Germany from the patent holder of LiFePO4 battery technology (Phostech).

Really?...

Prove it.

No problem Tyler!
Just let me know when you have your flight to Taiwan booked so I can arrange for you to be given a guided tour of the factory and you can see first hand the Phostech LiFePO4 powder being used to manufacture BMI cells in a factory (which is not the PSI factory as many people seeem to claim).
I will gladly arrange this for you personally!

You can report back your findings and this should once and for all silence all the skeptics who think I am not telling the truth and who can do nothing better than criticise without knowing the facts of the matter.

 

[TylerDurden]

I found this company page. And they DO mention LiFeBatt! So what was the controversy with lifebatt? The amount of powder that's actually used from Phostech?

Just because somebody bought a barrel of powder from Thorsten Lars, doesn't mean they put it in any batteries or put enough in to make a difference; if it even can make a difference. A whitepaper with comparative analysis between with and without could offer data to support the claims, but don't hold your breath.

Besides, the UT/HQ patent was revoked in the EU for lack of novelty, so big whoopie on the magic spooge.
"On Dec 9th, 2008, European Patent Office revokes Dr. Goodenough’s LiMPO4 patent, patent number 0904607. This decision basically reduces the patent risk of using lithium iron phosphate in automobile application at Europe. The reason of this decision is believed to be based on the lack of novelty. While UT can still appeal the EPO decision, this result encourages the electric vehicle makers to pursue on lithium iron phosphate battery technologies. [7]"
http://en.wikipedia.org/wiki/Lithium_iron_phosphate#LFP_Patent_Wars

 

[BMI]

BMI cells use LiFePO4 powder manufactured under licence in Germany from the patent holder of LiFePO4 battery technology (Phostech).

Really?...

Prove it.

I was checking out the phostech lithium site at

http://www.phostechlithium.com/prd_LiFePO4P2_e.php

and they apparently list their customers' information in the links of the two LiFePO4 versions. They mention K2, Saft and ATL and no one else, so that does seem to be pretty damning evidence for LiFeBatt's claims.

WAIT!

I found this company page.

http://www.phostechlithium.com/clients_list_e.php

And they DO mention LiFeBatt! So what was the controversy with lifebatt? The amount of powder that's actually used from Phostech?

Damningly, I see no mention of BMI but yet there are inconclusive stars dotting the map indicative of unnamed customers.

Also, interestingly, LiFeBatt also claims use of Dumont Zytel material for their battery case corroborating some similar origins of BMI and LiFeBatt.

Also, this page, http://www.alibaba.com/product-free/103234400/High_Discharge_20C_Quality_LiFePO4_10Ah.html
claims that this battery is the brands for LifeBatt, BMI and EnviroSafe by Mr.Armin. Who is Mr.Armin?

And I've found a page with an interesting discussion with the relationship between BMI, PSI and LifeBatt. http://www.evdl.org/archive/index.html#nabble-p23498313

I don't know how many times I have to say it but PSI cells and BMI cells are NOT the same thing!
If they are the same we will expect Bob Mcree's testing to reveal the same performance from both PSI green and BMI grey cells. If Bob's results prove there is a considerable or marked difference between the cells I hope people will finally come to realise they are two different cells...end of story!

 

[miro13car]

yes,
my 36V 10Ah HPS pack delivers 11Ah at real life use on Tidal Force Canadian model 500W restricted.
Real life means:
no constant speed, 25A on several take offs, long streaches of road at 8-10A, one 100-150m steep hill where I turn to NORMAL MODE for exercise/300-400W draw/, pedalling just for show , many regen breaking and mild accelerations .
I am 150 pounds.
11Ah on conditions:
weather above 20C
well balanced cells.
my pack has no problem to deliver 52 A for couple of second when I do full throttle from standing start , but of course even 30A cannot be sustained due to 31km/h limit which I reach in just under 10 second .
from my experience :
pack will bearly deliver 9Ah if pack is not balanced /more than 0.3V difference beween cells/.
pack will deliver progressivly less with temp drop. in winter at minus 5C I estimate maybe 8Ah??
MC

 

