Alkaline battery behavior when used in series question.

[AvroArrow]

Just a question for those more knowledgeable than myself regarding batteries. How do alkaline batteries behave when used in series? Basically what I want to know is will they be able to drive a boosted tri-lux setup, and if so, for how long?

Here's my proposed build setup:

Mag 3D
modamag PTS2-D
3x TX1K Lux3 stars
Fatman set for 700ma drive current
3x D cell alkalines or
4x C cell alkalines + PVC pipe holder + cut-down spring

Here's the math I've done for the build:

3 x 3.63v (Vf) x 0.7A (I) = 7.623W power output
7.623W / 0.9 (driver efficiency) = 8.47W power input
8.47W / 4.5v = 1.88A input current req'd from 3 D cells or
8.47W / 6.0v = 1.41A input current req'd from 4 C cells

So my question is how is the 1.88A drained from 3 D cells? e.g. does the driver drain 1.88A from each D cell, or does it drain 0.627A from each cell since it's running in series (1.88 / 3 = 0.627)? Also, how does the internal resistance of the alkaline chemistry affect this math?

I've looked at the Duracell Alkaline PDF spec sheets on discharge behavior so I know how the cells behave when drained at differing currents, but I need to know how much is being drained from them to figure out the rest of the math.

For those wondering why I'm using alkaline cells it is because this light is going to be for my dad and anything more complicated than loading a bunch of Duracells is not going to work (so no fancy Li-Ion or Ni-MH cell setups). The only thing I'm missing from that setup is the Fatman so I can change the driver selection of it'll make a difference, but I already have the rest of the parts and would like to finally put them to use before my T-bin lux3s become any more obselete.

 

[DonShock]

In series, the same current flows though the cells, 1.88A in your example. This is also why you connect the LEDs in series. So that even though the Vf of each LED may be slightly different, you put the same current through all of them. Alkaline cells do not handle higher drain as well as NiCad, NiMh, or Li-ion cells. So as you pull higher rates of current from them, they have a harder time suppplying the demand, and the voltage of each cell will drop more. I have run Tri-Lux3 setups before off of 3D alkaline cells successfully. I was using a Shark driver at the time and the lowest recommend Vbatt is about 1/2 of Vout. Three Lux3s give a combined Vf of around 10v, so the 4.5V from 3 alkaline D cells is close to this limit. It worked for me, but I never did a full battery run test, so I don't know if it may have developed problems as the battery drained. And I don't have a clue if the Fatman has similar recommendations. In general, I try to get my Vbatt as close to the Vf as practical when using a boost driver like the Shark to minimize power losses in the driver and maximize runtime.

 

[AvroArrow]

Ok, so all 3 alkalines will be sucked at 1.88A. Hmmm... that does not look good. According to the Duracell D cell spec sheet, at a 2A discharge level, the batteries are only good for 2 service hours before the cell is down to 0.8v. The Fatman requires a minimum of 2.7v for regulation, so once the 3x Ds are down to 0.9v each, that's it, game over man.

If I go for the 4x C setup, then the drain is only at 1.41A. And the Duracell C cell spec sheet says at a 1.5A discharge rate, the batteries are only good for 1.75hrs before hitting 0.8v. So it looks like the D Cells are the way to go... but the runtime will only be about 2hrs. Not quite what I was expecting.

I think at this point I should just scrap the driver idea and go direct drive. I have built a 3x Lux3 (TV1L) in a 3D Mag (wired in parallel) in the past and that worked out pretty well with really long runtime (more than 2hrs). Also according to modamag's old Direct Drive Reference Chart, 3x D alkalines feeding 3x K-Vf Lux3s should give them about ~600ma. I was hoping to make a brighter light this time around with slightly shorter runtimes (not 2 hrs short), but I guess that's not going to happen if I only feed it alkalines. Oh well. Thanks for the info DonShock.