izeman
1 GW
this product was discussed in the long a123 thread before. surely quite inexpensive but surely nowhere as top grade as agniusm's kits. still for little money a good deal.jimw1960 said:
choose a123 20ah batteries for a matched pack
- Thread starter izeman
- Start date
izeman
1 GW
thanks. i will take that into considerationarkmundi said:
with 24s i have a lot of cells to put in the middle of the pack *ggg*
izeman
1 GW
John in CR said:
lol. 24 cells in parallel. this will be a lot of kff connecting all those cells with crocodile clamps (and not making any shorts
i guess as written some posts back: they where charged to 3.3v at the supplier and arrived all at the voltage. so they seem to have passed this test.
dnmun
1 PW
3.3V is not charged, and the manufacturer never charges the cells when they are manufactured. assembled, tested, shipped.
3.3V is where they end up after sitting around for a long time.
to bring all the cells into balance when you build a pack from scratch, you have to make sure that all of the cells are shorted together to bring them all to the same identical voltage initially. then when you assemble tha pack, you can use the bulk charger under the BMS to fully charge and balance the pack.
3.3V is where they end up after sitting around for a long time.
to bring all the cells into balance when you build a pack from scratch, you have to make sure that all of the cells are shorted together to bring them all to the same identical voltage initially. then when you assemble tha pack, you can use the bulk charger under the BMS to fully charge and balance the pack.
izeman
1 GW
ah ok. so all cells @3.3v means nothing? i read a lot about cells being at 1v or even 0v when delivered.
and there will be no bms. just bulk charging and 4x 6s balance adapters for balancing cells as needed.
i will charge them all now and connect them in parallel and see how they change voltage. thanks!
and there will be no bms. just bulk charging and 4x 6s balance adapters for balancing cells as needed.
i will charge them all now and connect them in parallel and see how they change voltage. thanks!
999zip999
100 TW
He means hook up 6 cells in parelle and charge to 3.65v and let set. Or so.
izeman
1 GW
i tried to measure/calculate IR today.
i have a 0.1ohm resistor. resting voltage is 3.33v.
i connect the resistor and voltage drops to 3.21v. this is a delta of 0.12v.
IR= DeltaV / Current = 0.12 / 31.1 = 3.7mOhm. not so bad. it really seems that the icharger can't measure really well.
3.7mOhm is an IR i can surely live with. i will do some testing with other cells.
one more question: how long should the cell keep those 3.65v? i drops really fast (some minutes) to 3.43v. is this ok? what voltage should a cell charged to 3.65v have after one hour, one day?
i have a 0.1ohm resistor. resting voltage is 3.33v.
i connect the resistor and voltage drops to 3.21v. this is a delta of 0.12v.
IR= DeltaV / Current = 0.12 / 31.1 = 3.7mOhm. not so bad. it really seems that the icharger can't measure really well.
3.7mOhm is an IR i can surely live with. i will do some testing with other cells.
one more question: how long should the cell keep those 3.65v? i drops really fast (some minutes) to 3.43v. is this ok? what voltage should a cell charged to 3.65v have after one hour, one day?
vodk1c3
100 W
Voltage should stay more or less stable at the charged voltage for good cells. If you charge at 3.65 it should be no less than 3.6x after couple of days I think.
I have some short tabbed cells from victpower and I begin testing too (yours short tab? ). Most of them will rest between 3.45-3.6.
Voltage was dropping a bit for some cell after 1 day, a bit less after 2 day and again but even less after 3 day.
I did this discharge test on a 4s pack. Plug in a 12v car inverter, 204w of lights and let it discharge to 2.8. End voltage of cells and during discharge was very consistent.
The test ran for 68 minutes. Resting voltage of the cells after 5 minutes was 3.10v.
I guess it would give me somewhat 17+ah (+-90% power efficiency on the inverter) but what is the exact formula to caculate this?
Also what did u use for resistor and could you explain more precisely : IR= DeltaV / Current = 0.12 / 31.1 = 3.7mOhm
I feel like those cell I received for victpower are somewhat not perfect (mine look like trash, ripped out of packs and alot of emboss) but still deliver some nice power.
I have some short tabbed cells from victpower and I begin testing too (yours short tab? ). Most of them will rest between 3.45-3.6.
Voltage was dropping a bit for some cell after 1 day, a bit less after 2 day and again but even less after 3 day.
I did this discharge test on a 4s pack. Plug in a 12v car inverter, 204w of lights and let it discharge to 2.8. End voltage of cells and during discharge was very consistent.
The test ran for 68 minutes. Resting voltage of the cells after 5 minutes was 3.10v.
