arkmundi
10 MW
Yea, sorry to have hijacked ... taking that to on the road solar chargingRLT said:
Battery Space 20AH Prismatic LiFePO4 tests
- Thread starter RLT
- Start date
RLT
10 kW
Ok, I've finished the basic 'desktop' testing of the followup batch of 28 of these cells, and results were... Umm, variable. I guess as you would expect for low to medium-end Chinese manufacture.
On the bright side, there wasn't as much variation between cells as there was among the early Headway 38120 cells.
I tested this batch on the CBA only at 10A and then at 20A, with just a couple of insignificant exceptions. Most slightly exceeded their ratings, so that is good. I still haven't tested any at more than 25 Amps, (other then the one that I accidentally shorted out with a wrench that got a good chunk zapped out of it... and I didn't have a meter attached so I have no idea if they will really pull the 3C continuous and the 10C momentary current that they are rated )
I had two batteries that pulled over 21AH @10A and about 20.75AH at 20A
And unfortunately I had four others that didn't make it all the way to 20AH at 10 A (19.695. 19.918, 19.915, 19.662 AH respectively), and a couple more that were a bit under 20AH at 20A..
Most of them were in the neighborhood of 20.3 -20.5AH at 10A and 20.2- 20.4 @ 20A
Internal Resistance definitely fell short of Battery Space's claimed ~2.5Mohm, (Which I thought from the very begining was highly optimistic) but mostly wasn't bad. My lowest one was 4.3Mohm and the highest 13.0Mohm. I haven't calculated it, but I'd guess both the average and median IR was about 6.0 to 6.2 Mohm.
Even the higher IR batteries only got a few degrees above ambient temperature after an hour at 20amps, so at normal e-bike applications, you shouldn't have to worry about that.
If anyone has any specific questions, or wants to see more detailed results, let me know.
I'll update with any further stuff I find. Got a Cromotor coming in in a couple of days, so their first "Real World" test may be on that, with a 200A Kelly controller, or maybe I'll try them out on my current platform, a 5305 and Lyen 18 FET controller.
Either one should give them a good workout
On the bright side, there wasn't as much variation between cells as there was among the early Headway 38120 cells.
I tested this batch on the CBA only at 10A and then at 20A, with just a couple of insignificant exceptions. Most slightly exceeded their ratings, so that is good. I still haven't tested any at more than 25 Amps, (other then the one that I accidentally shorted out with a wrench that got a good chunk zapped out of it... and I didn't have a meter attached so I have no idea if they will really pull the 3C continuous and the 10C momentary current that they are rated )
I had two batteries that pulled over 21AH @10A and about 20.75AH at 20A
And unfortunately I had four others that didn't make it all the way to 20AH at 10 A (19.695. 19.918, 19.915, 19.662 AH respectively), and a couple more that were a bit under 20AH at 20A..
Most of them were in the neighborhood of 20.3 -20.5AH at 10A and 20.2- 20.4 @ 20A
Internal Resistance definitely fell short of Battery Space's claimed ~2.5Mohm, (Which I thought from the very begining was highly optimistic) but mostly wasn't bad. My lowest one was 4.3Mohm and the highest 13.0Mohm. I haven't calculated it, but I'd guess both the average and median IR was about 6.0 to 6.2 Mohm.
Even the higher IR batteries only got a few degrees above ambient temperature after an hour at 20amps, so at normal e-bike applications, you shouldn't have to worry about that.
If anyone has any specific questions, or wants to see more detailed results, let me know.
I'll update with any further stuff I find. Got a Cromotor coming in in a couple of days, so their first "Real World" test may be on that, with a 200A Kelly controller, or maybe I'll try them out on my current platform, a 5305 and Lyen 18 FET controller.
Either one should give them a good workout
imorton
1 kW
Hi, I want to really thank you for taking the time to do all this testing. 
Now for a non-techie guy like me, it looks like he's batteries are relatively good and meet their specs.
The only question I have, will they meet the 1000 cycle spec, that's the most important one to me. In other words if I buy them Will I be able to discharge them at 15 to 20 amps, and have them last 3-4 years.
Thanks again for taking the time to test him and sharing it with us
IAN
Now for a non-techie guy like me, it looks like he's batteries are relatively good and meet their specs.
