A123 20AH cells source?

[frodus]

Hey Phil, lets try to meet this week. The CBA/CC400 is ready for you and has some connectors that clamp down onto the tabs (we tested a Kokam cell).

Got a temp probe too, so we can monitor temperature and see how she goes.

got my number right?

 

[webfootguy]

@$50 a cell that's $0.78125 per wh.... that's around the price point of the 35-40c lipo.....

yep, and could be cheaper than many lipos when you add the freight -and the same wh per kg as lipo too, basicly wins on every count especially by haveing 10-20+ times the life.

ok webfootguy, now we do have an odd situation, my cell (from the same poeple a123rc) was marked up made in korea and yours usa, do they collect seconds from everywhere? allthough yours might not have the self discharge that mine does? can you charge it to 3.6v, hold it there for a good 1/2 hour+ (till the amps drop to .1a) then take it straight off the charger -good cells take overnight to get below 3.5v but mine from a123rc took 10 mins.

a123rc are now not responding to any of my 5 emails about larger numbers..

Ok, so we have 2 data points (you got Korea cells, I got USA cells). Who else is going to take the leap to investigate this?

I bought 2 cells. Both say USA on them and are from the same lot based on serial number. I used my iCharger 3010B to charge one to 3.65 volts until current dropped to C/20. The current drop off was dramatic. It was pulling the full 10 amps very easily on charge during the CC phase. It only took about 2 minutes from the time it hit 3.65v until the current dropped below 1 amp.

After 12 hours, the voltage was at 3.51 volts. I need a better setup (fatter wires) because I measure about 0.2 volt drop at 20 amp discharge between the cell tabs and the number reported by the 3010B. The stock wires are like 14 AWG at best and they connect to fairly wimpy clip leads. Anyone have advise on what I need to make to clamp securely to the tabs with a low R path to 4mm bullets? I'll test the second cell in the same way (12 to 24 hours after charge and measure voltage). So far, both cells look perfect to me. There was no evidence the tabs had been previously clamped or welded to me but it did not come in the "factory" shipping package. The tabs had black protective shrink wrap on them (slip fit) and the cells themselves were wrapped in a thick plastic wrap (much like food wrap but not very sticky and much thicker). If these cells are QA rejects, then I am starting to like QA rejects

Has anyone else gotten their orders? Has anyone else risked the money? I think paypal will protect you if you get something that is not as advertised.

I had no luck getting a123rc to respond to any of my emails as well. I went to the order page and there is a link to ask a question about a particular order. When I used that, they responded back to the page. You have to click the link again to see their response. There is no indication that they have responded, you have to poll for their response. This may be typical for Hong Kong vendors who are not fluent with English.

At this point, it might be cheaper to go with OSN directly as they offered cheaper prices if you purchase in quantity and they did respond when I asked them for a quote. My understanding is that a123rc is just buying cells from OSN anyway. I did not go that route due to no $$ protection. Paypal offered that if I ordered via a123rc because they accepted paypal.

For a large setup, buying a bunch of cells and testing them to get matched cells would be the way to go. We need cell_man to offer this service and weed out IR variances / bad cells for us. That would be worth paying for.

 

[webfootguy]

Hey Phil, lets try to meet this week. The CBA/CC400 is ready for you and has some connectors that clamp down onto the tabs (we tested a Kokam cell).

Got a temp probe too, so we can monitor temperature and see how she goes.

got my number right?

Yup. I'll bring a cell over (and call first).

 

[frodus]

I can't wait to see the results! Would really like to get some of these for my bike, but that might have to wait a year or so. Headway's first, then maybe something like this.

Let me know if you want to meet today or weds. I'm busy tomorrow and Thursday. Just need about 30 minutes to solder new leads on the CBA-II (old ones were shortened and changed to Deans).

 

[webfootguy]

Followup on resting voltage. Charge to 3.65 volts, rest for 24 hours and now voltage is 3.555 volts.

 

[jonescg]

How was the weight? Getting consistent numbers?

 

[webfootguy]

How was the weight? Getting consistent numbers?

Both my cells are exactly 505 grams.

 

[toolman2]

Followup on resting voltage. Charge to 3.65 volts, rest for 24 hours and now voltage is 3.555 volts.

so your made in usa cell has near to 0 self discharge, very good. mine not as good.
the i-charger can do a rough resistance check when any half decent connections (aligators would do) are made to the cell with the ballence wires hoked up as well.
BUT, you will get a higher ir reading if you check it when over 3.38volts, gotta load it just a bit to get the "surface charge" off them.