I guess it would give me somewhat 17+ah (+-90% power efficiency on the inverter) but what is the exact formula to caculate this?
Also what did u use for resistor and could you explain more precisely : IR= DeltaV / Current = 0.12 / 31.1 = 3.7mOhm
I feel like those cell I received for victpower are somewhat not perfect (mine look like trash, ripped out of packs and alot of emboss) but still deliver some nice power.
Doctorbass
100 GW
I would suggest to match with capacity since the Ri is also affecting capacity so capacity is like both parameters together ( ri and capacity)
You can have good Ri cells with bad capacity and vice versa... but if you have good capacity usually Ri will be good..
It depend alot on the range of current you plan to use with your cells.
A cell with high RI will have his capacity to vary alot depending on the current you draw to it ( a kind of peukert effect like on the SLA battery)
On the other hand a cell with good low Ri will not be affected by the current you draw so the capacity will remain the same for all curent range.
A cell with high Ri will have a good original capacity stored into it but problem is once you draw high current.. what happen is that to output that high current the cell will have a bigger ratio of capacity dissipated in heat and lower capacity output in electrical energy.
so that's why i suggest that if you plan on drawing like 2C or lower, then match your cells in capacity.
if you plan on using a wider current range, then the Ri influence is becoming more important . so matching in capacity and Ri is the best.
search on the forum, there is few guys like e that made their own software to match cells. I made one with Mathlab wich is very accurate, but i dont ofer it for the community at the moment since i have friends also that have some right on that application too. But ther is some that are free to use and that should give great results.
Just search for cell matching or capacity matching on the forum :wink:
Doc
You can have good Ri cells with bad capacity and vice versa... but if you have good capacity usually Ri will be good..
It depend alot on the range of current you plan to use with your cells.
A cell with high RI will have his capacity to vary alot depending on the current you draw to it ( a kind of peukert effect like on the SLA battery)
On the other hand a cell with good low Ri will not be affected by the current you draw so the capacity will remain the same for all curent range.
A cell with high Ri will have a good original capacity stored into it but problem is once you draw high current.. what happen is that to output that high current the cell will have a bigger ratio of capacity dissipated in heat and lower capacity output in electrical energy.
so that's why i suggest that if you plan on drawing like 2C or lower, then match your cells in capacity.
if you plan on using a wider current range, then the Ri influence is becoming more important . so matching in capacity and Ri is the best.
search on the forum, there is few guys like e that made their own software to match cells. I made one with Mathlab wich is very accurate, but i dont ofer it for the community at the moment since i have friends also that have some right on that application too. But ther is some that are free to use and that should give great results.
Just search for cell matching or capacity matching on the forum :wink:
Doc
Doctorbass
100 GW
btw, the A123 full tab mad ein USA cells from Victpower that i got are measuring between 0.65 to 0.80 miliohm. all measured with top industrial performance Ri tester ( Hioki 3550)
[youtube]Nxi5rQnN-b0[/youtube]
Doc
[youtube]Nxi5rQnN-b0[/youtube]
Doc
izeman
1 GW
vodk1c3 said:
I have some really bad looking cells. Like being ripped out of a pack without taking care of them. Some even have red marks on them - like blood!! Some look really nice. It depends. They are all used short tab cells. Some even have those refitted tabs where a new tab is folded over the old one and spot welded.
I wanted to have and 30a load which is about the same current my bike will pull constantly.
So U=R.I or R=U/I=3.3v/30A=0.1Ohm
This gives a 30A load at 3V and ambient temp. During discharge voltage drops and so does current.
To determine internal resistance you need a big load. The higher the better as this gives more precise results. I let the I charger monitor voltage. Testing voltage was 3.32v. Then I connected the 30A load and voltage dropped/sagged to 3.20v. This is a delta/difference of 0.12V. Current at 3.20V was 31.1A. Then use the formula above to calculate IR. Hope that was clear.
