1st Gen. Kokam 3270 discharged in parallel with a 3rd Gen. TP ProLite 2000 - RC Groups
Thread Tools
Apr 17, 2005, 12:59 PM
Registered User

1st Gen. Kokam 3270 discharged in parallel with a 3rd Gen. TP ProLite 2000


The graphs contrast the useable capacity of the cells at each of two discharge rates.

For this test, total current into both cells was set by the West Mountain Radio CBA, used as the load. Individual cell currents were measured with a Sears Pro clamp-on ammeter, voltages with a Fluke DVM.

Cell amperages total a little less than the CBA amperage, possibly due to variation between the CBA and the Sears and perhaps due to wiring losses. In any event, the distribution of current between the cells is the measurement of interest here, as are comparative voltages. Since all individual cell amperage and voltage measurements were taken with the same instruments, the distribution proportions are accurate, even if the absolute value of the measurements may not be.

At a five-amp combined discharge rate (CBAís figure), the current distribution was: Kokam 2.2, TP 2.7 (readings rounded to nearest tenth). Voltages were identical at the tabs of each cell, during all the discharges.

At 10 amps CBA: Kokam 4.6 amps, TP 5.1

At 15 amps: Kokam 6.7, TP 7.8

At 20 amps: Kokam 8.7, TP 10.8

At 25 amps (CBAís power dissipation limit): Kokam 10.7, TP 13.9

From this point on, the cell combo was run down to 3-volt cutoff and the cells immediately disconnected and allowed to stabilize. The stable voltages were: Kokam3.82, TP 3.45,
but this difference does not imply any harm, because neither cell went below 3 volts.

Without re-charging, the cells were then re-connected in parallel and again discharged at a CBA setting of 5 amps.

Distribution now: Kokam 4.3 amps, TP .4

The discharge continued until the total amp hours discharged was approximately the sum of the rated capacities of the two cells. The residual resting voltages were: Kokam 3.4, TP 3.1

I conclude that these two cells in parallel will discharge at a rate according to some function of their internal dynamic impedances, probably involving their capacities, as well. If it were simply proportional to capacity, the Kokam would have discharged at a higher rate than the TP.

At relatively high discharge rates, an arbitrary cutoff voltage (3 volts) will cause the combo to discharge less than the sum of their rated capacities. At relatively lower discharge rates, the entire sum of their ratings will be discharged.

No damage is done to the cells, provided that the maximum discharge rating of the higher-rated cell is not substantially exceeded, because neither cell goes below 3 volts.


- RD
Sign up now
to remove ads between posts
Apr 17, 2005, 01:58 PM
Registered User
IA-Flyer's Avatar
Great test RD!
Did you graph the parallel cells toghether when they were discharging?
Just curious if a voltage drop "step" could be seen near end of the discharge, after the TP dumped and the Kokam took over the load?
Thanks again!
Jim
Apr 17, 2005, 02:12 PM
Permanently Banned
Larry3215's Avatar
Excelent data RD.

I'll post my results shortly - have some family duties to attend to...

Larry
Apr 17, 2005, 04:05 PM
Registered User
Jim,

I saw the graph of each discharge but didn't keep any, because I was using the CBA merely as a selectable amperage load. I do remember that each curve was smooth and nothing looked out of the ordinary.

The new curves I graphed for this thread were of the Kokam 3270 at 8 and 12 amps, to match the discharge rates of the TP cell's curves, which I had stored a few days ago.

- RD
Apr 17, 2005, 08:48 PM
Registered User
Jim,

Here's a complete graph of the combo @ 10 amps.
I forgot to disconnect the cells, so I didn't get to measure the resting voltages, but the Kokam's would have been higher than the TP's. Sorry. Note that the combined mah discharged is not C1 + C2, 5,270 mah, because the Kokam @ 10 amps cannot discharge its full C. Interpolate on the Kokam graph in the first post, between 8 & 12 amps. It drops to 3 volts well short of its capacity.

- RD
Last edited by RD Blakeslee; Apr 17, 2005 at 08:53 PM.
Apr 17, 2005, 08:54 PM
Permanently Banned
Larry3215's Avatar
Very interesting graph RD. Very informative how the TP's contribute more at the beginning and then the Kokams take over at the end and yet neither pack goes below 3.0 volts.

I'll put my results in the next few posts.

