Battery Space 20AH Prismatic LiFePO4 tests

Finally posting the results of my tests on two of the Battery Space hardcase 20AH cells:
http://www.batteryspace.com/lifepo4-prismatic-module-3-2v-20-ah-10c-rate-64-wh-6-0---un38-3-passed-dgr.aspx

BSP1 Arrived with 3.316V on it, BSP2 arrived with 3.314V.

The following graphs are after about 10-12 cycles each of the batteries previously, but due to various problems with my CBAII, and a few errors of my own, I figured it would be a waste of time to post them. I finally worked out some of the problems with CBA, with heavier wires, better connectors, Arctic Silvering the FET to the heatsink, and more cooling.... So I think the following tests are pretty accurate. even overcoming the Internal resistance of the CBAII ... With these batteries, every time I checked voltage with a good meter, it was within .02V of what the CBA indicated

I accidentally over-discharged both batteries down to 2.0V a couple of times each, and it doesn't seem to have hurt them in the short term. Mostly once I got the CBA running right, I tried to keep the cutoff voltage at 2.5V.

BSP1 @
5A = 20.939AH
10A = 20.659AH
15A = 19.343AH*
20A = 20.369AH
25A = 19.530AH

* I think what caused this anomaly, the less than 20AH results is that for that test, I had charged it at 20A on my Turnigy 400watt charger, terminating at 3.60V, while all the other tests were either charged or at least topped off with a 'Voltphreaks' 2A single cell charger which charges more gently and terminates at ~3.8V. The same thing goes for the 20AH test on BSP2 below.

BSP2 @
5A = 21.379AH
10A = 21.073AH
15A = 20.860AH
20A = 19.957AH*
25A = 20.731AH

I didn't try going higher than 25Amps... although with the modifications to the CBA I might be able to do a test at 30A if anyone really wants me to.

Here are graphs of the tests:

.


Here are links to the actual CBAII test files if anyone is interested in looking closer:
(Right click, Save As, )
http://www.az123.com/E-S/BSP1.bt2
http://www.az123.com/E-S/BSP2.bt2

You don't need a CBA to read and manipulate the test charts... Just download the software from here:
http://www.westmountainradio.com/kb_view_topic.php?id=OT39.

I calculated the Internal resistance of the batteries several times, and never got consistent results... They ranged from .0028 to .0086 ohms, depending the amperages I was comparing. Maybe it was my calculations, rather than the measurements...
With the CBA software and the chart files you can put your mouse at any points on any curve, and get the voltages at that point in time, and calculate them yourself, if you wish...
Or wait a few days... I have a Turnigy meter coming that measures IR, and hopefully it will be more accurate than my calculations.

GENERALLY, it seems as if still after around 15 full charge and discharge cycles that the AH output is growing a bit, all other factors remaining the same.

Here are a couple of photos so you can get a visual idea of the BSP 20AHs, compared to old pre-terminal Headway 38120LiFe and 18650LiIon



Questions?
Opinions?

Do these seem like decent batteries, considering the $35.00 ea. price point?

Thanks for posting these results..

A 48v 20Ah pack of these would seem that they would work well in my scooter…

IAN.

imorton said:

Thanks for posting these results..

Click to expand...

You are welcome, Ian.

Interesting. 70 page views, and yet only your comment (and now this one)
I was kind of hoping for some substantive feedback by now.

I would've done a more extensive post, but I was on my iPhone

So from what I can see, they can be an excellent replacement for the headway cells.

Personally, I would love to grab 16 of the cells and put them together with a BMS and see how they work on a 48v scooter.

Heh, I'd be lucky to type that much with an I-phone. I hate those little touch screens.

I plan on ordering 25 -30 or so of them for a 72V pack+ spares, if some of the experts chime in and say they look decent, performance-wise / cost.

(And maybe buy a couple of the 40 AH ones to test, although I'd assume they were the same, except size and capacity. The 40A ones might work in the Suzuki motorcycle chassis I have. )

I'd go with the a123 prismatic cells for performance, but I'll be beating the heck out of them, so I like the idea of a hard case, and swapping back and forth, reconfiguring packs, so I like the screw terminals.

