Lithium Battery Reports & Tests' 9/08

[dnmun]

11 yrs is nothing, my ford outside is 38 years old now. i finally unloaded the 4500 lbs of gravel which had been sitting in the back end for 2 months until the rain stopped. what a relief.

is that the original lifebatt cell you got with gary last year? i thought i had pulled 12mohm from your original discharge data on doc's discharge thread.

selling packs is gonna be a cutthroat business. i can see that headway is gonna try to see how many they can sell over here, alotta people seem to be offered deals recently. i do think they will perform well for longer than the 3 years, we'll see.

maybe jason will get you one of the prismatic cells in the pack that gaston reviewed, and doc may have, or andy can supply, the psi cell to compare to the headway. and also the yesa cell. would be neat to see different testing corroborate the data.

 

[Nimbuzz]

Dr Bob, Even though some think it is a tough business I think quality, dependable Lithium packs made by someone in this country who has done their homework and will stand behind them with a warranty will be who I buy from. I've already spent $thousands$ on Lithium packs that were touted by supposedly reliable engineers. All of these packs died in a year and there was no recourse across oceans with non-English speaking companies. If your marketing can show your thorough testing and if it is backed up by a warranty like Optibike has (3 years!) then I hope you will have a market -- of at least us guys here.

 

[BMI]

Dr Bob, Even though some think it is a tough business I think quality, dependable Lithium packs made by someone in this country who has done their homework and will stand behind them with a warranty will be who I buy from. I've already spent $thousands$ on Lithium packs that were touted by supposedly reliable engineers. All of these packs died in a year and there was no recourse across oceans with non-English speaking companies. If your marketing can show your thorough testing and if it is backed up by a warranty like Optibike has (3 years!) then I hope you will have a market -- of at least us guys here.

Nimbuzz I am really sorry to hear of all the money you have spent on those failed lithium packs which only lasted a year.

I will add my two cents worth here. I have always believed it is better to pay more for quality up front rather than have premature failures with no recourse from the manufacturer.
I wan't aware of the Optibike warranty. I wonder what the conditions are for this warranty and what cells are used in their packs? Anyway BMI factory battery packs when supplied with proper BMI charger (fully automatic dedicated LiFePO4 quality charger with protection from reverse polarity, over voltage, over current and over temperature) come with the same warranty of full three years for EV applications and five years for standby power type applications or 3000 full cycles (whichever comes first). I will actually honour this warranty if certain other brands of chargers are used but I only approve this on a case by case basis. For exapmle I provide a full 5 years warranty to people who use the packs for solar power applications. In this case only solar panels are used for charging so this is an application where no mains charger is used at all yet the five year warranty still applies.
I believe this is the best warranty offered any battery manufacturer at the present time?

Of course the other factor to consider is that you are not dealing with a representative in China who may or may not have a clear understanding of the English language. You are dealing with someone who is from a western country and understands the service expectations and ideals demanded by westerners.

If BMI cells weren't premium quality and didn't provide the highest performance (other than an A123 cells which are smaller so can be expected to have a higher discharge rate for the same equivalent capacity) they would not of been the LiFePO4 cell of choice by the European manufacturer of the world's fastest and most powerful production e-bike (the "el-cycle"). A top speed, not speed limited as is the case with the production bikes so they can legally be riden on the road,of close to 100kph (60mph) from only a 36V pack is not bad performance in anyones language I would of thought. Just imagine what it would do with a 48V or 72V pack.
Specs on the el-cycle are here- http://www.evalbum.com/2295

 

[Deepkimchi]

That graph seems like a crock of s---. :?

Been running those green Lifebutt cells at 20-30 A for 15-20 min, felt cool to the touch.

We don't have a lot of hot air in Korea right now like other places, maybe that's it...the air cooling..

 

[recumbent]

That graph seems like a crock of s---. :?

Been running those green Lifebutt cells at 20-30 A for 15-20 min, felt cool to the touch.

Somethings definately not right, I run my green PSI cells at 40amps peak, 30amps for many minutes fully loaded bike, plus my 200 lb arse and never warm. And thats concealed in a pannier bag to boot. Maybe you have a dud cell to test.

