Powering my P7 Battery suggestions

So right now I'm powering my P7 D---I bin with a setup like this -



Using AMC drivers in parallel to give me 2.8 amps when I close both switch 1 and 2, or 1.4 amps when I close either switch separately. My battery pack consists of 2 packs of 4 AA's (these -http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=2552)

When not hooked up, fresh off the charger, I'm looking at 5.2-5.4volts off the pack. Hit switch 1, it drops to right around 4.6, and I'm getting 1.4amps almost exactly to the LED. Hit switch #2, and I'm down to 3.5-3.4 volts, and looking at maybe 2.3-2.4amps at the LED.

So would switching to 4c's fix this? (This is in a SCUBA canlight, so the batteries go in a PVC canister, form factor is not super important) I'm also only getting maybe 45 minutes of burn time out of the light before I'm only able to pull 1.4amps. I'd really like to just switch to 4D's, but that won't quite fit in my canister. Would I be able to do it with 3D's? I'm worried that the voltage sag would happen with the D's and I wouldn't even be able to get 1.4amps.

Anybody with any good suggestions I'm willing to listen. Maybe to go Li-ion? Can you wire Li-ions in parallel? If I'm going to the trouble of purchasing a new battery pack I'd like to try and get my run time up to 3 hours or so.

Any particular brands of batteries to go for/stay away from?

Thanks!

Sounds like the batteries just cant deliver.
I ran a P7 off 6 D NiMh with no issue using a KD 2.8a driver.
Paralleled 18650 Li Ions should work fine in the above situation. You should get 45 min run time per 2500ma cell you use. But Id consider using the KD driver and running 6-8 serriesd parellel 18650s at 7.4v.

There are probably much better drivers to use but these are cheap.

Going to 4xC might do it, especially with good cells, but not 100% sure. I'm running my P7 Mag on 4xD with the same drivers as yours (but one switch for both, natch), and it pulls 2.85A on fresh NiMHs (Tenergy 10 Ah). Unfortunately, I don't have a convenient way to measure voltage sag under load -- wish I'd made more measurements before assembly. I haven't done a burn-time test with it (on any batteries), but it should be just over 3 hours, and my experience using primaries supports that. (The NiMH and charger are a recent acquisition, I haven't even run them one full cycle yet.) So I'd say 4xD NiMH will definitely get you the runtime you're after.

FWIW, mine pulls 2.1A on 3 fresh NiMH and 2.2A on 3 fresh primaries, but I'd guess you get rather short runtime before it drops below 1.4A; 3xD is probably better wired with one switch left on a 1.4A driver, and the other as direct drive (or possibly resistor-drive, depending on your battery and switch resistances and LED bin).

With these drivers (or the KD 2.8; AFAICT, it's just two of these back-to-back), Li-ion doesn't work real well; you can use 1 cell and get almost no life, or two cells and burn half your energy off in the regulator, so I'd tend to avoid that, but if you do: As Packhorse said, 6 or 8 18650s (or maybe 4 or 6 25500s) in a 7.4 pack is the best option. Either use protected cells or monitor runtime strictly, because it probably won't dim at all before the cells over-discharge.

I thought there were issues running Li-ions in parallel? Thanks guys. Seems like there's still not really a good solution. If I can safely parallel 18650 cells, can I just run 4 of them in parallel and go DD?

Actually, there are more issues with Li-ion in series than in parallel, and particular with charging a series pack together (better to break them up and charge each cell separately). Parallel is fine.

As for 4p direct-drive, sure, subject to the same issues as direct driving it off one high-capacity cell. Make sure your bin is appropriate -- if Vf is too low, you'll overdrive at full charge, and if it's too high, it'll go dim even earlier than it should. (Direct drive's going to fade early anyhow.) I don't remember whether the I bin you have is the "right" bin or not, and in a dive light, you can probably overdrive it some without worrying as long as you've got good heat path to the housing, so YMMV.

And to keep your two levels, you need to hook up one switch as resistor or regulator drive; reusing one of your 1.4A regulators for that probably makes sense.

When I did my P7, I looked at it and concluded the only good Li-ion option was a 7.4V stack with a good buck convertor, and since I didn't want to spend much on a driver, and liked the ability to run on primaries, I went with a NiMH design (and naturally proceeded to run primaries for several months ). But there are a lot of people happily DDing on a 3.6v battery, so it must not be as bad as I like to think...

I am running a P7 DSXOJ Direct drive from 2 C li-ions in parallel (4.2V,6600mah) no problems here and its very bright.

AlexGT

Where did you get your Li-ion C cells? I'm starting to think that going with 4 18650's in a parallel pack is going to be the way to go. Leaving the regulators in there ought to get me two levels, and full brightness.

I got some from AW and some from Lighthound when they were available.

AlexGT