Discharge curve tells it all.
Obviously you want it to be flat, flat. You want this "knee" drop at the end of discharge as sharp as possible.
At 20A discharge curve of A123 20Ah pouch tested on the same CBAII is much more flat, voltage sag is much smaller.
I ride on A123 12S , QC rejects from Victpower, I am sure those $70/pouch no-rejects from A123 are even better.
Hard to beat A123 pouches.

miro13car said:

Discharge curve tells it all.
Obviously you want it to be flat, flat. You want this "knee" drop at the end of discharge as sharp as possible.
At 20A discharge curve of A123 20Ah pouch tested on the same CBAII is much more flat, voltage sag is much smaller.
I ride on A123 12S , QC rejects from Victpower, I am sure those $70/pouch no-rejects from A123 are even better.
Hard to beat A123 pouches.

Click to expand...

How much would a 48v 20Ah battery of these A123 pouches cost?

The GBS 20Ah cells x16 works out to @ $500

What does one weigh? I'm curious what a 16 s pack of them would be. A ping pack the same capacity would be about 16 pounds including some kind of protective cover.

Certainly the hard shell is worth something, as well as the ability to swap out a weak cell. But it's not super affordable, and certainly won't meet limn standards for size and weight.

Still, if size and weight matters little a reasonable choice for 20 amps controller bikes and scoots.

I think the price is fairly reasonable when you compare the price to a 48v 20Ah Ping pack. (@ $834 usd with shipping)

No disrespect to Ping.

And it's a USA supplier

These seem more appropriate for a larger ev application like a golf cart or lawn mower. The size and weight alone are deal killers for my ebike applications. :?

wineboyrider said:

These seem more appropriate for a larger ev application like a golf cart or lawn mower. The size and weight alone are deal killers for my ebike applications. :?

Click to expand...

I agree about them being more suitable for larger applications however, they are ideal for scooters, trikes, and bikes that are able to handle a larger battery pack.

My avatar "scooter", comes with 48v 12Ah lead, so these are identical replacements

Maybe Santa Claus will bring me a pack of four for Christmas ...

Nice cell mini-review, don't see many posts about LiFePO4 stuff these days, wonder why that is?
Just reading this thread makes me feel like building a LiFePO4 pack.

dogman said:

What does one weigh? I'm curious what a 16 s pack of them would be.

Click to expand...

The site says 750 Grams / 1.65 lbs each although my calculations say that is closer to 1.5 lbs each if the 750 gram figure is accurate. I don't have a scale around here that will weigh things in that neighborhood with precision... but the old 'grocery scale' I have does say it is right at 1.5 pounds... Two of the old non-screw terminal Headway 38120s come in at about 1.5 ounces lighter.

Hmm, I do have a precision scale that will weigh a single Headway... 10.95 Oz
X2 = 1.39 lbs +1.5 oz... Yeah, the Battery Space Prismatic is real darn close to 1.5 lbs

So your 16 S pack would come in at just a tad over 24 pounds.

RLT said:

dogman said:

What does one weigh? I'm curious what a 16 s pack of them would be.

Click to expand...

The site says 750 Grams / 1.65 lbs each although my calculations say that is closer to 1.5 lbs each if the 750 gram figure is accurate. I don't have a scale around here that will weigh things in that neighborhood with precision... but the old 'grocery scale' I have does say it is right at 1.5 pounds... Two of the old non-screw terminal Headway 38120s come in at about 1.5 ounces lighter.

Hmm, I do have a precision scale that will weigh a single Headway... 10.95 Oz
X2 = 1.39 lbs +1.5 oz... Yeah, the Battery Space Prismatic is real darn close to 1.5 lbs

So your 16 S pack would come in at just a tad over 24 pounds.

Click to expand...

I had a look, and a Ping pack (48v 20Ah) weighs @ 22lbs, so the GBS cells/pack would only be slightly heavier.

I like the idea of local (USA) supply for the GBS cells, and the repairability as it would make the difference between a broken down pack, and one that is easily repaired.

Now if readers here would like to donate $500 towards me doing some real life testing, I would consider sacrificing my "scooter" for science… Lolll

IAN.

Well, I bit the bullet and ordered another 28 of these. Was kind of hoping for more input by the experts, but I guess someone has to actually put them to work to know for sure.

I'll probably build a 23S 1P for a nominal 72V 20 AH pack, and use the rest for spares .. (Or maybe go with 24S.. I guess my Kelly controller will handle the extra few volts as long as I let them sit for a while before I turn on the key).... and distribute them fore, midships and aft on my strong old mountain bike frame that used to be my 2WD, but now will run a 5305 in back..