 

[olaf-lampe]

BUT, you will get a higher ir reading if you check it when over 3.38volts, gotta load it just a bit to get the "surface charge" off them.

With higher you meant better? Otherwise I wouldn't expect a higher IR whit an additional surface charge.
-Olaf

 

[toolman2]

BUT, you will get a higher ir reading if you check it when over 3.38volts, gotta load it just a bit to get the "surface charge" off them.

With higher you meant better? Otherwise I wouldn't expect a higher IR whit an additional surface charge.
-Olaf

higher ir being worse, (the cell dropping more voltage under load) its just that the lifep04 cells all charge to say 3.65volts, but very easily drop to 3.4v from even i tiny load for a sec -so the ir will appear very high above 3.4v, needs to be a bit lower volts to be a fair test.

 

[miro13car]

Again,
only meaninfull Internal Resistance test done with factory IR meter 1000Hz would convince me thay are not QC rejected cells.
A123 own report clearly says they are QC rejects out there.
MC

 

[pgt400]

What about measuring V and I at say .5 and 1 C plus some math to determine battery Z? Accurate / Reliable??

 

[frodus]

thats how I do it, two different constant currents, and measure the volts (and amps) with a meter. Neither are lab calibrated though, but its accurate enough for me. I just don't feel like investing in something I'll use once or twice on these cells. My BMS calculates it as well.

I think for MOST people DeltaV/DeltaA is good enough.

 

[toolman2]

ok, ive just done a pretty decent ir test,
i loaded the cell to 200amps measuring the cell voltage separately with clips to the terminals.
first quick hit to set up test with the cell still cold at 16 deg c gave 1.8m ohms.
then at 20 deg gave 1.5m ohms, the cell was then warming up a bit, to around 26deg and all the wires and connectors were melting.
so i heated the cell to 35 deg c, this was after about 10ah had been taken out of it, and the final test at this point gave 1.0m ohms.
under this kind of use (10c, or flattening it from full within 6 mins) the cell is clearly going to warm to over 35 deg, so 1.0 is the number to go
by for me.

i put it on the i-charger and then got consistant 1m ohms from it also (it gave 2 when the cell was cold)
-i know its not a factory ir meter, but if you look (i think its rc group forums) into it miro13car, youl prolly find it does the 1000hz thing or similar?

so 1m ohm is about what i remember from liveforphysycs massive ir test setup from these cells, to me it means you could pull 330a and lose 10% of your voltage, or people with a mental disorder could prolly go over 400a if the cells are under compression.

so its not a second in the ir department

 

[miro13car]

toolman,
on Falcon EV you can see he clearly uses milvolmeter so no 1000Hz thing?
And HIS cell IR is about 0.6 miliOhm.
From comparing 2 voltages at 2 loads you come up with
1.5 miliOhms at 20C temperature, on Falcon he measure obviously at ambient so around 20Calso so very close to your temp.
So Falcon's cells has three times less internal resistance that your cells.it is clear.
Falcon knows his stuff if he would not offer Agrade cell he would have not charge $80/cell, right?
He charges 80 dollars for a reason.
MC

 

[oatnet]

He charges 80 dollars for a reason.

Andy has never been the cheapest vendor.

I personally would not wait, and then pay 60% more, to have "A" stock. The a123 prisimatics I got from cell_man - which came with tags listing the IR, values from 1.15-1.6 IIRC, have been quite adequate for extreme applications, not to mention supply 16+ AH on a 15AH cell. I can see how the bragging rights would be nice, but do you have a specific design criteria driving you to need super-super-cells, instead of mere super-cells?

 

[miro13car]

oatnet,
agree with you on one side that you need serious non-ebike application like pack for ecar to take full advantage of A grade 20Ah 200A continous cell. But for maximum life you need A grade.
Not exactely 60% more you, exagerate.
Is $50 from A123RC versus $80 from FalconEV.
A123RC price is no more 40 bucks.

MC

 

[frodus]

Lets have a little lesson in math miro.....


Step 1: Cells are ~$50 from A123RC, right?

Step 2: Now multiply that by 1.60 (that means 60% more).

What do you get? $80

What did we learn here today? Well, we learned that 80 is 60% more than 50.