[BMI]

yes,
my 36V 10Ah HPS pack delivers 11Ah at real life use on Tidal Force Canadian model 500W restricted.
Real life means:
no constant speed, 25A on several take offs, long streaches of road at 8-10A, one 100-150m steep hill where I turn to NORMAL MODE for exercise/300-400W draw/, pedalling just for show , many regen breaking and mild accelerations .
I am 150 pounds.
11Ah on conditions:
weather above 20C
well balanced cells.
my pack has no problem to deliver 52 A for couple of second when I do full throttle from standing start , but of course even 30A cannot be sustained due to 31km/h limit which I reach in just under 10 second .
from my experience :
pack will bearly deliver 9Ah if pack is not balanced /more than 0.3V difference beween cells/.
pack will deliver progressivly less with temp drop. in winter at minus 5C I estimate maybe 8Ah??
MC

Thanks for that MC.
I am quite surprised to hear you are actually getting 8Ah from the pack at minus 5 degC.
I would of thought it would be less. All LiFePO4 batteries are affected by cold temperatures but not to the same extent that lead acid batteries are affected. I think if you were using an equivalent capacity lead acid pack the performance would be far worse than what you are getting with the LiFePO4 pack.

 

[Deepkimchi]

Bob,

I could have worded things nicer. I've been around construction types, probably picked up some bad habits.

BUT: Would like to know if you actually measured the surface temperature, or just calculated it. You just posted the temps on the graphs without any backup.

I don't sit on my cells. I do have a 40 A limit, do check my CA, and after running the bike rather hard for 20 minutes, I stop, open the cover up, touch the cells with my 98.6 hand, and they felt cool, consistently.

If you want to be treated like a scientist, you might omit the stuff like "fry an egg" on the cells, or talk about the "destroyed" cells with little smilies attached.

And yeah, maybe you're right about me being pissed... just pissed away $ 130 on the great Infineon controllers, then find out about a week later there's a problem with soft start. I give notice I want to buy some cells on 17 May, nobody says jack, I buy them, then you come out with the "great test" a week later with your little smilies.

Maybe they should rename the forum Endless Vendors...

 

[BMI]

Bob,

I could have worded things nicer. I've been around construction types, probably picked up some bad habits.

BUT: Would like to know if you actually measured the surface temperature, or just calculated it. You just posted the temps on the graphs without any backup.

I don't sit on my cells. I do have a 40 A limit, do check my CA, and after running the bike rather hard for 20 minutes, I stop, open the cover up, touch the cells with my 98.6 hand, and they felt cool, consistently.

If you want to be treated like a scientist, you might omit the stuff like "fry an egg" on the cells, or talk about the "destroyed" cells with little smilies attached.

And yeah, maybe you're right about me being pissed... just pissed away $ 130 on the great Infineon controllers, then find out about a week later there's a problem with soft start. I give notice I want to buy some cells on 17 May, nobody says jack, I buy them, then you come out with the "great test" a week later with your little smilies.

Maybe they should rename the forum Endless Vendors...

I am really sorry to hear about your Infineon controller problems Deepkimchi. Don't feel so bad about your green Lifebatt cells. They are far better cells than the majority of cells available. It is only my guess but I think your cells will show they provide considerably better performance than all the guys who are running Headway and Ping packs so don't feel so bad.
This is only my personal opinion based on my experience and knowledge so I may be proven to be wrong on this point but once Bob has a chance to test some other cells and not just the Green Lifebatt cells we will know the facts.

 

[liveforphysics]

I absolutely trust the data and graphs that Bob shares with us. Bob has been nothing but a first class guy, and a damn smart guy that has taught me a ton. I know things always look fishy anytime anyone goes from just being a product user to being a vendor. This always gets my "suspicious" flag waving as well. However, Bob is such an impeccable guy, and a throughly qualified guy, and a damn smart guy, I can still trust his data 100%.


I hate internal resistance. I personally choose to run high C-rate LiPo cells just because I hate Ri so much. I hate my pack's energy going into heating the pack rather than the motor, and I hate seeing the voltage drop under load.

A cell that has half the Ri of another cell is definitely not a product that can be directly compared, as they have very different application possibilities and limitations.

This gives me an idea...

Would the perfect LiFePO4 E-bike packs to make be low Ri grey cells for power, and then headways cells in parallel at the individual cell level for economical energy storage? Matching chemistries can certainly all be grouped in P at the cell level, and discharged and charged fine together, regardless of the Ri of the grouped cells.

Rather than $1.37/wh for just BMI cells, or $0.47/wh for Headways low power cells, you could make a hybrid pack with good power and capacity for $0.92/wh for 20ah packs, or $0.77/wh for 30ah packs.

I can understand from a vendor perspective how this would make a vendor cringe to think about mixing products like that. From an E-bike battery pack consumer perspective, it seems to be a more appealing product than either cell type singularly in a pack. Just something to think about

Best Wishes,
-Luke

 

[AndyH]

Welcome back, Bob.

