Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Behaviour of cells in parallel.
#21
Looking at the cell discharge covering time period 16th 12:00

Will try to just show charts...

 

 

 



Ideally the cells need the bus splitting so the voltages can deviate... the cross connection is hiding some data due to the common cut-off point as this changes the discharge rate of each cell through the cycle.
If you can't quantify how much they cost, it's a deal, I'll buy 5 of them for 3 lumps of rocking horse ......
Reply
#22
(08-16-2019, 07:01 PM)completelycharged Wrote: Ideally the cells need the bus splitting so the voltages can deviate... the cross connection is hiding some data due to the common cut-off point as this changes the discharge rate of each cell through the cycle.

Well yes bus splitting would give us the cells individual characteristics.
That would require 4 separate discharge modules and at this time the object of the exercise was to see how the cells operate in parallel.
I can of course discharge each cell individually for those results.

Or if someone wants to build a (Phd) Wink board like mine I will gladly share the sketch. Who knows what improvements can be made to it.

I think total cost was around $50 if that much.

PS: Looks like the battery pack that I had it hooked to has been depleted as its no longer injecting data to influx.
Well when I get home in a couple of hrs I will plug it in (got to get me a 12V to 3.3V buck) and recharge the cells that are on it and start a new series of cells.
Wolf
completelycharged likes this post
If 18 X 650 = 2200+mAh then we have power! 
May all your Cells have an IR of 75mΩ or less Smile
Last count as of 8/7/2019
Total Number of Cells Recorded and processed                 6149
Total Cells required for PowIRwall                                   2856
Total Cells ≥2200mAh, ≥80%, ≥35mΩ, ≤75mΩ, ≥4.12V   2760
For Info Google Drive
Not your average Wolf       
            Cool
Reply
#23
I had just reached a stage in my own battery journey where I was starting to wonder about parallel performance of batteries, particularly differing IR, so while I'm late to the party, I'm glad one had already started. Trying to think it through,, does it make sense that lower IR cells provide more of the Wh in a high drain scenario and that power from higher IR cells is split between meeting power draw and backstopping the higher draw cells?

In situations where IR is nearly the same and power draw is low I suspect the effect is negligible. In the case where IR is very different (mixed high/low drain chemistry) and power draw is high what happens? Can chemistry be mixed to obtain a desired high Power Drain/long power draw tradeoff? I am particularly interested in what happens to the cells when mixed chemistry cells are cycled between High and low loads. I suspect the low IR cells drain down preferentially and then are recovered by the high IR cells (Wolf's initial testing seems to confirm this), but does this mean the High IR cell is cycled more? Does this cycling accelerate cells aging?

Thanks Wolf for lifting yet another rock in the battery world. There are all kinds of creepy crawly things under it and I look forward to seeing what the community finds.
Mobilis in Mobili
 
Cell count as of 10/10/2019
234 Cells >2000mAh, >80% Rem. Cap., 14 day resting voltage >4.12V
191 Cells of Everything Else
68 In progress
 
Aiming for 8 cells tested a day
More info on my Google Drive
Reply
#24
In general, what I've observed is power cells provide most of their power in the first half of the voltage curve (where their lower IR means they have a higher effective voltage under the same load than the high IR cells), and the capacity cells provide most of their power in the second half of the voltage curve. So this is kind of the worst of both; your battery must be sized such that even the low drain cells can provide most of the load since they will be towards the end. If you are draining your pack in the >0.5C range of load, I would size it such that your weakest/high capacity cells can still handle the load.
Reply
#25
(09-09-2019, 09:20 PM)rev0 Wrote: In general, what I've observed is power cells provide most of their power in the first half of the voltage curve (where their lower IR means they have a higher effective voltage under the same load than the high IR cells), and the capacity cells provide most of their power in the second half of the voltage curve. So this is kind of the worst of both; your battery must be sized such that even the low drain cells can provide most of the load since they will be towards the end. If you are draining your pack in the >0.5C range of load, I would size it such that your weakest/high capacity cells can still handle the load.

Ok.  reading through some of the test results from Wolf and others I'd agree that seems to be the case.  It could be possible to determine a custom discharge cut-off voltage for the mix to avoid over-drawing on the high capacity cells at the end of charge, but I suspect it would limit the battery life so significantly that there would almost be no point.  If this is true there may be a niche application where the end use requires a couple short spikes of high power that is not feasible to handle with high capacity cells in parallel, but not enough spikes to render the additional life of the high capacity cells pointless.

This is promising for me because I am looking at using Li-ion for an EV, but the thought harvesting the required cells to attaining decent voltage with 300-400p packs is daunting.  Perhaps by making each pack 150p and 10% high drain cells I can manage the acceleration spikes without significant loss of total capacity.
Mobilis in Mobili
 
Cell count as of 10/10/2019
234 Cells >2000mAh, >80% Rem. Cap., 14 day resting voltage >4.12V
191 Cells of Everything Else
68 In progress
 
Aiming for 8 cells tested a day
More info on my Google Drive
Reply
#26
(09-09-2019, 09:47 PM)Nemo Wrote: This is promising for me because I am looking at using Li-ion for an EV, but the thought harvesting the required cells to attaining decent voltage with 300-400p packs is daunting.  Perhaps by making each pack 150p and 10% high drain cells I can manage the acceleration spikes without significant loss of total capacity.