I will do some more testing to see how much voltage drops from a fully charged cell.
vodk1c3
100 W
izeman
1 GW
yes same here
if doc says, that his cells are <1mOhm IR than i definately got crap
my cells sag for 0.12V @30A. his cells (or good new cells) sag only 0.022V. which is about a sixth. that's about NOTHING.
i bought those cells because i wanted NO sag. i wanted the battery to keep the voltage. only thing that calms me, is that i paid only $15. no idea how much better those cells for $20 from osn would have been. i'll never know.
if doc says, that his cells are <1mOhm IR than i definately got crap
my cells sag for 0.12V @30A. his cells (or good new cells) sag only 0.022V. which is about a sixth. that's about NOTHING.
i bought those cells because i wanted NO sag. i wanted the battery to keep the voltage. only thing that calms me, is that i paid only $15. no idea how much better those cells for $20 from osn would have been. i'll never know.
izeman
1 GW
@doc: thanks for your long explanations. even though they made me sad *ggg*. you took the risk, paid the higher price some month ago and got rewarded. as the saying goes: who buys cheap buys expensive.
izeman
1 GW
Doctorbass said:
how can that be? i have cells with 19.900mAh discharge and still IR in the lower single-digit range.
i guess it's not that bad, as my discharge current is quite low. but those cells never ever will be able to deliver 600A. even though the calculation is not 100% correct, but as an estimate:
DeltaV / Current = 3.5mOhm -> DeltaV = 3.5mOhm * 600A = 1.8V. So a charged cell with 3.4V will immediately drop down to 1.6V, which is below it's allowed limit. So what do you think: Is a paypal dispute worth the effort?
999zip999
100 TW
Izeman what are they going in and how many amp draw are they going to be use are? The OSN short tab cells are now 18.00usd.
vodk1c3
100 W
Personally I probably won't go the dispute way if the pack deliver 17ah+ @5C.
100A on a bike is already overkill. My 3540 would probably melt.
Wish I had invested a bit more for full tab also.
But who know, maybe the full tab from his newer batch are worse... greymarket is greymarket...
100A on a bike is already overkill. My 3540 would probably melt.
Wish I had invested a bit more for full tab also.
But who know, maybe the full tab from his newer batch are worse... greymarket is greymarket...
dnmun
1 PW
i just don't believe his numbers. as a manufacturing engineer, i know there is not gonna be more than .5-1% variation in the internal resistance or the capacity so when he pulls up numbers like this i just attribute it to measurement error of the little balancer he uses. i would use the doc's numbers for these cells, not the balancer.
vodk1c3
100 W
@Izeman :You should look at Steveo trike video on youtube, he had higher than expected sag from shortabbed victpower a while back. Probably bad IR too... maybe connections but I doubt seem really well made.
@dnmun : Our cell are clearly not new from factory so intenal resistance may vary...
https://www.youtube.com/watch?v=da-x9xpkTRE&list=UUgKhgvTNQEWS4fRRiVkWnyg&index=9
@dnmun : Our cell are clearly not new from factory so intenal resistance may vary...
https://www.youtube.com/watch?v=da-x9xpkTRE&list=UUgKhgvTNQEWS4fRRiVkWnyg&index=9
vodk1c3
100 W
Also this one : https://www.youtube.com/watch?v=j6l4MnPlkMs&list=UUgKhgvTNQEWS4fRRiVkWnyg
izeman
1 GW
Yes. Charger's measurement for sure was wrong. But voltage sag under load is real. And if a 1/3C load causes a high voltage drop then this is a sign of high IR.dnmun said:
izeman
1 GW
2C load max. OSN cells probably would have been the better choice. I still can start a PayPal dispute.999zip999 said:
megacycle
100 kW
Are all your cells from the same batch? From researching appears the first thing to check.
IR and capacity are complex, if you could measure them accurately with one method and they are close to each other, a suck it and see arrangement like you have done with capacity testing, but yours is pure dc draw not pulsating load like an ebike, might have different results.
Then expect your monitoring connections would all have to be made well, what would be the result of measuring at a place which is not on the tab but through a bolted connection, 0.1mOhm connection and neighbour has a loose 0.2mOhm connection, draw 90A, cell logs show inacuracies.
IR and capacity are complex, if you could measure them accurately with one method and they are close to each other, a suck it and see arrangement like you have done with capacity testing, but yours is pure dc draw not pulsating load like an ebike, might have different results.
Then expect your monitoring connections would all have to be made well, what would be the result of measuring at a place which is not on the tab but through a bolted connection, 0.1mOhm connection and neighbour has a loose 0.2mOhm connection, draw 90A, cell logs show inacuracies.
docnjoj
1 GW
After about 40 cycles or so I have noticed some sag in my A123 pack. I need to put my celllog on t find out how much, and the cells balance and hold their charge for several days of non-use. This was around a 1500-1700 peak watts on a 5 % hill. It didnt really slow down but would not accelerate even with some low gear serious pedaling. We go out tomorrow again so I will meter up and see what I can find. Total pack draw was about 13 AH for the 17 miles we rode. Maybe this is as good as it gets with these cells. They are Victpower full tab cells in an agniusm kit.
otherDoc
otherDoc
