IMAA's Marez on parallel battery systems

More rainy day fun OK some testing is in order.

Pack A 617 mAh actual when discharged at C/5 to 3.6 volts Pack B 612 mAh actual when discharged at C/5 to 3.6 volts

Charge A @ 500 mA to cut off OCV after 15 min rest = 5.72 volts Discharge B @ 500 mA to 3.6 volts.

Connect A in parallel with B and record current. Current at start - 290 mA Current after 8 minutes - 60 mA Area under the curve = 17 mAh

Discharge B @ 500 mA to 3.6 volts after 24 minutes connected to A. Capacity delivered 27 mAh

Pack A open circuit voltage after 15 minutes rest - 5. 23 volts Pack B open circuit voltage after 15 minutes rest - 4.68 volts

Amount of energy drained from A in 24 minutes - 4.3%

Leaving you with 95% of the good battery to support flight - if you FAILED to do an ESV pre-flight and if you FAILED to charge the other pack in the first place.

Next test to see how much current drain a 3 cell pack puts on a fully charged 4 cell pack (simulating one cell shorted - of course trying to blow

3 amps through a shorted cell would in most likelyhood blow the short away and you are back to the above scenario)

-- Red S. Red's R/C Battery Clinic

us out for "revolting" information.

Yo, Red-

This is the same guy that 'corrected' misinformation spread on the internet by informing us in his MA column that signal demodulation in our R/C Rx's is performed by the decoder.

Abel

| More rainy day fun | OK some testing is in order. | | Pack A 617 mAh actual when discharged at C/5 to 3.6 volts | Pack B 612 mAh actual when discharged at C/5 to 3.6 volts

To be fair, since Marez is an IMAA guy, he's probably talking about larger batteries than AA packs.

| Charge A @ 500 mA to cut off OCV after 15 min rest = 5.72 volts | Discharge B @ 500 mA to 3.6 volts. | | Connect A in parallel with B and record current. | | Current at start - 290 mA | Current after 8 minutes - 60 mA | Area under the curve = 17 mAh

Of course, the current flow will depend _a lot_ on the internal resistance of the batteries and on the resistance of the wiring between them.

Ironically, the eventual (given enough time) `area under the curve' should not :)

| Leaving you with 95% of the good battery to support flight - if you | FAILED to do an ESV pre-flight and if you FAILED to charge the other | pack in the first place.

Sounds correct to me. The results would be totally different for LiPo cells, but then again, you can't plug LiPo cells directly into most receivers without a voltage regulator anyways.

As for the 95% left, don't forget that the `dead' battery didn't throw that 5% away -- it was being charged. Only a small percentage of that

5% is actually lost. | Next test to see how much current drain a 3 cell pack puts on a | fully charged 4 cell pack (simulating one cell shorted - of course | trying to blow 3 amps through a shorted cell would in most | likelyhood blow the short away and you are back to the above | scenario)

This test would be more interesting. Since a 3 cell NiCd/NiMH pack would probably be charged at 4.2 to 4.5 volts, the good pack would not stop (over)charging it quickly like it did in your first test. The rate would depend on the internal resistance and the wiring. Eventually the good pack would discharge to around 4.2 to 4.5 volts, which hopefully is enough to power your receiver for a little while.

[ Having a low battery alarm would be good in this case. ]

But this isn't a very common failure mode anyways, and you're right -- a high amperage would probably unshort the cell anyways.

| > The Cermark Company to the rescue - it has recently introduced its | > new 'Power Backer', which takes a secondary normally less capacity | > battery

Sounds like a paid endorsement to me.

If you wanted to do it really cheap and yet not worry about cross-charging, you could get two diodes and two 5 cell packs (5 cell to give you some extra because of the voltage loss of the diode) and do it that way. Of course, you'd need diodes that could handle large discharge rates. I'm not even sure this is worth the extra complexity, but maybe.

The big advantage that his `Power Backer' gives that I see would be that you know if something went wrong without having to measure yourself.

About three or four years ago I ran this test to satisfy myself. I took a three cell pack and paralleled them and measured the current drain. There was an 80 mah current flow from the 4 cell pack to the three cell pack. The voltage dropped to about 4.1 volts IIRC. As Red says, it takes a lot of voltage difference to charge a Nicad and one cell difference is not a problem.

Contrary to another comment, the pack size would not effect the out come, it's a voltage thing. I tested equal packs and unequal paks such as a 300 mah in parallel with a 700 mah pack etc. It's been so long I don't remember all of the scenerios I tested, but I have many planes outfitted with dual packs and dual switches. I build my own switch harness with 4PDT switches that disconnect both the battery leads. This allows a special charge harness to series the two packs so they can be charged as an 8 cell pack.

Dan Thompson

| Contrary to another comment, the pack size would not effect the out | come, it's a voltage thing.

You must be referring to my comment, and if so, you apparantly misunderstood what I was saying. Maybe I wasn't very clear.

(According to Red) Marez said 3 amps flowed between the empty and the full battery pack when hooked up. Red measured it, and found only 290 mA flowing.

The difference is probably that Red appears to have done his test with

600 mA NiCd AA cells, and if Maraz measured it, he probably did it with subC or so cells. He probably also used thicker wire, as you'd use on an IMAA plane. That could explain the discrepancy between 300 mA and 3000 mA -- subC cells have a much lower internal resistance.

I didn't dispute Red's outcome at all, except by saying that even with that 5% `lost', most of it wasn't really lost.

In article , Doug McLaren wrote: | | The difference is probably that Red appears to have done his test with | 600 mA NiCd AA cells, and if Maraz measured it, he probably did it | with subC or so cells. He probably also used thicker wire, as you'd | use on an IMAA plane. That could explain the discrepancy between 300 | mA and 3000 mA -- subC cells have a much lower internal resistance.

Also, Maraz said `the good battery charged the defunct battery at 3 amps'. Since it sounds like he's trying to sell these circuits, he could have picked a battery with several shorted cells and still not be `lieing' (though it would be misleading, since this almost never happens.)

Red testesd one specific form of `defunct' battery -- not being charged (the most likely flaw, to be fair) -- and found that it wasn't really a big deal. Another common problem (perhaps the next most common problem) is a cell that fails open -- in this case, it's pretty obvious that the cross-charge rate will be zero -- no problem there.

If one cell fails shorted, then there's some room for a problem, and Red said he'll test this next. I suspect that the charge rate will be a good deal higher but that it won't burn things out, and I suspect you'll still have 15 minutes of flight time to land your plane in most cases. (Might be a problem if you're going after your 8 hour LSF Slope Level 5 task, however.)

If more than one cell fails shorted at once, then there's a big problem, but this is so incredibly unlikely to happen all at once in a pack that's not been seriously damaged (like in a crash) it can pretty much be ignored.