 

[BMI]

That graph seems like a crock of s---. :?

Been running those green Lifebutt cells at 20-30 A for 15-20 min, felt cool to the touch.

Somethings definately not right, I run my green PSI cells at 40amps peak, 30amps for many minutes fully loaded bike, plus my 200 lb arse and never warm. And thats concealed in a pannier bag to boot. Maybe you have a dud cell to test.

Maybe Bob needs to do the same test again with a different green cell to see if the same result is recorded. Perhaps there is some problem with the green cell which was used?

I would of expected similar heating at low discharge rates and only see a marked difference at high rates of discharge.

 

[bobmcree]

Bob,
on your chart, why starting voltages are not the same?
At 0Ah used green cell start well under 3.3V.
Why?
MC

the voltages are not the same because the impedance of the green cell is over twice as high, which means that there will be more loss inside the cell leading to lower voltage out to the terminals.

i will be happy to repeat the tests with a couple of other green cells i have, but if you go back to the graph i published when i briefly tried to sell the green lifebatt cells you will see that the graphs are very close.

i just do not have time to collect the data tonight because i am preparing to host a kickoff meeting for Organizing for America to begin the fight for access to affordable health care for every american. this is just a bit higher priority tonight but i promise tomorrow to have 50 and 75 amp discharge data on the green and grey cells, and maybe 100 if i can get the resistance of my test leads down low enough. i built a very simple programmable load, using a .010 ohm 1% resistor, the heat sink from a sun workstation, and the transformer and lcd are the same as the ones that i put in my power supply kits. Three 4110's will handle up to 1000W and i think the heat sink will handle that much for a few minutes, i have only taken it up to 75A because my test leds started to melt

as for the temperature rise, it is obvious that if the green cell has twice the impedance there will be twice as much heat generated. i would point out that when you use the cells on your bike you do not discharge them at a steady 25A for the whole capacity, but rather give plenty of time for cooling. Independent of any measurement errors, if you look at the area between the two curves, where do you think all that energy is going? It is not producing any x-rays or other high energy photons i can detect, so the only alternative is for it to turn into HEAT. .005 ohms for the lifebatt green cell and .0025 for the grey bmi cell means twice as much heat at the same current.

this increase in internal heating is quite obvious at higher currents. I have finished testing my high current load, so in the next couple of days i promise more curves with more complete data.

As for Deepkimchi's comments that my graph must be invalid because his cells are "cool to the touch" after "20-30A for 15-20 mins". i submit you are the one whose results are a crock of whatever. My tests are based on repeatable measurements made with equipment with traceable calibration. I reported the starting and ending temperatures, the room temperature, and provided a curve which i will stand behind any day, whereas your opinions are based on the temperature of your ass and you have no real data at all. Sorry if i do not believe your measurements are as valid as mine. If you want to send me one of your cells i will be happy to provide unbiased test data, as i will for anybody who wants to send me a cell for comparison. I am sorry that because you have invested in a bunch of green cells that you feel threatened by data that proves the grey bmi cells superior, but life is tough sometimes, better get used to it because your green cells will look a lot worse at 50 amps in a day or two. They are still a good product, there is just a better one out now and yours just will not be as good no matter how many times you insult me. Buy some test equipment and prove me wrong before you resort to infantile name-calling, huh?

 

[BatteryMooch]

[i just do not have time to collect the data tonight because i am preparing to host a kickoff meeting for Organizing for America to begin the fight for access to affordable health care for every american. this is just a bit higher priority tonight but i promise tomorrow to have 50 and 75 amp discharge data on the green and grey cells, and maybe 100 if i can get the resistance of my test leads down low enough. i built a very simple programmable load, using a .010 ohm 1% resistor, the heat sink from a sun workstation, and the transformer and lcd are the same as the ones that i put in my power supply kits. Three 4110's will handle up to 1000W and i think the heat sink will handle that much for a few minutes, i have only taken it up to 75A because my test leds started to melt

Watch out at those higher power levels, the IRFB4110 is spec'd for over 300W only for short pulses or infinite "ideal" heatsinks.