Your graphs make me really wish I had a CBA and a decent clamp-on meter

Larry
Apr 17, 2005, 09:02 PM
Permanently Banned
Larry3215's Avatar
First of all a disclaimer - I DO NOT recomend running packs in paralell that are of different capacity or brand or age under hi loads.

Next point to consider in evaluating my results - I do not own a decent/reliable clamp on meter or a CBA and my DVM is only mid grade and NOT a Fluke. In my opinion it is reasonably close tho.

The only direct amp reading device I own is a watt meter and I did not use it in line with the individual "packs" because it introduces excess resistance into the circuit and throws off the results - more on that later.

My tests and equipment only allow me to infer an average amp draw for each
pack over the period of the discharge based on the mahr durring re-charge.

Rd's results on the other hand show the actual current delivered by each cell durring the discharge. The bottom line is that RD's testing setup and equipment are superior to mine.

One more point to consider - my packs were different from RD's and that probably has a lot to do with why my results are somewhat different from his.

My cells were 10-12C rated Tanics and 8-10C Irate cells. I suspect (but have no proof) that the packs I used have closer IR values than the cells RD used.

Data in the next post.

Larry
Apr 17, 2005, 09:09 PM
Permanently Banned
Larry3215's Avatar
OK here is the data - sorry no graphs since I also dont own a CBA - yet

First test used a single 6S2P 3100 Tanic pack (1550 cells) in paralell with 2ea 3S1P home made Irate 2200 packs in series. The Irate packs are about 1 1/2 years old and the Tanic pack is about 10 months old. Both packs have lots of cycles on them.

I discharged the combined "pack" at a starting rate of 15 amps and quickly went up to 30 amps (max with this motor/prop combo at 6S) monitoring pack voltages along the way.

At all times the individual packs had exactly the same voltage.

About 75% of the discharge was done at 30 amps.

Durring the first run I ran out a total of 2410 MAHR on the watt meter (placed in-line with the combined pack - NOT one of the single packs).

I stopped the discharge when my motor got to 150 degrees and imediately disconected the packs so they would not equalize each other. After waiting 5 minutes resting voltages were less than .01 volts apart. 23.0 on the 3100 pack and 23.1 on the 2200 pack.

Recharge on the 3100 took 1415 mahr and the 2200 took 950 mahr.

In this test the two packs delivered a total mahr roughly proportional to thier relative capacities - the 3100 pack delivering the larger portion by roughly 5% depending on how you do the math - based on delivered mahr vrs nominal capacity.

The next test used the same 6S Irate 2200 "pack" (2ea 3S1P 2200's) in paralell with 1ea 2S1P Tanic 850 and 1ea 4S1P Tanic 850 - A 6S 2200 pack paralelled with a 6S 850 pack.

This is the biggest difference in capacity/brand/age I could come up with.

The 850's are less than 6 months old and the 2200's are the same 1 1/2 year old ones from the other test.

In this test I was able to exceed the nominal 10C rate for either pack - I went to 30 amps again on the discharge.

I didnt let the motor cool all the way so I had to end the discharge a little sooner - 2010 mahr total - when the motor got to 150 deg - my personal limit.

After the discharge I again disconected the packs imediately to prevent any equalizing. Resting voltages were 22.4 for each pack - exactly ther same.

The 2200's took back in 1428 mahr and the 850's took back in 537 mahr.

This time the packs relative contributions were even closer to thier relative capacities than in the last test.

Note that neither of these tests went all the way to LVC for the jumbo "pack"
- however as I noted in this thread

https://www.rcgroups.com/forums/showthread.php?t=360087

some of the individual cells within BOTH sub packs DID go below 3.0 volts per cell in the second test. But thats a different issue and a different thread

My conclusions are very much in-line with RD's. Packs discharged in paralell seem to follow some function of thier Ir in combination with individual pack capacity. RD's examples look to me to follow Ir and mine seem to follow capacity more. But since we dont know exact Ir values of any of the packs/cells used this is just a wild guess/assumption on my part.

In either case - this looks to be a relatively safe thing to do if not carried to extreems - however I still DO NOT recomend it.

The following post goes into more detail as to why.

Larry
Apr 17, 2005, 09:12 PM
Registered User
Larry, the ante's only around $150. Join the game! Tomorrow, I'll do a 20 amp graph. The TP will put in a much larger early proportion at that amperage, I think. The Kokam might not contribute much at all - well see. If so , the combo will probably discharge less than 4 amp-hours before it goes down to 3 volts.