What are the serial numbers for the green LB and grey cell? How many cycles on each?

thanks, my back is doing so great i am still a bit incredulous. i keep thinking i am going to wake up in pain again one morning; but until then i am trying to make up for lost time.

the green cell i am using for the charts is 4BF1K731-0320 and it has about 25 cycles on it, most of which were initial conditioning cycles to 40%. There seems to be some evidence now that discharging only to 40% for the first 30-50 cycles may provide maximum service life. If this turns out to be true it would be easy enough to build a device to condition cells before building them into packs. I have definitely found increase in capacity over the first 10 cycles or so, but my experience is limited to the very early green cells and this one later bmi grey cell. The other green cells i compared for impedance were in the same sequence from 309 to 460.

the bmi cell armin just sent me is 4BF10I13-1039, and that is one-oh-eye-one-three, a bit different format...

Bob,

I own cell 4BF1K812-0188. I purchased it more than a year ago from Yesa. This was a return - I replaced it under warranty when a customer broke a terminal. This cell, with a broken terminal, no nut on the terminal, some leaking electrolyte, and six cycles to 2.1V LVC at 5-20A doesn't heat or perform as poorly as the green LifeBatt cell you tested. You might want to seriously consider selecting another cell for comparison.

Andy

 

[dnmun]

i had measured a really high internal resistance on it too, in the note over on drbass's original yesa,psi, headway discharge thread. that was taken from your original post on that cell though, not recently.

 

[bobmcree]

Bob,

I could have worded things nicer. I've been around construction types, probably picked up some bad habits.

BUT: Would like to know if you actually measured the surface temperature, or just calculated it. You just posted the temps on the graphs without any backup.

no problem, my friend. i am an ex-sailor myself. i measured the cell temperature with an IR non-contact meter, partly because it is so easy, and partly because any other method has a significant Heisenberg uncertainty. ( When i use a very accurate lm34 or 35 to graph temp, it takes some heat out of the measured spot on the cell and sends it down the wire, so the manual measurements with my ir meter that average a fairly large and repeatable area are more accurate. The meter matches my calibrated lab mercury thermometer within 0.1F but even if it did not i am measuring the cells with the same meter.

I don't sit on my cells. I do have a 40 A limit, do check my CA, and after running the bike rather hard for 20 minutes, I stop, open the cover up, touch the cells with my 98.6 hand, and they felt cool, consistently.

If you want to be treated like a scientist, you might omit the stuff like "fry an egg" on the cells, or talk about the "destroyed" cells with little smilies attached.

My comment about sitting on your cells was a misunderstanding of your statement that the bike 'hauled your 200 lb ass around". The other comments were my attempt at humor. I try not to take myself too seriously. I do not want to be treated like a scientist, even though i worked as an engineer and metrologist for most of my professional life. I want to be treated as just another guy with an ebike, and that is why i do my best to validate any measurements i provide, rather than have anyone just accept them, and why i have not asked that my name be colored as an expert or mentor.

I will not pay West Mountain Radio $119 for software that would let me use their thermistor to record temperature, and i do not think it is necessary to measure it more often than every Ah which is what i did, and am content to do manually.

And yeah, maybe you're right about me being pissed... just pissed away $ 130 on the great Infineon controllers, then find out about a week later there's a problem with soft start. I give notice I want to buy some cells on 17 May, nobody says jack, I buy them, then you come out with the "great test" a week later with your little smilies.

Maybe they should rename the forum Endless Vendors...

I am not a vendor. I do not officially sell anything or represent any product. I did briefly try to represent Lifebatt USA some time ago, but nobody ever bought any cells from me and i only built one custom pack. I parted ways with them over a dispute on warranty coverage quite some time ago.

I do some custom work and repairs, and i will be happy to investigate the soft-start problem you mention even though i have nothing more to gain from doing so than hopefully contributing to helping you become another happy ebike user. Please understand that is my only goal, and the reason i try so hard to make my tests completely unbiased. I believe you will find many people here with the same attitude as mine, and yes you will find some vendors as well. You will also find that vendors who are a bit too liberal with the truth will be called out and do not last long.

The timing of my tests was based on when i received a BMI cell from armin for testing, and on the recent improvement in my ability to provide the data after a medical procedure that has reduced my back pain tremendously. You did not buy the cells from me, or i would have advised you that there might be better cells available. I expect that the person you did buy them from could not supply you with any better product, and hopefully they did not represent the cells to be anything other than what they are. I don't know what you paid for your green cells, but i expect it was significantly less than you might have paid for the bmi grey cells. I have stated several times that the green lifebatt cells are a good product; the bmi cells are just better.