With the 4110 having a theta-js of 0.9 degrees-C/W, even with an infinite heatsink and a dangerous 150C junction temperature rise you can only have each FET handling 166W (150C / 0.9 degrees-C/W = 166W) for a total of about 500W for the three FETs. But, that's for a heat sink that doesn't exist.

Assuming a good 0.3 degree-C/W theta-sa heat sink, 25C ambient temp., three FETs, and a 150C max temp rise, you get the following numbers...
(0.9 degrees-C/W theta-js per FET) / (3 FETs) = 0.3 degrees-C/W theta-js
(0.3 theta-js) + (0.3 theta-sa) = 0.6 theta-ja
(150C rise) / (0.6 theta-ja) = 250W to reach 175C junction temeprature

250W will bring your three FETs to their maximum rated junction temperature. And that's a dangerously high temperature. It should be held to a 110C rise, at the most. That will bring you back down to 183W.

If you have an absolutely, incredibly efficient sink, then (150C rise) / (0.45 theta-ja) = 333W at 175C junction temp. and 244W at 145C (110C rise).

The FETs might last for months (heck, even longer) or blow in 2 minutes. But, if you can, I'd recommend another heat sink and three additional FETs for loads greater than 200W or so. But, the SOA graph for the IRFB4110 shows that at 4V, each FET is good up to about 85A before exceeding its rating when used as a load (DC curve) . If they don't burn up, they can safely handle very, very high current levels.

Looking forward to your test results and I hope the kickoff meeting went well!

 

[bobmcree]

i have finished printing all the forms and info packets i will need tomorrow, so i thought i would check back in. i designed the load with the intent of using it mainly to test single lifepo4 cells at up to 100A, so at that level there will be 100W in the 100w kelvin resistor, which is fan cooled, and if the cell drops to 3.0v that leaves just another 200w in the 3 fets. The heat sink has liquid filled tubes that conduct the heat across it, and the fan is a monster at about 7W. With 10 Ah cells the system will only need to handle 100A for 6 minutes and i think it will do that easily, and the internal wiring is silicone or brass strips, so the fets can handle the current down to pretty low voltage. If i want to measure higher voltages i will just add series resistance to handle most of the power and let the load regulate the necessary few volts.

Of course i do not expect the 3 4110s to handle 1000w for any length of time. i thought about using some scrap parts i pulled out of controllers for upgrades, but the 4110 has much better thermal conductivity from the die to the package than most of the other devices, and i have several hundred in stock. If i were designing the load from scratch i would probably use to-247 package devices. As it was i used 3 because there was room on the heat sink, and i used 10 ohm gate resistors just to work as fuses if one shorts and to make it easier to isolate a failed part.

i probably should not have resorted to saying deepkimchee's test results were based on the temp of his derriere, but i take my measurements very seriously and i was personally insulted when he made the claim that the charts i took the time to publish were not valid while presenting no data of his own. I have worked as a metrologist in medical manufacturing operations and set up NIST traceable calibration labs all over the world, so i know how to make these measurements, and to insinuate as he did that i would deliberately publish bogus data hurt my feelings and made me angry. Anyone with a CBA2 and a green lifebatt cell can reproduce my results at 25 amps. If someone at Lifebatt wants to send me a cell i will provide the same level of testing as i have for the bmi cell and for the Lifebatt cells i still have.

 

[BatteryMooch]

i have finished printing all the forms and info packets i will need tomorrow, so i thought i would check back in. i designed the load with the intent of using it mainly to test single lifepo4 cells at up to 100A, so at that level there will be 100W in the 100w kelvin resistor, which is fan cooled, and if the cell drops to 3.0v that leaves just another 200w in the 3 fets. The heat sink has liquid filled tubes that conduct the heat across it, and the fan is a monster at about 7W. With 10 Ah cells the system will only need to handle 100A for 6 minutes and i think it will do that easily, and the internal wiring is silicone or brass strips, so the fets can handle the current down to pretty low voltage. If i want to measure higher voltages i will just add series resistance to handle most of the power and let the load regulate the necessary few volts.

Ahh...the resistor changes everything.
Sounds like a nice design, looking forward to the results!

 

[Nimbuzz]

Bob, Great tests! Just what we need.

-- Are there any tests that can tell us cycle life or does a low internal impedance already suggest a high cycle life?