The only question I have, will they meet the 1000 cycle spec, that's the most important one to me. In other words if I buy them Will I be able to discharge them at 15 to 20 amps, and have them last 3-4 years.
Thanks again for taking the time to test him and sharing it with us
IAN
RLT
10 kW
imorton said:
Heh, I'll let you know in about 5 years.... or 3 years if gasoline goes up to ten bucks a gallon.
imorton
1 kW
Because from what I can tell, they meet their specs at +/- 3%.
Now if they only achieve 250 charge cycles instead of the estimated 1000 charge cycles, that would make them really poor.
IAN..
PS: unfortunately there is no way to test this…
Now if they only achieve 250 charge cycles instead of the estimated 1000 charge cycles, that would make them really poor.
IAN..
PS: unfortunately there is no way to test this…
RLT
10 kW
Well, there are ways to test it, but too time consuming and/or expensive equipment for me to do it.
imorton
1 kW
Because from what I see, making a 48v 20ah pack of these cells, we are looking at @ $725
$500 for the cells
$50 BMS
$50 wiring, plugs etc…
$50 charger
----------
@ $650
tax & shipping
--------
@ $725
Not a cheap battery, so I would want the usage/calendar life to at least attain 750-1000 charge cycles…
$500 for the cells
$50 BMS
$50 wiring, plugs etc…
$50 charger
----------
@ $650
tax & shipping
--------
@ $725
Not a cheap battery, so I would want the usage/calendar life to at least attain 750-1000 charge cycles…
dmwahl
10 kW
They appear to be composed of these cells, which claim 2000 cycles to 80% DOD, 3000 to 70% DOD. I wonder if they only claim 1000 cycles for this pack since there's no BMS and the cells are therefore more likely to see over(dis)charge. Seems like a good option though for building a pack, $30 per cell this way vs $35 individually and you get some connection straps.
I have a 100Ah 8S pack using GBS cells and they include heavy aluminum plates banded around 4S blocks, so while it may not be necessary on these due to the smaller size, it's probably still a good idea to have some compression and certainly won't hurt. The lack of compression could also be part of the reason for the lower cycle count.
I have a 100Ah 8S pack using GBS cells and they include heavy aluminum plates banded around 4S blocks, so while it may not be necessary on these due to the smaller size, it's probably still a good idea to have some compression and certainly won't hurt. The lack of compression could also be part of the reason for the lower cycle count.
dnmun
1 PW
compression has been shown to be essential to cycle life since the development of the gel VRLA batteries back in the late 1930s when SLA batteries were invented.
dmwahl
10 kW
Do you know how to quantify how much compression is required or what an acceptable level of swelling is? The plastic cell casing will provide some compression, and the smaller the cell the more it will provide assuming the 20Ah has the same thickness walls as the larger cells. It's difficult to apply a general principle like "use compression" without some kind guideline about how much compression is actually required. If I tell someone to "water your garden" but don't take into account what is planted, soil type, weather, etc, then it's largely a useless recommendation.
dumbass
100 kW
I have been using these cells for about 5 years now. Actually, my old (5 years) cells are the yellow ThunderSky cells. Then I think they changed their name to GBS. I have always bought the 4 packs and built them into 24v packs. Then coupled 2 together for my 48v packs. But I always removed them from the plastic case they come in. And I can't really say that I have them under any intended compression. And I have no bulging at all. I don't run them through a lot of cycles so I can't speak of their life span. I do not use a BMS but use a Cellog8 with an alarm for both charging and discharging. I currently use individual chargers for charging. In my untechnical opinion they are difficult to keep balanced. I am so frustrated with the balance problem I am going to try parallel charging only. Then convert back to serial connect for use. Here's my thread on the subject;
http://endless-sphere.com/forums/viewtopic.php?f=3&t=68602
I currently have 36 cells. 16 old (ThunderSky - 5 years of use) and 20 cells (GBS) I will be putting in service in a few weeks. These "new" cells are actually 2 years old but I never used them. But I sure will now!
I used to buy from here;
http://ev-power.com.au/webstore/index.php/12v-lifepo4-batteries/gbs-12v-lifepo4-batteries-1/12v-20ah-lifepo4-battery.html
http://endless-sphere.com/forums/viewtopic.php?f=3&t=68602
I currently have 36 cells. 16 old (ThunderSky - 5 years of use) and 20 cells (GBS) I will be putting in service in a few weeks. These "new" cells are actually 2 years old but I never used them. But I sure will now!
