Maglite ML50L specs for upgrading.

I'm curious what the actual inside dimension is on the latest (3rd or 4th gen?) ML50L tube both length and diameter [beyond the threaded section]. Will a 26650 with normal dimensional tolerance fit without risk of jamming? Can the cap spring compress enough to fit two 21650 in the 2cell? (I am thinking of a parallel circuit setup not series.) I read in an old thread that even slightly oversized 26xxx cells are too tight in the ML25L. But as the ML50 is newer with a longer body and switch and the ML150LR uses a LiFePO4, there may be a chance that maybe mag wasn't too daft and gave the ML50L that extra 0.5mm ID. Of course they didn't give it a true off switch so there is loss in storage, and the H/M/L selection order is backward regarding saving both night vision and charge, so maybe I'm too optimistic.
I know the tubing thickness is drastically thinner than on classic mags, which is why they changed to fine pitch threads, so boring or honing for ID is likely out.

Also, do the 50L 2 and 3 cell versions use the same driver electronics? Wondering mainly about min voltage cutoff and max allowed voltage, and which would work better for either a single cell Li-ion or LiFePO4. I no longer have access to an equipped electronics test bench and being unemployed I can't justify trial and error costs. My intent is to buy a light now because I need one, probably run alkaleaks or ni-mh, then upgrade it in the future. Just juggling the various pros and cons of the four ML25/50 2c/3c models.

I'm likely over thinking things again and by the time I get to doing any upgrades my options will have changed again and I'll be able to afford to do my own testing. So in the mean time, latest 3c 25l or 2c 50L/x... hmmm I may be resolving my own questions already but I'm going to post the tech questions anyway out of curiosity.

26650's are too fat for the ML25/ML50 and there is no way two 21700's will fit in a standard 2 cell light and more than one Li-ion will likely blow the electronics anyway. A single protected 21700's, such as the keep power 5000 mAh cell in a adapter will give longer run times in both 2 and 3 cell lights than NiMH's with much less weight. Since most 26650's have little if any greater capacity than a good 5000 mAh 21700, I don't see any real reason for bothering with 26650s.

There is always some risk running Li-ions in a 2C light, since you are running about a Volt higher than a pair of alkalines, but the 2C ML50 and the 192 lumen multi mode 2C ML25 seem to handle the extra Voltage (at least the ones I have do). I don't recommend running the old 177 lumen/single mode 2C ML25 on Li-ions because it seems to have a lower Voltage range (unlike the other 2C Mags, the single mode 2C /177 lumen version will run on a single 1.5V cell). A single Li-ion should be no problem for 3 cell Mags as the Voltage is just about perfect.

I've used power ex C cells with success.

Probably the simplest way to run these lights well is on NiMH AA eneloops or fujiloops in a DIY spacer tube - performance is good, though run time will be less than on good Li-ions or C cell NiMH's.

I went with ML50L 3 cell. The other half said no twisty switch, one handed button on the side is where it's at. And while I like 2 cell format the 3 cell run time is so much better and it really isn't much longer as it is a house light. (I have AA minimag for a pocket light.) I here the latest ML50L has a nice beam but I also ordered a diffuser lens just to play with.

Your right the 21xxx are more common and made better, balancing out the smaller package. I think I'm more irritated about why ML made the tubes so slightly undersized, it's like the ML engineer just measured some random C cell and went with that rather than using the actual IEC R14 standard(which allows max of 26.2mm), plus some allowance for heat/swelling/leaking. The only dimensional tolerance I could find for 26650 gives 25.96 +-.5 so max 26.46mm, 0.26mm larger than a C, that is 10thou. (And I found one for an 18650 that gives +-0.05mm on dia but that may be for the one brand only or a typo.)

Anyway I was mostly thinking of LiFePO4(3.2v) not other 3.6v Li-ions in the 2 cell. I never said I was going to put 2 lithium cells in series, I just wanted to know if there was enough give in the spring to physically fit them.
For the near future I won't be doing anything fancy, but at some point I'm going to test both 2 cell and 3 cell output curves and a proper bench power source and nail down just how the drivers operate, and plan from there. Hopefully I remember to post the findings.

Matt Smith (Advernture Sport Flashlights) is a member here and has some old YT videos, as well as info on his company website on Li options, battery fitment, extension tubes, dummy cells, tailcap mods, etc. on Maglites.