-- How consistent would you expect the quality of the grey cells to be in different batches?

-- Looking forward to your pack offering -- time line?

Thanks for your efforts and info.

 

[miro13car]

Bob,
as usual your excellent explaination why green cell is hotter and why they don`t start from the same voltages.
Obviously resting voltage of green one must be lower than grey one.
it is simple laws of electricity.
Every cell can be imagined as actual cell plus internal resistance connected in series.
And now is cristal clear why green one is warmer and why its resting voltage is lower.
In my LB/BMI HPS pack my grey LB/BMI cells /12 0f them/ read all 3.4V if perfectly balanced.
I imagine in 12 cell pack of green they would read more like 3.2V if perfectly balanced?
BTW my grey cells served my well for over 2500KM of riding now and still delivering solid 11Ah at I estimate 10A average all summer, less in winter time. They just love heat, but every degree drop in temp affect them.
MC
MC

 

[miro13car]

And forgot to add :
to TF Google forum :
there is no Lithium battery available for our bikes?
Wake up and start reading, you are still laggin in Nickel chemistry era!!!!
MC

 

[Nimbuzz]

Ok Mr Miro -- Please tell us of the proven Lithium packs for our bikes. Bob here is finally defining a good cell. Before he did this everyone was squabbling about grey and green and various sellers' claims and on and on. This is awesome that Bob has cleared much up but only for cells not packs -- yet. Many of us on the Google Tidalforce forum spent $thousands$ on Lithium several years ago from established ebike companies run by Electrical Engineers. All of these packs died in a year or so often at less than 1,000 miles. We've been in the 'Lithium Era' longer than most, are now cautious AND require warranties or proven track records. When I bought my first Point One lithium pack in 2006 the majority here at Endless Sphere were running SLAs -- perhaps because they were too wise to be the ginea pigs that we were (but I don't mind being an experimenter on the cutting edge -- just don't want to keep doing it) The big hubs are drawing 30+ amps and the regen from a big hub is also hard on packs. The robust packs needed for a Tidalforce X are similar in quality to what is needed for electric cars and are just now emerging. Praise the Lord and pass the proven Lithium! Go Bob Go!!

Where might one buy a 36 v grey cell BMI pack right now -- preferably with a substantial warranty?? My Google searches come up empty handed.

 

[bobmcree]

Bob,
as usual your excellent explaination why green cell is hotter and why they don`t start from the same voltages.
Obviously resting voltage of green one must be lower than grey one.
it is simple laws of electricity.
Every cell can be imagined as actual cell plus internal resistance connected in series.
And now is cristal clear why green one is warmer and why its resting voltage is lower.

close but no cigar just yet. (this is an old expression from carnivals where if you guessed the guy's weight closely enough you won a cheap cigar, always worth less than the cost of guessing in my experience

it does not follow necessarily that a cell would have lower resting voltage if it had higher impedance, unless you use a very very cheap meter. If you use a modern meter with a typical 10 meg input and your cell is resting at 3.4v there will be .34 microamps drawn by the meter, or 3.4 e-5 amps, so for every milliohm in the cell the voltage will drop by 3.4 e-8v which you cannot measure. Even with the cheapest meter you can find which is 1000 ohms per volt you would have 1 ma current drain so for every milliohm of internal cell impedance you will see 1 microvolt of voltage drop, which you could measure, but not with that meter.

the significant voltage drop due to cell internal impedance only happens at higher currents. I just finished testing several of the green lifebatt cells, and the ones that are not destroyed are 5.5 to 5.8 milliohms which is what i measured originally over a year ago and within the Lifebatt spec. I only have one bmi grey cell, and it measures 2.2-2.5 milliohms over its full discharge range, making it half the Lifebatt cell. So at 1 amp you will see 5.5 mv drop in the lifebatt cell and 2.5 mv in the bmi cell. In a pack of 16 cells this is still only 90mv and 40mv, but at 25A it becomes 25x greater, at 50A you would expect to see over 4v drop in the lifebatt pack and about 2v in the bmi pack. This equates to 200w of heat in the lifebatt pack and 100w in the bmi pack.

if when they make the cells it is like the a123 cells, there is a "jelly roll" of plate and electrolyte, and there are conductor strips at several points along the jelly roll, which is then rolled up and the tabs welded together. the difference in impedance may be as simple as more conductor strips in the bmi cells. more contact area from the guts to the outside world produces lower impedance. i do not think this would affect the resting voltage of the cell significantly. differences in the formulation of the powder are the most likely reasons the cells would have different voltages when no current is flowing. I hope this makes it clear, if not i will be happy to come up with some more explanation.