 

[toolman2]

toolman,
on Falcon EV you can see he clearly uses milvolmeter so no 1000Hz thing?
And HIS cell IR is about 0.6 miliOhm.
From comparing 2 voltages at 2 loads you come up with
1.5 miliOhms at 20C temperature, on Falcon he measure obviously at ambient so around 20Calso so very close to your temp.
So Falcon's cells has three times less internal resistance that your cells.it is clear.
Falcon knows his stuff if he would not offer Agrade cell he would have not charge $80/cell, right?
He charges 80 dollars for a reason.
MC

it would be "obvious" in a photo of my a123rc cell test that it was done at ambient temp too, but it wasn't..
three times 0.6 is not 1.5.
this type of test is fraught with variables, all of which can be juggled depending on what your after, one of the differences here is im just some curious dude -and he is selling cells. :wink:
also 0.6m ohm does not agree with any other figures anyone else has for any of these cells.
and yea your right, im sure no one would charge strong money for anything thats not A grade :lol:
by the way is $80 a cell including free postage worldwide?

anyway, it would be good to test one of falcons cells if anyone wants to give it a shot, cos they may well be better.
i actually think my a123rc cell IS a second, not for ir though, -it arrived at 2.38v and thats not cool.

 

[DOMEPEACE]

He charges 80 dollars for a reason.

Andy has never been the cheapest vendor.

I personally would not wait, and then pay 60% more, to have "A" stock. The a123 prisimatics I got from cell_man - which came with tags listing the IR, values from 1.15-1.6 IIRC, have been quite adequate for extreme applications, not to mention supply 16+ AH on a 15AH cell. I can see how the bragging rights would be nice, but do you have a specific design criteria driving you to need super-super-cells, instead of mere super-cells?

i want those bragging rights! EVs arent very popular where i live, and i'd be the only one in 500 miles with legit A-grade prismatics! its not their power density that has me sold, its the lifespan. im some future EV designs I have, if i were to go with anything but 7000+CL id be changing my pack more than i would oil!(that is if it had oil, or petrolium for that matter!) i poured over the last few pages, and i couldnt find who you were quoting, dosent matter tho, I just want to know where i can get legit prismatics or the ANR026650M1 @ no less than A grade, bulk discount would be nice. $30 extra isnt nothing for QC pass uber batteries! ill just skip buying some chronic lol. oh and if i can specific order the US made cells, thats even better! love and props to whoever is willing to provide this info!

 

[miro13car]

well,
Falcon EV picture clearly shows cell on the top of HP miliohmmeter, range is 1miliohm. These are facts. What is temperature in room? There is no reason to believe there is 30degC in the room. for sure.
Should we say A123RC cell has double IR comparing with Falcon?
if 0.6 miliohm agree or not agree with any known.....meter clearly shows it in picture unless you think it is deliberate set up/tempering with miliohmmter?
Yes , I never heard of LiFePo cell shipped at below 3V terminal voltage.

 

[webfootguy]

Here is an update. I went over to Travis's place (frodus) tonight with the two 20Ah cells. The setup for the cells that I had charged both of them to 3.65 volts at a C/2 rate (10A) until current dropped below 1A. The the cells then rested for 48 hours prior to the test. The first cell we setup to discharge at 100A (5C) rate until it reached 2.0 volts for a full discharge that should last around 12 minutes (5C is 60/5 = 12).

The test setup is a CBA II/III with a CC400 current amplifier. The discharge at 100 A was done by setting the CC400 to 90 A and the CBA to 10 A so 100 A combined. The 100 A current was confirmed with a clamp on DC amp meter (it measured about 104 A which is within instrument error margins).

NOTE: on the graph, the x axis reads 10x too small. This is because of the CC400 amplifier. The CBA only sees 10% of the current. To get the actual values for the x axis, multiply the values by 10.

The cell started out at 3.569 volts and ended up at 2.000 after 707 seconds (11min 47sec) at a constant discharge rate of 100 A. The temperature went from 72.8F to 103.4F during the same time. The capacity measured at 19.592 Ah (note the factor of 10 from the graph).

The second cell we discharged at 50A for about 30 seconds and noted the current and voltage, then dropped the current to 20A and waited about 30 seconds and measured the current and voltage again. The idea was to get a handle on the internal cell resistance using the delta V over delta I method (rather than the 1000 Hz AC method which requires a special test instrument to do properly).