You would not expect the Ford dealer to suggest that LandRover was a better product and send you away. You would not take your car back if they came out with a better engine before you bought it, and you had not done the homework to learn about it before you purchased, would you? Would you then blame AutoWeek for not reviewing the new engine sooner or making sure you knew about it? I am just trying to help my fellow ebike users find the best batteries, and to learn the differences between the products available. I am retired, and this is just part of my way of earning my social security benefits.

Apparently i was wrong when i said i thought the BMI cells were made in the same PSI factory as the Lifebatt cells. Armin has corrected my misinformation. If i were a shill for BMI i would probably have known better, don't you think?

 

[bobmcree]

Bob,

I own cell 4BF1K812-0188. I purchased it more than a year ago from Yesa. This was a return - I replaced it under warranty when a customer broke a terminal. This cell, with a broken terminal, no nut on the terminal, some leaking electrolyte, and six cycles to 2.1V LVC at 5-20A doesn't heat or perform as poorly as the green LifeBatt cell you tested. You might want to seriously consider selecting another cell for comparison.

Andy

let's see your data then, and your justification that it is probably as correct as mine. all you have done is say that it "doesn't heat or perform as poorly". i have tested several other green lifebatt cells, all which have leaked but otherwise seem ok, and the results are consistent.

i will be happy to provide my own unbiased data if you want to send me the cell, as i have offered to do for anyone. it would of course mean more if you provided your own data.

 

[bobmcree]

I absolutely trust the data and graphs that Bob shares with us. Bob has been nothing but a first class guy, and a damn smart guy that has taught me a ton. I know things always look fishy anytime anyone goes from just being a product user to being a vendor. This always gets my "suspicious" flag waving as well. However, Bob is such an impeccable guy, and a throughly qualified guy, and a damn smart guy, I can still trust his data 100%.

Shhh! you are going to get me kicked out of the Church of the Subgenius again. I was so foolish as to go off the path and join MENSA briefly, but i quickly got rid of those pesky extra brain cells and returned to the Church of Bob and the search for SLACK

i am not a battery vendor. i build a few custom packs and i will use the best batteries available whether i like the person selling them or not. While i appreciate the comments, i do not expect anybody to trust me. That is why i posted photos of a .005 ohm precision resistor and the .0025 and .0058 readings of the grey bmi cells and green lifebatt cells. of course the photos could be fakes, but i offer to repeat the tests on other peoples' cells and tests made with a different meter provide the same results.

it is no secret that don h. and i are not friends. that has nothing to do with who has the best batteries. i have said, and i will repeat, that i will happily test ANY cell from ANYBODY, as long as i am free to release the data. If i cannot release the data it becomes consulting and you will have to pay.

and definitely if you need the capacity using two strings of .005 ohm cells will greatly improve the performance under load, and may be an economical alternative. When you get down to cells with .002 ohms though, even the interconnect resistance becomes significant, so you will never get the same performance as you could from twice as many of the better cells, though it might be just fine. When the first batteryspace 10 Ah D cells came out at .008 ohms vs. the .005 spec for the SAFT cells that cost 2-3x as much, i ran the batteryspace cells in parallel with schottky diodes, and i still have a set of them running on one of my bikes that are 3 years old and doing fine.

 

[dnmun]

looking forward to your testing the headway cells in the group buy. you use a different technique from the way i do it following justin's rule, delta voltage for delta current pushed out of the battery. i still think it would be neat to characterize the impedance using your technique and changing the frequency as well as the voltage of your source, to see if there is point where charge impedance and discharge impedance are equal, seems like an interesting region of the battery behavior. and the different electrolytes may show different properties at different frequencies too, as well as temperature. just curious.

 

[bobmcree]

looking forward to your testing the headway cells in the group buy. you use a different technique from the way i do it following justin's rule, delta voltage for delta current pushed out of the battery. i still think it would be neat to characterize the impedance using your technique and changing the frequency as well as the voltage of your source, to see if there is point where charge impedance and discharge impedance are equal, seems like an interesting region of the battery behavior. and the different electrolytes may show different properties at different frequencies too, as well as temperature. just curious.

justin's suggestion to measure the voltage drop at two different currents and use the difference is of course the same way i would do it to measure dc cell impeance. the problem with this technique is that the cell voltage is dynamic, and how soon after you change the current to you measure? the indurstry standard is to do it at 1 kHz and i have always found the impedance at 1 kHz to be very much a predictor of the dc performance.

i am packing up my truck now so this will be the last post for a day or two, but the bmi cell is cycling twice a day and will be well broken in when i return in about a week.