My impedance measurements were made with my HP 4328a meter. This 35 year old marvel is as accurate as when it was new, and a joy to work with. The circuitry is fairly simple, but the accuracy and reliability comes from special transformers, special transistors, a world class meter movement, and it is totally state of the art for its day, which is to say 2% accurate at .001 ohms and with 20 micro-ohm resolution. You can take an inch long piece of copper wire and connect it to the probes then measure the change in resistance due to heating when you breathe on it. close enough for our purposes even if not digital.

i used the same scale (.010 ohms max) for 3 measurements to avoid ranging errors. i measured the bmi cell, the lifebatt cell i used for the 25A curve, and several others which were about the same, then i measured a .005 ohm 1% resistor to validate the measurements. the pictures speak for themselves.

View attachment one jaw supplies voltage the other measures it.jpg
View attachment bmi .0025 ohms @ i kHz.jpg

View attachment 1

 

[bobmcree]

Where might one buy a 36 v grey cell BMI pack right now -- preferably with a substantial warranty?? My Google searches come up empty handed.

you can ask Armin (bmi) about the packs; i am not aware of any. i am working up a business plan right now to try and get financing to bring in a batch of cells at a great price he has offered me to get started, and build them into packs. money is not easy to come by these days, so no promises. I agreed my initial order would not be a bulk cell order for a bunch of guys, but anybody who wants to invest in the venture can contact me and see what we can work out, i doubt it will happen before the end of summer.

if you want to buy the cells from Armin we could discuss me building them into a pack for you, and between us we probably could supply a warranty for the right customers. it is really hard to know someone is not going to swing their batteries around their head and use them to smash small animals or almost anything else one can imagine. I have 6 of the original lifebatt 4 cell modules that would take that kind of abuse and more, but they are too big and too heavy. other than that they are great. see photos posted long ago.

 

[AndyH]

Welcome back, Bob.

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

 

[BMI]

Ok Mr Miro -- Please tell us of the proven Lithium packs for our bikes. Bob here is finally defining a good cell. Before he did this everyone was squabbling about grey and green and various sellers' claims and on and on. This is awesome that Bob has cleared much up but only for cells not packs -- yet. Many of us on the Google Tidalforce forum spent $thousands$ on Lithium several years ago from established ebike companies run by Electrical Engineers. All of these packs died in a year or so often at less than 1,000 miles. We've been in the 'Lithium Era' longer than most, are now cautious AND require warranties or proven track records. When I bought my first Point One lithium pack in 2006 the majority here at Endless Sphere were running SLAs -- perhaps because they were too wise to be the ginea pigs that we were (but I don't mind being an experimenter on the cutting edge -- just don't want to keep doing it) The big hubs are drawing 30+ amps and the regen from a big hub is also hard on packs. The robust packs needed for a Tidalforce X are similar in quality to what is needed for electric cars and are just now emerging. Praise the Lord and pass the proven Lithium! Go Bob Go!!

Where might one buy a 36 v grey cell BMI pack right now -- preferably with a substantial warranty?? My Google searches come up empty handed.

Where else would you purchase any BMI battery products but from BMI (whether they be loose cells, complete battery packs or chargers)!
The BMI battery comes with full 3 years or 3000 cycle life warranty (whichever comes first) so long as the battery is purchased with a proper BMI battery charger which will switch off automatically so the battery can never be over charged if you forget to switch it off in time as can happen with a manual charger.