To put some numbers on this - for the sake of discussion....

My system is 7kw PV,  40kwh 18650 battery bank (14s360p), and 12,000watt inverter.  The max charge is  48v@120a (333ma per cell).   The max discharge is 48v@250amps (695ma per cell) enforced by a 250a circuit breaker.  The system has not yet gone beyond 50% max (340ma per cell) and on average runs 250ma / cell on discharge.

If you're doing 300-400p @ 14s then you're battery bank would likely have similar numbers.   What kind of 'spike' are you envisioning?   Even at 48v@500a it would only be 1,400ma per cell - i.e. high but not outrageous and you could plan to increase the battery size.
Reply
#27
OffGridInTheCity Wrote:If you're doing 300-400p @ 14s then you're battery bank would likely have similar numbers.   What kind of 'spike' are you envisioning?   Even at 48v@500a it would only be 1,400ma per cell - i.e. high but not outrageous and you could plan to increase the battery size.

I'm not exactly sure what to expect, but spikes up to 600 amp seem reasonable for an EV system (I think controllers could limit this, still learning).  How long these spikes last is also important, and I don't have a sense of that yet either.  EV also typically operate at 72V+.  lets say 72V, that ends up as roughly 20s300p = 6000 cells.  At my current rate of supply this would take years.  If you want to continue talking EV battery design head over to this thread https://secondlifestorage.com/showthread.php?tid=3712 so that the conversation here can stick to the behaviour of cells in parallel.
Mobilis in Mobili
 
Cell count as of 10/10/2019
234 Cells >2000mAh, >80% Rem. Cap., 14 day resting voltage >4.12V
191 Cells of Everything Else
68 In progress
 
Aiming for 8 cells tested a day
More info on my Google Drive
Reply
#28
Let me explain my experience I had one good battery and one dead battery if you place a dead battery with good battery for interval of 10 seconds remove it ,once again place it for. 10 untill you get 2-3v. I revived Samsung but couldn't revive LG as it was burning the good cells .Even though if you revive you get only 400-700 mah .please place if the cells are of same pack .
Reply
#29
(09-17-2019, 09:49 PM)Henith10 Wrote: Let me explain my experience I had one good battery and one dead battery if you place a dead battery with good battery for interval of 10 seconds remove it ,once again place it for. 10 untill you get 2-3v. I revived Samsung but couldn't revive LG as it was burning the good cells .Even though if you revive you get only 400-700 mah .please place if the cells are of same pack .
I believe what you are referring to is called "bump charging" which can be used to get a 0v cell up to a voltage that a charger will recognise and charge the cell.
The process to try to recover 0v/very low voltage cells which is most recommended here is for very low currents to be put into the 0v cell.
Bump charging could have higher currents which are not recommended.
This thread is about something completely different to bump charging.
This thread is about investigating how different types of cells (ie high current power tool cells vs low current laptop cells) behave when used alongside each other in cell packs.
Ibiza likes this post
Reply
#30
Alright after a short hiatus I am back, testing cells in parallel.

This test is on 4 
7222 LGABD11865 4.20 48.10 2451
7223 LGABD11865 4.20 47.50 2444
7224 LGABD11865 4.20 47.50 2560
7225 LGABD11865 4.20 46.20 2445
2A Discharge to 2.8V


These are 3000mAh cells with a reasonable IR and right on the edge of 80% SOH.
So far so good. I am running a discharge trace on them right now and when done I will run a charge trace.

The chart so far:


Quick update:
Pretty uneventful as the cells are relatively evenly matched.
Slight rise in cell temp at the end of the discharge.


Wolf
If 18 X 650 = 2200+mAh then we have power! 
May all your Cells have an IR of 75mΩ or less Smile
Last count as of 8/7/2019
Total Number of Cells Recorded and processed                 6149
Total Cells required for PowIRwall                                   2856
Total Cells ≥2200mAh, ≥80%, ≥35mΩ, ≤75mΩ, ≥4.12V   2760
For Info Google Drive
Not your average Wolf       
            Cool
Reply