Or I could go with 15S 2P to replace or supplement the 48v/40AH SLAs in my chinese trike-truck, if it proves to be to heavy and bulky for the bike.

From their posted datasheet: http://www.batteryspace.com/prod-specs/6332.pdf

The 2000+ cycles stated is beyond industry expectations. Only A123 can make that claim and on the basis of nanophosphate R&D and patent. That would be the test I'd like to see, hooking it up to a respectable charge, discharge test machine to have vendor independent verification of the data. If they prove out, yea, since they are half the cost of the A123 AMP20 prismatic cells.

I'll be happy with 1000 cycles even 600 cycles would not make me go insane with anger. I never even noticed the 2000 times claim.
My only real interest in these was the basic capacity, easy modularity and the physical durability. I'm not a big fan of the 'pop-tart' packages, (as Marty called them on another thread).

Your purchase & investment in the make has to make sense to you. But if cycles were in the order of 1000, I'd opt for the A123 cells. The cost savings are lost without those cycles.

arkmundi said:

Your purchase & investment in the make has to make sense to you. But if cycles were in the order of 1000, I'd opt for the A123 cells. The cost savings are lost without those cycles.

Click to expand...

Isn't the industry standard for discharge testing something like less than .2C (like 4 amps in this case). So couldn't they in theory with a 4 amp load be able to provide 2000 cycles going to 80% DOD…?

imorton said:

Isn't the industry standard for discharge testing something like less than .2C (like 4 amps in this case). So couldn't they in theory with a 4 amp load be able to provide 2000 cycles going to 80% DOD…?

Click to expand...

You have to take the word of the manufacturer in their datasheet and they are saying 2000+ cycles. But there are independent testers like http://www.jbicorp.com/. A high volume order would almost certainly carry with it a requirement for such testing on a periodic basis, to insure there is not slippage during manufacture. We don't know anything about this battery. Seems Battery Space has contracted for production from a Chinese source and put their label on it. Hence, some initial scepticism is warranted.

As far as what an independent test might do, its up to the buyer, for their application and expected rigor:

jbicorp.com said:

Life Cycle Testing

JBI can perform Life Cycle Testing using constant current, constant power, constant resistance, and/or constant voltage. JBI has over 300 programmable charge / discharge testing circuits.

In the 0-18 volt range, Life Cycle Testing up to 300 amperes charge and discharge can be performed on calibrated programmable circuits at temperatures from -40°C to 80°C.

Test Temperatures from -40°C to 25°C are obtained using cold boxes with calibrated controls.

Test Temperatures from 25°C to 80°C are obtained using waterbaths, ovens, or environmental chambers with calibrated controls.

Life Cycle Testing with a range of 0-75 volts and 0-500 amperes is available. A range of 0-2000 amperes, 0-18 volts discharge, with 0-100 amperes, 0-18 volt charge is also available.

Click to expand...

arkmundi said:

imorton said:

Isn't the industry standard for discharge testing something like less than .2C (like 4 amps in this case). So couldn't they in theory with a 4 amp load be able to provide 2000 cycles going to 80% DOD…?

Click to expand...

You have to take the word of the manufacturer in their datasheet and they are saying 2000+ cycles. But there are independent testers like http://www.jbicorp.com/. A high volume order would almost certainly carry with it a requirement for such testing on a periodic basis, to insure there is not slippage during manufacture. We don't know anything about this battery. Seems Battery Space has contracted for production from a Chinese source and put their label on it. Hence, some initial scepticism is warranted.

As far as what an independent test might do, its up to the buyer, for their application and expected rigor:

jbicorp.com said:

Life Cycle Testing

JBI can perform Life Cycle Testing using constant current, constant power, constant resistance, and/or constant voltage. JBI has over 300 programmable charge / discharge testing circuits.

In the 0-18 volt range, Life Cycle Testing up to 300 amperes charge and discharge can be performed on calibrated programmable circuits at temperatures from -40°C to 80°C.

Test Temperatures from -40°C to 25°C are obtained using cold boxes with calibrated controls.

Test Temperatures from 25°C to 80°C are obtained using waterbaths, ovens, or environmental chambers with calibrated controls.