I am expecting to have the new full range of battery packs with specifications in PDF on the BMI website by the end of this week but at the moment there is general info on the website on the battery pack page here-
http://lithbattoz.com.au/index.php?page=battey-packs

Of course the warranty doesn't apply for acts of stupidity such as droping it or otherwise physically damaging it. Also it must not be excessively over discharged past OLV (over low volts) at which time the battery will warn you by "beeping" and a red warning LED will light up on the front of the battery.
If properly looked after a BMI battery will provide several years of reliable, trouble free use as many people from all over the world will confirm the reliability. What's more, when you buy a BMI battery pack you are getting what you pay for so if you buy a 10Ah pack you know you are getting AT LEAST 10Ah of useable capacity. If you buy a 20Ah pack you are getting AT LEAST 20Ah of useable capacity. We don't exagerate out claims as do many of the Chinese battery maufactrurers.

I supply the same batteries to the plug in hybrid electric car market and occassionaly offer them on ebay for this purpose. Feel free to check out my ebay advert and in particular check out my detailed feedback and what others have to say about me. My 100% positive feedback rating didn't happen by accident and is a result of a great deal of continued hard work in keeping customers informed and staying up till after midnight most nights relpying in detail to the many emails I receive from all over the world. I always treat others the way I would like to be treated.

http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&item=280351225061&sspagename=STRK%3AMESELX%3AIT&viewitem=

I regularly receive phone calls from all over the world and you can always give me a call to solve any problems (which rarely occur) and which can always be resolved in a quick and simple manner and since I speak good english so communication problems are non existant. I am not in the habit of lying or being dishonest in any way to my customers. Why would I, since I am a customer too and if I purchase any product from a seller I too like to be treated with the same honesty and respect just like you do.

 

[bobmcree]

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.

i am just setting up now to run some curves at 50A and higher, where i expect to see even more advantage to the grey cells. I think building packs with the PSI 'lego' blocks is the easiest way to put them together into packs, and i am increasingly in favor of a multi-output charger that charges each cell fully each time and a lvc only bms on the pack. Anything else is an unnecessary compromise. I have a charger on my bench now built from 12 2A switchers i got from armin, and each has a led that shows full charge. Most people don't know that pack mounted "balancing" bms systems can only improve pack balance by a small amount each charge cycle, and that if a pack is out of balance by more than 1 Ah it may never get rebalanced. In order for balancing to work charge cycle time must be extended. With a dedicated switcher for each cell pack charging can be completed in less than an hour with every cell perfectly charged, then a cell-level low voltage cutoff will prevent any big surprises but likely will never be needed unless a cell fails prematurely.

Surprisingly (to me anyway), a couple of the lifebatt cells that were discharged to <1v by the defective vms eventually recovered enough to accept a charge. A couple of them seem good as new, but one has doubled in impedance. Even so with cells approaching $50 each a dedicated lvc on each cell seems like a good investment, then coupled with a multi-pin connector the pack bms needs to be just the voltage detectors and optos or other combining circuitry, avoiding most of the parts that cause bms failures on the road. This is the approach i am taking with the lifepo4 packs i am building now; with either the 2A modules which are 110/220v in or some 20A modules that require 48v.

i am glad to hear armin will have packs listed soon. i can only do a few and usually they are custom designs, and i am not counting on having anything for this bike season at all. it matters not that armin is in oz and the batteries in taiwan, he can still serve us better than somebody in las vegas or where is it now?

 

[BMI]

I decided to perform a quick internal impedance test from my most recent shipment of BMI cells which arrived only last week. I thought it would be good to see how my quick test measurments taken with an impedance test meter would compare with Bob's results using his much more sophisticated and no doubt more accurate testing rig.

My first reading was to measure an A123 cell to see how close the reading as measured by my inexpensive impedance test meter came to the published A123 specifications for an A123 M1 cell. As you can see in the photo the A123 cell is brand new and does not have any spot weld marks on it. The specification for a new A123 cell says the internal impedance should be "8 milliohms typical".



As you can see my reading of 8.91 milliohms seems pretty close to specification so my test meter musn't be too far out with regard to accuracy.



Then I measured all of the cells in a tray of cells which arrived from the factory only last week. All cells in this tray measured out at under 2.8 milliohms and most were in the region of 2-2.5 milliohms which is pretty typical of my readings which I take from random shipments of cells. I am yet to see a cell measure out at greater than 3 milliohms and ocassionally a cell will measure out at just under 2 milliohms.