I used to buy from here;
http://ev-power.com.au/webstore/index.php/12v-lifepo4-batteries/gbs-12v-lifepo4-batteries-1/12v-20ah-lifepo4-battery.html
dmwahl
10 kW
I found a paper related to cell compression and cycle life, http://www.princeton.edu/~spikelab/papers/103.pdf
It doesn't give specific recommendations, but seems pretty clear that some compression is beneficial to cycle life and too much is worse than none at all.
It doesn't give specific recommendations, but seems pretty clear that some compression is beneficial to cycle life and too much is worse than none at all.
dnmun
1 PW
there was a mention of it in the A123 pdf. there was a graph showing an optimum pressure which i do not recall.
for the SLA they discovered that when the electrodes were stacked horizontally like pancakes there was an improvement in cycle life.
i now mount all my SLA packs with the plates horizontal. i think it will more than double cycle life by reducing the damage to the gel/electrode interface by reducing the movement of the interface with the gas formation during charging.
i think the same principal is applicable for the lithium cycle where there can be gas formation at the interface.
for the SLA they discovered that when the electrodes were stacked horizontally like pancakes there was an improvement in cycle life.
i now mount all my SLA packs with the plates horizontal. i think it will more than double cycle life by reducing the damage to the gel/electrode interface by reducing the movement of the interface with the gas formation during charging.
i think the same principal is applicable for the lithium cycle where there can be gas formation at the interface.
RLT
10 kW
Just an update... I finally got around to using some of these batteries... in a 16S configuration for a 3 wheel electric 'pickup' cart.
For about five charge cycles they worked fine: An occasional burst of 45 A for maybe 20 seconds at a time, mostly 22-25 A for a minute at a time. and then 15-18 A the rest of the time; and not used for more than five minutes at a time. Alternated bulk charging with single cell charging to keep them balanced, and never discharged below 90%... I only use it to haul tools and materials around my place.
Anyway, I don't know what happened, but over the last two days somehow the cells drained from 3.25 V per cell to around 0.35V per cell, None of the external wiring shows any evidence of a short The circuit breakers didn't trip.... I haven't opened the controller to see if something happened inside there. No odor of 'magic smoke' at all. on the outside
11 batteries swelled badly (but haven't split the plastic ), two swelled a little bit, and three seemed to not have swelled. The pack was well compressed, sides and bottom, wrapped in fiberglass tape as tight as I could get it, first in In 4 cell packs, then the whole 2 X 8 cell pack tightly wrapped again, Then in a tight wooden box built around it, which in turn fit into tightly into a steel frame of angle Iron and square tubing. I had a heck of a time getting the battery pack out of the metal frame. Had to disassemble the wood case a piece at a time and pry each piece out individually..
The maximum ambient temperature has been about 92F and the thing sits in the shade mostly.
For about five charge cycles they worked fine: An occasional burst of 45 A for maybe 20 seconds at a time, mostly 22-25 A for a minute at a time. and then 15-18 A the rest of the time; and not used for more than five minutes at a time. Alternated bulk charging with single cell charging to keep them balanced, and never discharged below 90%... I only use it to haul tools and materials around my place.
Anyway, I don't know what happened, but over the last two days somehow the cells drained from 3.25 V per cell to around 0.35V per cell, None of the external wiring shows any evidence of a short The circuit breakers didn't trip.... I haven't opened the controller to see if something happened inside there. No odor of 'magic smoke' at all. on the outside
11 batteries swelled badly (but haven't split the plastic ), two swelled a little bit, and three seemed to not have swelled. The pack was well compressed, sides and bottom, wrapped in fiberglass tape as tight as I could get it, first in In 4 cell packs, then the whole 2 X 8 cell pack tightly wrapped again, Then in a tight wooden box built around it, which in turn fit into tightly into a steel frame of angle Iron and square tubing. I had a heck of a time getting the battery pack out of the metal frame. Had to disassemble the wood case a piece at a time and pry each piece out individually..
The maximum ambient temperature has been about 92F and the thing sits in the shade mostly.