Life Cycle Testing with a range of 0-75 volts and 0-500 amperes is available. A range of 0-2000 amperes, 0-18 volts discharge, with 0-100 amperes, 0-18 volt charge is also available.

Click to expand...

Click to expand...

Sounds pretty severe and serious testing going on there. Like the OP mentioned, if they last 1000 cycles at 80% DOD, that would make them ideal for vehicles that have the space and weight carrying ability.

Personally for me, I would KISS Santa Claus if he brought me 16 of the 20Ah cells…. Well, maybe I wouldn't kiss him, but I would let my woman sit on his knee… Lolll

I bet JBI would charge more for testing a couple of the batteries than the cost of the 30 batteries did. Even if it only cost half as much, it would be pretty stupid of me to spend that kind of money for testing, and have nothing to show for it. If I was going to buy a thousand of them, it would be different, and I'd do, or pay for a more rigorous testing.

Now, the CBA tests that I did are very limited, I'll admit, and I have been too lazy to set up a brute force tester and run it through manually 2000 times, or to spend the couple to several thousand of dollars of equipment to do it automatically. But what testing I did do convinced me that it was worth the risk to buy enough to do a 'real world' test. I never expected these cells to be the performance equivalent of A123 cells, nor would most reasonable people expect them to be, given the price point.

After the experience with the early Headways, I learned to not take the manufacturer's or reseller's words for anything. That's why I ordered two to test first. And within the limitations of my testing abilities, the batteries preform, at least in the short term, as advertised, and within the acceptable limits of what I intend to ask of them. If they start dying off after a hundred cycles, then you can say "Well, you were stupid to have purchased them", and I'll agree.

I would have purchased A123s if their physical dimensions and structural durability were closer to what I wanted. I'm actually still tempted to buy a couple of boxes of them to have on hand should I decide on a different configuration and to go through the commitment to build a durable shell for them, and hardwire them into a specific configuration. I learned long ago to not 'nickle and dime' when it comes to batteries. And I'm actually TRYING to spend a lot of cash, while the US$ is still worth the 7 cents worth of goods that it bought in my youth. So it isn't a matter of money that I chose to try a batch of these cells.

All I ever hoped for on this thread was that the experts judge them on the basis of what my CBA tests found, not what the seller claims for specifications, or 'Oh, they aren't A123's, so therefore are no good'.

RLT said:

I bet JBI would charge more for testing a couple of the batteries than the cost of the 30 batteries did. Even if it only cost half as much, it would be pretty stupid of me to spend that kind of money for testing,...

Click to expand...

Yes, yes, of course. I was making no such suggestion. Merely answering a question. I'm saying with A123, you know exactly what you're buying, because those independent tests have been done. There's no misinformation, no false claims. With these Battery Space cells, we don't know shit about them except certain claims. You can trust it or not at this time. And if you do, let us know your experience, good or bad, as you've become "our" independent tester. Good luck with that. Just don't be disappointed if you get far less than the 2000+ cycles claimed.

Testing continues with the new 28 of these cells. I'm about 1/4 through Mostly they are close to the original 2, although a couple of them so far are approaching 22AH of capacity at 10A, and somewhat over 21AH at 20A. (1C) IRs so far are between 3.5 and 5.1 mohm.

This afternoon, I tried to see how many I could fit into a Toppeak MTX bag (the 480c.i) one Very conveniently, 12 of them fit in quite nicely, with a bit of room left over for switches fuses etc.:





The bag WILL hold 15:



although you can't quite completely close the zipper:


Not too bad. The original plan for this bag (years ago was to hold 24S Headway 10AH 38120s... which was a tight fit too. So, it about holds the same wattage capacity of the prismatics, if only half the voltage of the headways.
I may try a front and rear battery pack again on my old Mongoose, newly reconfigured to rear wheel drive.



from its old 2WD build
Will still be lighter up front than it was when I had a motor and 44.4V 15AH of LiIons on it

So it looks like the GBS cells are a solid and worthy of being used in applications where size and weight are not critical.

I would love to grab 16 of these and do a 48v 20ah pack. They would easily fit in my scooter and provide easily a 40 km range. Repairability is the best thing that I like.

If I was to order (4) of the 12v 20ah pack, I wonder if/or I should order some single cells as spare parts. Or maybe I should order a complete 12v20ah pack as a spare (drop-in) replacement should some cells go bad…?

IAN.