In some instances I supply certain special customers with "matched cells" for racing applications which measure out at exactly the same internal impedance. Even without this matching of cells the cell quality is incredibly consistent from one shipment to the next.

 

[bobmcree]

my higher current discharge curves will be delayed about a week. my dad is in poor health and i am going down to california to take him to a new doc and help with some other issues. i will run the higher current tests as soon as i return.

thanks to armin for publishing the impedance data. wouldn't it be nice if other vendors of the 40138 cells would report on their latest products? i'm sure don is reading this.

 

[RoughRider]

a question?

this graph from bob...do i read it right...the cells do NOT deliver the full 10Ah??

but BMI says on the webside that the cells will deliver the full Ah and a bit more...


i dont think, that is makes sense to buy such expensive cells...

even WHEN they can do the 3000cycles...what do i need it for?? what do i need 3000cycles for??

what is the price of one cell?? 40$?? 3,2V+10Ah= 32Wh = 0,8Wh/$...

thundersky / sky energy: 44$ for 3,2*40AH= 128Wh ==> 2,9Wh/$....more than 3 time of BMI

lets think, they last for 2000cycles...at least 1500cycles...

so, that means, at least 5years or even longer....

EXAMPLE OF A PACK
lets see for a 48V 40Ah pack:

BMI: 60cells => 2400$ only for the cells
SE / TS: 15cells: 660$

i save 1740$

BMS comes on top...more cells for BMI means more expensive BMS ... price on ebay from BMI for 48V and 10Ah with BMS: 990$...so you will pay near 4000$ for the pack of 40Ah...and TS / SE pack less than 1000$


in 5years when my TS/SE-pack is down at 80%...i can buy a new one, maybe with new / better cells and the prices will be down at maybe half of the price today because of mass production...

so buying the BMI cells means, to have very much of dead capital...

better put your money to the bank and let it work for you at 4% in 5 years it will be over 700$ and that means, that you got your batterypack from TS or SE cells FOR FREE...compare to BMI

think about that...

 

[dnmun]

WOW!

 

[Ypedal]

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....

 

[BMI]

a question?

this graph from bob...do i read it right...the cells do NOT deliver the full 10Ah??

but BMI says on the webside that the cells will deliver the full Ah and a bit more...


i dont think, that is makes sense to buy such expensive cells...

even WHEN they can do the 3000cycles...what do i need it for?? what do i need 3000cycles for??

what is the price of one cell?? 40$?? 3,2V+10Ah= 32Wh = 0,8Wh/$...

thundersky / sky energy: 44$ for 3,2*40AH= 128Wh ==> 2,9Wh/$....more than 3 time of BMI

lets think, they last for 2000cycles...at least 1500cycles...

so, that means, at least 5years or even longer....

EXAMPLE OF A PACK
lets see for a 48V 40Ah pack:

BMI: 60cells => 2400$ only for the cells
SE / TS: 15cells: 660$

i save 1740$

BMS comes on top...more cells for BMI means more expensive BMS ... price on ebay from BMI for 48V and 10Ah with BMS: 990$...so you will pay near 4000$ for the pack of 40Ah...and TS / SE pack less than 1000$


in 5years when my TS/SE-pack is down at 80%...i can buy a new one, maybe with new / better cells and the prices will be down at maybe half of the price today because of mass production...

so buying the BMI cells means, to have very much of dead capital...

better put your money to the bank and let it work for you at 4% in 5 years it will be over 700$ and that means, that you got your batterypack from TS or SE cells FOR FREE...compare to BMI

think about that...

BMI cells are true 10Ah rated cells. Please remember that the cell under test is brand new and will slightly increase its capacity once it is cycles a few times. LiFePO4 cells also gain capacity in warm weather.
Speak to people who are actually using the batteries and have REAL LIFE experience with them as to their capacity test results.
micro13car has said his pack delivers 11Ah and this is just one of many examples since micro13car brought this up himself earlier in this thread.

There are many differences between Thundersky and BMI batteries. TS are a low discharge rate cell which are rated at less than 5C while BMI cells will provide more than double this continously and 20C peaks. Therefore you need an extremely large TS battery to be able to gain enough useable capacity which is why they are limited to car conversions.