RLT
10 kW
And another update:
After extensive tests of the 16 S pack that had been accidentally discharged down to around .35 V per cell (Turned out my controller had about a cup of water in it from a long, nearly horizontal rainstorm... water must have run down the wires and seeped through the silicone seal..., and I had left the master switch on, so there was a continuous three to seven watt draw for a few days after the storm, before I went to use the vehicle again.)
Anyway , all but three recovered enough to be usable. ranging from about 17.3 to 19.1 AH on the CBA, tested several times each from 25A to 10A. The three rejected ones were averaging 14. 16 & 16.5AH The swelling on most of the cases went down a bit with cycling, which was kind of a pleasant surprise.
I had another 14 of them sitting on a shelf (fully charged, DOH!!!) for the last few years, never used, just the original CBA tests... I ran those through a similar CBA test program, and they were mostly equivalent in range (19.5AH to 17.5AH) to the ones that had been so abused.
Then I ordered five new ones from BatterySpace.. and tested them a few times on the CBA at 25 to 10A, They have made some significant improvements since I bought the original 30 batteries. The WORST of the new ones were about 1 AH more then the best of the old batch, and the best of them was about 3.5 AH better than the worst of the old ones on the original CBA tests., ranging from 21.118AH to 22.285AH @ 20 A draw.
Physically the cases have changed from the old ones. The new ones are about 1/8 inch taller than the old ones, and the plastic ridges are differently arranged. And instead of a screwed in vent caps like in the old ones, the new ones have a glued in cap.
I can supply the CBA test files if anyone is interested enough to want to look at them.
After extensive tests of the 16 S pack that had been accidentally discharged down to around .35 V per cell (Turned out my controller had about a cup of water in it from a long, nearly horizontal rainstorm... water must have run down the wires and seeped through the silicone seal..., and I had left the master switch on, so there was a continuous three to seven watt draw for a few days after the storm, before I went to use the vehicle again.)
Anyway , all but three recovered enough to be usable. ranging from about 17.3 to 19.1 AH on the CBA, tested several times each from 25A to 10A. The three rejected ones were averaging 14. 16 & 16.5AH The swelling on most of the cases went down a bit with cycling, which was kind of a pleasant surprise.
I had another 14 of them sitting on a shelf (fully charged, DOH!!!) for the last few years, never used, just the original CBA tests... I ran those through a similar CBA test program, and they were mostly equivalent in range (19.5AH to 17.5AH) to the ones that had been so abused.
Then I ordered five new ones from BatterySpace.. and tested them a few times on the CBA at 25 to 10A, They have made some significant improvements since I bought the original 30 batteries. The WORST of the new ones were about 1 AH more then the best of the old batch, and the best of them was about 3.5 AH better than the worst of the old ones on the original CBA tests., ranging from 21.118AH to 22.285AH @ 20 A draw.
Physically the cases have changed from the old ones. The new ones are about 1/8 inch taller than the old ones, and the plastic ridges are differently arranged. And instead of a screwed in vent caps like in the old ones, the new ones have a glued in cap.
I can supply the CBA test files if anyone is interested enough to want to look at them.
imorton
1 kW
Wow, thanks for the update.
Looks like these are very suitable batteries for a scooter that needs 20-30a draw.
Thanks again, IAN.
PS: sorry to hear about your rain soaked (extremely drained) batteries...
Looks like these are very suitable batteries for a scooter that needs 20-30a draw.
Thanks again, IAN.
PS: sorry to hear about your rain soaked (extremely drained) batteries...
RLT
10 kW
You are welcome Ian.
They seem to be pretty tolerant batteries. I haven't cycled any of them a whole bunch of times but at least their shelf life seems to exceed everyone's expectations. I wish I had known about 'storage charge' when I put the ones away three years ago fully charged..
If you can afford the weight and bulk they are much less worrisome than LiPo and LiIon and a lot lighter than lead..
They seem to be pretty tolerant batteries. I haven't cycled any of them a whole bunch of times but at least their shelf life seems to exceed everyone's expectations. I wish I had known about 'storage charge' when I put the ones away three years ago fully charged..
If you can afford the weight and bulk they are much less worrisome than LiPo and LiIon and a lot lighter than lead..