Who read this thread?
100 User(s) read this thread:
daromer (10-09-2019, 04:35 AM), Sean (10-08-2019, 10:57 PM), mike (08-31-2019, 10:42 PM), 18650Shrink (08-16-2019, 11:46 AM), Majorphill (09-11-2019, 01:54 AM), Korishan (10-11-2019, 10:08 PM), A + A (08-14-2019, 08:30 PM), PAF (09-12-2019, 08:25 AM), jdeadman (10-09-2019, 03:17 AM), wim (10-03-2019, 02:03 PM), watts-on (10-08-2019, 10:39 PM), Jack Daniels (08-31-2019, 06:07 PM), Hanssing (09-17-2019, 11:23 PM), Stefaan De Ridder (08-17-2019, 09:47 AM), Geek (10-09-2019, 01:13 AM), SWEwall (08-31-2019, 08:16 AM), Jon (08-16-2019, 06:02 PM), Ivo Staelens (08-16-2019, 08:02 PM), Grumplestiltskin (08-19-2019, 06:04 AM), emerca (08-17-2019, 03:12 AM), ChrisD5710 (10-03-2019, 02:40 PM), Sonic01 (09-19-2019, 09:46 PM), Tentious (10-22-2019, 04:26 AM), Oz18650 (11-02-2019, 01:00 AM), rev0 (10-08-2019, 09:53 PM), AndyMc1970 (09-16-2019, 11:11 PM), djuro (09-18-2019, 05:31 PM), jm1 (08-16-2019, 11:26 PM), drchips (09-18-2019, 11:34 AM), HughF (08-15-2019, 03:09 PM), benwilliam (08-15-2019, 02:52 PM), gpn (10-09-2019, 03:23 AM), Crimp Daddy (09-10-2019, 12:45 AM), jonyjoe505 (08-15-2019, 02:58 AM), completelycharged (10-17-2019, 05:28 PM), Ibiza (10-04-2019, 08:29 PM), thunderheart (08-16-2019, 07:38 AM), Bubba (10-08-2019, 11:34 PM), Cherry67 (09-12-2019, 05:02 AM), Walde (08-17-2019, 10:18 AM), Riplash (09-11-2019, 01:21 AM), Jim Jr. (10-03-2019, 12:12 PM), djotter (10-08-2019, 08:45 PM), Buzz0515 (09-18-2019, 12:40 PM), chuckp (10-09-2019, 01:44 AM), clupton (08-20-2019, 02:30 AM), Leipi35 (08-14-2019, 03:19 PM), Wolf (11-01-2019, 11:24 AM), Scottietheyoung (09-10-2019, 10:22 AM), Generic (10-04-2019, 03:37 AM), Solardad (08-16-2019, 10:58 PM), stevelectric (08-15-2019, 08:00 PM), ajw22 (10-03-2019, 02:30 AM), kevinjaye (10-11-2019, 02:06 AM), OffGridInTheCity (10-08-2019, 11:37 PM), Maniac_Powerwall (09-01-2019, 06:02 PM), Rhoos (09-09-2019, 02:47 AM), Overmind (09-24-2019, 08:45 AM), intra (10-03-2019, 11:17 AM), winny (09-13-2019, 08:44 AM), ironzap2 (08-16-2019, 08:25 PM), 100kwh-hunter (09-12-2019, 11:03 AM), Doin it (08-30-2019, 03:45 AM), gpearce52 (08-19-2019, 09:42 PM), Marrcin (09-22-2019, 05:12 PM), foamovercast (08-30-2019, 01:44 PM), Nemo (09-18-2019, 02:37 PM), Regalo (11-10-2019, 09:49 AM), bartk (08-14-2019, 09:04 PM), mackoffgrid (09-10-2019, 05:21 AM), smileypete (08-18-2019, 10:49 AM), solarsteve (10-05-2019, 07:01 PM), Jonisingt35r (09-18-2019, 03:35 PM), 3xtr3m3 (08-16-2019, 12:27 PM), LPG (09-26-2019, 01:19 AM), Leon240d (08-17-2019, 02:34 AM), dakiller (09-12-2019, 03:49 AM), berthfield (08-19-2019, 03:22 AM), Funkadelicious (08-24-2019, 03:12 PM), Bmr4life (08-19-2019, 08:36 PM), hschlech (08-28-2019, 04:06 AM), James-w (09-18-2019, 08:51 PM), Sleonard2649 (09-10-2019, 04:41 AM), Rabbithacker921 (09-18-2019, 07:59 PM), mdwool01 (09-06-2019, 02:21 PM), DirkB19 (10-15-2019, 09:12 PM), Eduardo Kang (09-19-2019, 08:13 PM), Robert024 (09-17-2019, 01:32 AM), EnergyOasis (09-27-2019, 01:22 AM), Niyotek (09-14-2019, 10:41 PM), huat (09-17-2019, 09:51 PM), Henith10 (09-17-2019, 09:49 PM), Ramos1970 (09-22-2019, 02:22 PM), Smarky (09-26-2019, 04:59 AM), mihit (09-25-2019, 06:34 AM), BatteryMooch (10-03-2019, 03:41 AM), Victor8 (11-07-2019, 11:48 AM), barry (10-21-2019, 08:17 PM), ttbloker (Yesterday, 04:55 PM), msdmaddog1 (11-11-2019, 08:34 PM)

Forum Jump:


Users browsing this thread: 1 Guest(s)