I am yet to see Thundersky batteries being used to power an e-bike or e-motorcycle but please correct me if I am wrong!

They are vastly different technologies in the battery management area. I have personally been told by a professional EV conversion company they don't like the TS BMS with the loose wires on the top of the battery and the fact the batteries must be installed upright.
This is not the case with BMI batteries. They can be installed in any orientation at all including upside down.
The VMS is far more sophisticated and has many advanced features which are far ahead of what TS uses in their BMS. The VMS is concealed inside the battery so there are no loose exposed wires as is the case with the Thundersky batteries

Does a TS battery have the ability to perform an automatic test of every cell in a large bank of batteries and perform this test automatically up to twice a day as can a BMI battery? Then if any cell in any individual battery is found to be out of specification or in some way requires attention or checking the owner of the battery can be sent a SMS text message to his phone or an email sent to him alerting him to check the battery in his electric vehicle and indicating the exact cell in the battery which requires attention?

The fact is the Chinese battery companies just don't have the technology yet to be able to manufacture very high quality LiFePO4 batteries. Unless you see how the cells are manufactured in the BMI factory you would not have any understanding of the complexity and why all cells are not the same (and why quality cells like BMI cost more).
This has resulted in the unfortunate case which I see only too often where people have outlayed many hundereds (if not thousands)of dollars on lithium battery packs from China only to have them prematurely fail.
Of course this has resulted in so many people being wary of LiFePO4 batteries. It is extremely difficult (from my point of view) to regain the confidence of people who are wary due to their bad expriences. Perhaps you fall into this category of people RoughRider which is why you appear to be so negative? I perfectly understand this and it is human nature to buy the "cheapest" of anything. But then when you do buy the "cheapest" and you find that it doesn't perform as expected and/or fails prematurely you are left with a sour taste in your mouth.
This is why I always say when it comes to LiFePO4 batteries to buy as good a quality as you can afford. Of course I totally accept not everyone can afford top quality LiFePO4 batteries but if you can afford them buy them. The principal objective should be to buy on quality and not on price...... unless you want to risk being left with a sour taste in your mouth.

The manufacturers warranty is important in showing the customer the manufacturer stands behind their product. This is why the 3-5 years/3000 cycle life warranty offered by BMI is to my knowledge the longest offered by any LiFePO4 battery manufacturer.
I have often wondered why Thundersky aren't prepared to stand behind their product and provide a similar manfacturers warranty? The 12 months only warranty seems inadequate for such an expensive outlay of batteries as required to fit out a full size EV conversion. From what I understand the manufacturer (Thundersky) does't provide any warranty whatsoever. The warranty offered is provided by the distributor in the country of sale because they are required by law to provide some kind of warranty which is the case in many western countries. I am talking about specifically in Australia here and the situation may be different in other western countries which understandibly I would not be familiar with.

I was recently made aware by a professional EV conversion company of a case where a car fitted with Thundersky batteries caught on fire in the owners garage while charging. The car was purchased from a professional EV conversion company so it was no "backyard job". The woman from Melbourne was charging her car overnight in her garage and it caught on fire somtime while it was on charge. It is not clear if the fire started in the batteries, BMS or charger but the fact is this was a TS supplied installation.
I am not aware of any case of a similar fire involving BMI batteries in any installation. If we take a worst case scenario and the BMI batteries were involved in a fire (perhaps such as if the car was involved in an accident) any such flames would soon self extinguish due to the flame retardant DuPont Xytel battery cases which BMI batteries are manufactured from.

It is the much high quality materials which go into the manufacture of BMI batteries which is why they cost more so it is not an equal playing field to simply compare batteries of the same AH capacity on price alone (as RoughRider is trying to do) without taking into account the quality of materials used in their manufacture.

BMI cells use LiFePO4 powder manufactured under licence in Germany from the patent holder of LiFePO4 battery technology (Phostech).
Common sense would say that any product (not only batteries) will be of better quality and provide superior performance when it comes from the person/company who actually invented it (and therefore has the most experience with it) compared to the same product which has come along some time later as a result of being copied in China (often to less than a satisfactory standard).