True, I didn't think of that, although it's hard for me to picture this without sketching it out or laying out the actual carriers. Ok, well just add 2 extensions (or one long one) and throw in 3 C's or 3 D's! Yes, at that point the Maglite sounds like the way to go. Besides, this host is too expensive to do so much extra modding too.
-Garry
No, no driver swap will be required! Nanjg driver is one of advantages of this flashlight, it have to remain 
As both battery carriers will be connected in parallel, voltage stays at 3.6V, capacity will double and current draw from each battery will be half >> runtimes comparison are in #1 post. Nothing else will be needed just the extension tube and DIY battery holder. Here is the picture how the holder will be made:
I agree with this, in case of 3 carriers 3p3s I would also rather mod Maglite.
Nice work. Should work very well. I always over think things. 
Or........
Buy a spare battery holder, keep three fresh AA's in it.
When the torch's batteries run low, swap it for the one in the light.
Problem solved. No custom parts, no wiring, no soldering.
Same result.
Umm . . . not if the stats in the O.P. are correct:
Look at the runtime on high comparison.
-Garry
Once you go to all that trouble (And cost), you might just want to get a member in the US, to send you a maglite. A 2D maglite will take the holders and you can have a 6AA that way. You would also be able to have a good sized heatsink too. It's just a suggestion. A 2D incan Maglite is $15USD and it should take another $15 to send. $30-$35 versus the cost of a C88 + an extension.
Also, those battery holders have a lot of resistance in them. If possible, it would be better to "make a home made holder for 6AA", or at least replace the cheap tin with copper and use gold plated springs.
It's a fine idea and I'm not saying it isn't a good idea. I'm just giving my thoughts on it.
Should be about the same, at least with eneloops. With primaries, you will get sag with three that you may avoid with 6, I agree.
I think the big advantage is sustaining the 2.8A output for much longer time (since you're only drawing 1.4A off the cells).
-Garry
I agree, and even more so if using alkalines. I just put a nanjg105c (1.4A) in the tractor supply 3xAA and run time with utilitech alkalines is nowhere near what I calculate it should be (from test discharging utilitechs at 1A w/ hobby charger). I guess they can't hack the ~40% higher rate as well as I'd hoped.
Anyway, if you wanted to make 2 battery holders (3S) into one that's 2p3s, it might be easier if you ditched the metal contacts on one end of both, epoxied the two into one and then made the appropriate connections.
anyway, Jmac is right, buy a 2D mag - I saw 2D and 3D on clearance at tractor supply this week (available in store only but listed on website).
About this mod vs. Maglite - if we consider modding maglite with XML there is this excellent mod made by jayrob on CPF, which I like very much. What I want to say is that to $15 incan maglite you have to add price of all other parts needed to mod, which results in
* Complete modified XM-L T6 Maglite - $129 (not including batteries or battery voltage monitor) Or add $25 for the Rebel build... (also not including battery holders)
Now C88 costs $45 with shipping, XM-L, good driver, pretty good press-fit heatsink which is able to divert heat from emitter well, and as was said in reviews, it is very nice looking, well build, feels good in hand, good mass for heatsinking, good beam... and personally I really like its design - simple and elegant. With extension tube attached it still has decent look, not very long (a lot of this is matter of taste of course)
Runtimes are better with maglite and two 5000mAh capacity 32600's, it's different category - but C88 with 6AA need not to be ashamed 
, it'll hold the same lumen output as good, just shorter - this is how I think about it
It is true that those battery holders have a higher resistance, but with freshly charged eneloops the default one shipped with Cyclone can give around 2amps at tailcap. As with parallel setup the overall resistance will be lower and max. current will be 1.4amp/cell - I think they will do okay as they are. But fixing them with a little solder or better springs would help to hold voltage¤t even longer. Or one can build its own 3s2p holder, no problem with that.
mio
I'd like to see a few more people weigh in on this - but if the OP is correct, I'm interested.
Jamio, JohnnyMac, Gary, dthrckt, et.al - I like the idea of carrying an extra battery holder and/or trying to mod a couple for efficiency. Can you guys tell from the eBay photos which would be the better one?
They look the same to me except for color. Both have similar construction with contact points and mods needed. Order one of each.
I think likewise, they are the same just different color and the whites are a bit cheaper. I already ordered two whites, when they arrive I'll post if they are different from the one shipped with cyclone.
Well, technically you are right about this when considering lower currents e.g. med 0.8amp. Because eneloops have very good discharging curve the difference between using one+one spare 3AA pack or one 3s2p will be very little in this case.
But we are all interested in high outputs and long runtimes, aren't we? (now with AAs)
Why wouldn't the OP be correct? It is very simple, once again example for full current 2.8A:
one AA battery pack = 2.8amps load per pack (and each battery in pack)
two parallel AA packs = 2.8a / 2 = 1.4amps load per pack (and each battery in both packs)
Then just compare on discharge graph (datasheets for eneloops are here) and look how much mAh it will give till voltage drops below LED Vin (+voltage drop on driver) at given current. Generally, the lower current draw from battery the better for battery, advantage of lower current draw with two parallel packs will be even more significant with lower quality AA, alkalines or in cold weather, when runnig the light on high with single pack wouldn't be possible, or very short only.
mio
Hello all, I had not much time last few days to write but was still working on this project to move it forward. I see there are few of us interested in this idea yet and I feel the problem is that I have no real data about runtimes and performance, nothing to show that it really works as expected. Now good new is that I've finally found a supplier who is willing to make a first prototype of the extension tube with a reasonable costs!
I estimate it will be done next week, and I'm still waiting for my two AA carriers. (I ordered almost month ago) So as soon I'll have all together and build, I'll run the runtimes tests with light output, temperature.. everyone interested hold on for a while!
I would also like to try and mod the default pill, it works good, but I'm curious if it could be better :)
So for now with a lot of not confirmed, let's wait till next week and then will see if this mod is actually worth it.
Yes, jayrob’s mod is nice, but of course it could be done for less than $50 for the whole thing, minus 3 or 4 AA batteries. Depends on who and how.
I wish you luck with it and I look forward to seeing what you have come up with. Modding is the essence of invention (I think), so good for you, for trying it out!
Yes the price is much better if you can save costs of international shipping for mag and all the parts needed. Also if you can do the parts for yourself.. Anyway, after cyclone will be finished, mag is very close to be another flashlight to mod, guess everyone here owns at least one 
Things have moved forward today and finally the ordered battery holders arrived! What is even better - the extension tube was finished meanwhile also and fits the Cyclone very well, have a look at pictures!
About the new battery holders quality - they are the same as the one shipped with Cyclone, as expected, I'm really curious how they will perform in parallel. The extension tube is not anodized yet and I will only paint it black, for testing will be good as it is. Now as soon as I build the batt.holder will do the tests and post the results!