As i said in an earlier post, i do not represent any cell vendor and will be happy to make packs for people out of any cells they want. it helps me feel like i am earning my social security to perform this service for my fellow ebikers, just as it makes me feel useful to be involved as i am in the fight for access to health care for every american. i urge anyone who has not spotted that thread in general discussion to give me your ideas so we can use them to fight the lies being spread by the opponents of health care for every american.

 

[AndyH]

Bob,

I own cell 4BF1K812-0188. I purchased it more than a year ago from Yesa. This was a return - I replaced it under warranty when a customer broke a terminal. This cell, with a broken terminal, no nut on the terminal, some leaking electrolyte, and six cycles to 2.1V LVC at 5-20A doesn't heat or perform as poorly as the green LifeBatt cell you tested. You might want to seriously consider selecting another cell for comparison.

Andy

let's see your data then, and your justification that it is probably as correct as mine. all you have done is say that it "doesn't heat or perform as poorly". i have tested several other green lifebatt cells, all which have leaked but otherwise seem ok, and the results are consistent.

i will be happy to provide my own unbiased data if you want to send me the cell, as i have offered to do for anyone. it would of course mean more if you provided your own data.

Things were happening here while you were in the body shop. I've provided all my data so far in this thread, as well as here. DocBass has done some excellent work as well here.

I'll happily provide more once my equipment comes back from the calibration shop. In the mean time, these three posts seem germain.

Andy

 

[bobmcree]

Things were happening here while you were in the body shop. I've provided all my data so far in this thread, as well as here. DocBass has done some excellent work as well here.

I'll happily provide more once my equipment comes back from the calibration shop. In the mean time, these three posts seem germain.

Andy

the link you provided does not show a graph i could find of a lifebatt cell at 25A and i did not see any temperature data on a cell discharged at 25A for its full capacity. I have not been "in the body shop" but i admit i do not have time to wade through every test. you challenged my specific data of a lifebatt cell at 25A, so unless you have the same test conditions in one of your tests and it shows much different data than mine i don't see what you are arguing about. i did see some curves at lower currents, and some curves that show no current on the cba. if you are going to challenge my results let's compare apples to apples and let's see your 25A data on a green lifebatt cell from the same serial number sequence.

the three posts you say are "germain" have nothing in them that contradicts my results, and pardon me if i do not read the one from someone who has been proven to be a liar. If Don would agree to release gary from his NDA and allow him to tell the truth about our warranty disagreement we might ALL know the truth at last. of course gary cannot say anything without don doing this. i have agreed to drop the issue as it gets threads locked and everyone loses.

my data on cell impedance shows that the bmi cells will always produce a higher output voltage at any current than the green lifebatt cells. Nobody has disputed that lifebatt green cells range from 5-6 milliohms and bmi grey cells range from 2-3. this means at 100A or 10C (which is the spec lifebatt dropped to after deciding their original 20c was not real), that a single bmi grey cell will have .3v higher output voltage and the green lifebatt cell will have an extra 30w of heat dissipation INSIDE the cell. Whether you believe my other tests or not, these results have not been disputed. Do the math and it is easy to see that the green lifebatt cells will be much hotter than the grey bmi cells after a 25A discharge, and i am not sure but i expect that i CAN fry an egg on a green lifebatt cell discharged at 10C for 6 minutes, and i will post a photo of the egg frying. This is of course not a valid scientific comparison, but it will be fun.

if you want to dispute my data of the one lifebatt cell tested at 25A compared with the bmi cell at the same current and temperature, all you need to do is post a graph of that data, not a bunch of irrelevant links to a whole library of tests nobody is going to wade through to find the one test that i could not find when i did take the time to look.

i do not doubt anything you have posted, but you are disputing my data without anything but opinions as a basis. I suggest that when you get your equipment back you test the green cell then we exchange cells and test each other's. if our data is less than 4% different there is nothing to argue about.

 

[Battboy]

Yay - We Qualified. Not bad for a College built bike on a small budget! http://www.wired.com/autopia/2009/06/electric-bikes/

No, there are no "Green Cells" in this bike. We haven't made those Cells in over a year. There is a whole new generation of XPS Cells now being offered in our new XPS Series Packs.