- RD
Apr 17, 2005, 09:13 PM
Permanently Banned
Larry3215's Avatar
I did one more test dischargeing dis-similar packs but this time I introduced a deliberate - but still small - extra resistance in the leads of one of the packs to see what the impact would be on the relative loads and voltages on each pack.

This was intended to represent a "worst case" for what can happen under hi load paralell discharges.

I chose the same packs as in the 2nd test above because they were the closest match as far as relative discharge contributions.

I placed my watt meter in line with the 850 "pack" and used two extra sets of Aderson Power pole conector blocks to wire this pack to the load. The 2200 "pack" was conected as close to the load as possible with only a single conector block.

The results come a lot closer to what RD saw in his test as far as relative amp contributions. The big difference is in the relative voltages of the packs under load.

The 850 pack stayed at 22.2 volts durring a brief 30 amp run but the 2200 pack dropped to 18.6 volts. Thats 3.1 volts per cell for the 2200 pack and 3.7 for the 850 pack.

My conclusion here is that if your conectors/conections are not in good shape AND closly matched for each sub-pack - you could easily over discharge one of the packs when run in paralell at hi loads.

The extra resistance in the 850 side caused a gross difference in the way the packs shared the load. This forced the 2200 pack to deliver way more than its fair share.

I am quite certain this would apply even if the packs were of the SAME capacity.

Larry
Apr 17, 2005, 09:16 PM
Permanently Banned
Larry3215's Avatar
Quote:
Originally Posted by RD Blakeslee
Larry, the ante's only around $150. Join the game! Tomorrow, I'll do a 20 amp graph. The TP will put in a much larger early proportion at that amperage, I think. The Kokam might not contribute much at all - well see. If so , the combo will probably discharge less than 4 amp-hours before it goes down to 3 volts.

- RD
hehehe its all "relative" RD

I OWE big time on taxes this year so Im "relatively" broke at the moment and what excess hobby funds I have are already ear marked for a soon to be released DR1 and motor and esc

However - they are in the budget for the near future!

Larry
Apr 17, 2005, 09:24 PM
Registered User
IA-Flyer's Avatar
Thanks for sharing the graph & results RD, thatís extremely interesting!
Like you were saying, it looks like a combination of lots of things!
Iíd "almost" guess this is like looking inside a single cell that has individual cells (if thatís what theyíre called) of different capacity and load handling abilities.
The sharp knee is gone, but the current difference sure doesnít look like a problem in your test.

Thanks again for doing this test. I have a headache now, and probably will for quit sometime thinking about this!

Jim
Apr 17, 2005, 09:31 PM
Registered User
IA-Flyer's Avatar
Quote:
Originally Posted by RD Blakeslee
Larry, the ante's only around $150. Join the game!
- RD
Apr 18, 2005, 07:32 AM
Registered User
Jim, The knee is gone because the Kokam is supplying most of the current toward the end of the combo's discharge and it doesn't produce a knee. Again, interpolate the last part of the Kokam curve in the first post, between 8 and 12 amps.

Twenty amp graph will be up shortly.

- RD
Last edited by RD Blakeslee; Apr 18, 2005 at 07:49 AM. Reason: accuracy, clarity
Apr 18, 2005, 09:30 AM
Registered User
Quote:
Originally Posted by RD Blakeslee
Larry, the ante's only around $150. Join the game!
It's not just a matter of money. It's a matter of time for some. Many of the more prolific testers are retired with a lot of free time on their hands. This is wonderful, and something I look forward to myself.

Some people have jobs or other responsibilities that can take up 12-14 hours a day, including commute. They have little free time, and to devote large chunks of it to testing anything requires tremendous dedication and sacrifice of other activities. I tried this last year with motor/prop testing, and it really eats up your free time.

I have great respect and appreciation for all those who take the time to do this type of testing and then share it with all of the forum members here. For those who have less free time available to do such testing, it's really helpful that those who do are willing to share.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Charging in parallel with the Cellpro 4S possible? JT_Redmist Batteries and Chargers 4 May 25, 2007 04:45 PM
Kokam super 30 ESC with 2 motors in parallel? cometravi Power Systems 0 Mar 14, 2005 04:31 AM
Can NiMh be DISCHARGED in parallel? stu78 Batteries and Chargers 2 Oct 22, 2003 11:54 AM
Kokam in parallel ? soren Batteries and Chargers 1 Apr 20, 2003 07:12 PM
Kokam 2-cell packs in parallel SoaringDragon Batteries and Chargers 1 Dec 31, 2002 09:19 PM