Now the data. The voltage and current were measured at about the 30 and 60 second point at 3.214 volts/51.2 Amps and 3.274 volts/20.2 amps respectively. One can calculate DC resistance using deltaV/DeltaI which in this case is (3.274-3.214) / (51.2-20.2) = 0.060 / 31 which is 1.9 mOhm if I did the math right. Remember this is not the same number you get with the AC test but is good enough for our purposes and seems to fall in the ball park of what others have seen with the AC test. I have yet to find an actual datasheet for these cells.

Now some pictures:

 

[webfootguy]

toolman,
on Falcon EV you can see he clearly uses milvolmeter so no 1000Hz thing?
And HIS cell IR is about 0.6 miliOhm.
From comparing 2 voltages at 2 loads you come up with
1.5 miliOhms at 20C temperature, on Falcon he measure obviously at ambient so around 20Calso so very close to your temp.
So Falcon's cells has three times less internal resistance that your cells.it is clear.
Falcon knows his stuff if he would not offer Agrade cell he would have not charge $80/cell, right?
He charges 80 dollars for a reason.
MC

In the picture at this link:

http://www.falconev.com/a123_20Ah_ohms.jpg

he is clearly using an HP 4328A. The specification for this instrument at this link:

http://www.ntecusa.com/docs/HP_4328A_info.pdf

shows it uses the AC method at 1000 Hz plus or minus 100 Hz.

Other can confirm this but using the AC and DC method produces different numbers for the internal resistance of a cell.
I can not speak as to Falcon EV's reputation or why he charges $80 as I have never bought anything from that site.
I have been around the block long enough to know that a higher price does not always mean a higher quality product.

Given the recent event of cells showing up in Hong Kong / China, he may be getting them from the same source as we are.
If he takes credit cards or paypal you have lots of protection against fraud. This is not true of many Far East vendors.
Even more, if he offers any kind of warranty, and especially if he will match the internal resistance of the cells than I would think that may be worth the $80 / cell vs $50 per cell. On the other hand, you could just buy extra cells and do the matching yourself and keep the extra cells as spares or for other projects that did not need a close match.

 

[AndyH]

toolman,
on Falcon EV you can see he clearly uses milvolmeter so no 1000Hz thing?
And HIS cell IR is about 0.6 miliOhm.
From comparing 2 voltages at 2 loads you come up with
1.5 miliOhms at 20C temperature, on Falcon he measure obviously at ambient so around 20Calso so very close to your temp.
So Falcon's cells has three times less internal resistance that your cells.it is clear.
Falcon knows his stuff if he would not offer Agrade cell he would have not charge $80/cell, right?
He charges 80 dollars for a reason.
MC

In the picture at this link:

http://www.falconev.com/a123_20Ah_ohms.jpg

he is clearly using an HP 4328A. The specification for this instrument at this link:

http://www.ntecusa.com/docs/HP_4328A_info.pdf

shows it uses the AC method at 1000 Hz plus or minus 100 Hz.

Other can confirm this but using the AC and DC method produces different numbers for the internal resistance of a cell.
I can not speak as to Falcon EV's reputation or why he charges $80 as I have never bought anything from that site.
I have been around the block long enough to know that a higher price does not always mean a higher quality product.

Given the recent event of cells showing up in Hong Kong / China, he may be getting them from the same source as we are.
If he takes credit cards or paypal you have lots of protection against fraud. This is not true of many Far East vendors.
Even more, if he offers any kind of warranty, and especially if he will match the internal resistance of the cells than I would think that may be worth the $80 / cell vs $50 per cell. On the other hand, you could just buy extra cells and do the matching yourself and keep the extra cells as spares or for other projects that did not need a close match.

Strictly from the picture, we really don't know much. AC Ri will be different than DC Ri. We don't know for sure that the milliohmmeter is in calibration, and we don't really know if the meter display is from the cell in the picture - I have different sets of leads for my 4328 as well. Ri will change with state of charge and ambient temperature.

It's a nice picture but it really doesn't give us any reliable information.

 

[frodus]

Another thing I was thinking about.

If someone doesn't let the battery voltage "settle" at each current level, they can easily skew the results. We let the voltages settle for 30s or so, until they kept more constant. If people aren't doing this when they measure, they can skew the results either direction, so be careful.

I think the best is to use a meter designed for this, and I might actually buy one for this purpose, so I can match cells.