Wow that is really smart, with the drill press and these bolts.
They are perfectly round.
Fitting protected cells will be a problem, probably impossible on the twisty version unless I can build a recessed tailcap with the spring set back from the contact plate that makes the connection to the rear face of the tube. From the rear face of the tube, to the top of the 7135s on the driver, is only 2.71", 70mm cells measure 2.755" or so. It would fit with nearly zero clearance with a single-sided driver with no components on the battery side, but I don't have any like that. 105C converted to single-sided won't work, as there has to be a total of 12 chips, and if they're triple stacked they won't fit in the pill.
If anybody absolutely must use a long protected cell (though when the driver has low voltage shutoff, I don't see the point in a single cell light, kinda like being upset your new Ferrari doesn't have provisions for a trailer hitch) you may have to wait for tivo532's DIY east-092-style driver, which is single-sided. The 'real' east-092s still exist somewhere but they're mixed in with big batches of the low output copies, and I don't know how to get real ones without buying 100 and sorting through them, maybe ending up with 10-15 that are usable, and a whole bunch of leftover junk.
This is the space left at the tail with a 65mm unprotected cell...
...using this longer-style small spring...
I will have to use the short version of the small spring on the driver even with an unprotected cell to leave room for a compressed spring in the tail.
This is how I did it on the ~93mm long TR-801
The contact plate is electrically isolated from the tailcap, so it doesn't depend on the threads to carry any current or require the anodizing to be 100% to make sure it shuts off reliably. When the plate touches the rear of the tube it turns on, when it doesn't it doesn't. The spring is stiff enough it can be unscrewed ~1/4 turn and be used in momentary mode.
The S6 tailcap is pretty similar, with a smaller diameter thread for the original brass switch retainer and a larger thread for the battery tube. That makes it easier as there's already a step at the location where the face of the contact plate needs to sit.
Can you cut the head less than .445” to fit a longer cell? Removing less material from the head should sit higher and lift the pill higher right?
Yes, that could be done, but I would like the battery tube to contact the rear of pill directly and not just rely on the threads alone for thermal reasons.
I could fill the gap with a copper spacer with a large enough ID for the cell to pass through, I suppose... though protected cells in a light meant to be as small as possible does seem a little silly. 0:)
A single-sided driver would gain just as much internal space without increasing the outside dimensions at all. I think that will be the ultimate solution if the long cells are a must.
Sinking the driver deeper into the pill isn't an option, either - a single 7135 stands .062" off the PCB, two are .124", driver compartment is only .200" deep. That can be done on the TR-801s as the pill is deeper to start with, and the way I hollow out the pill and replace it with a new AL top gains even more space, more room to sink the driver deeper and increase the distance between driver & tailcap.
I made the cut a little long on purpose so right now a bit of the scribe mark is still visible (you can see it best in pic #2). I need to trim maybe .010" off the head, which has been compensated for - for mock-up I'm using the supplied non-AR lens which is .065" thick (the AR lens is for the final build is .061"), and I still need to take a tiny skim cut on the top surface of the pill to make sure it's nice and square and that all 5 are exactly the same height, so all the heads can be cut in one go without having to move any of the fixturing.
Nice, already looks much more compact with just the head cut down.
I'm not going to do any twisty-tail versions of the S6, the tail threads are too good. To get enough slack for it to work in momentary mode I'd have to attack the threads with a file and I really don't want to do that. :(
It wouldn't end up as short as the TR-801, anyway.
I guess with the clicky tailcap though, you can fit protected cells and mount to pocket clip further back for a deep carry.
Eh... no, not really. This light is notorious for not fitting protected cells, where it doesn't have enough space is the one piece I'm not modifying, the battery tube itself. At the tail the threads are anodized so the brass switch retainer MUST contact the rear of the tube or else you get no light. The bare tube, total overall length face-to-face, is exactly 2.75", which puts a 70mm cell flush with both ends, no space for drivers or springs to extend into the tube at all. The head would have to be extended, with a spacer between the tube & pill, to fit a cell longer than 66-66.5mm long.
I really don't want to build each of the five lights different, kinda kills the whole idea of only having to set up the tooling once for each operation and end up with five lights finished. Doing one light start to finish, then the next one start to finish, then the next one... takes forever, literally 12-15 hours for each one if done that way.
Single cell lights don't need protected cells. That's my story and I'm stickin' to it. 0:)
It’s always better to stick it in unprotected. That’s my story and I’m stickin to it 
FYI, I don’t own ANY protected 18650’s or 26650’s 
-Jamie M.
I don't think the driver for the first assembled light is working right, it won't do more than 3.85-3.9A no matter what cell I throw at it. At first I used the board with the Nichias, figured the low current might be their fault (plus the dimness was just depressing), so I used the only other 3XP I have right now which has XPG2 1As on it, and got the same numbers. If it can't be fixed by re-soldering the 4 stacked chips I don't think it will be too much of an inconvenience...
I finally got a usable amount of light out of the Nichias...
...with SIXTEEN 7135s, and it requires a INR 20R. 4.95A at the tail. Maybe I'm fooling myself but they don't seem as icky-brown when driven hard. But is that really a practical combo? It gets rather warm at that amperage...
Anybody want to play 'guess that LED/tint'?
:D
When are ya buildin mine? EH??
-Jamie M.
5B, Nichia, 4C, 3D, 1A?
Hey man, go talk to Singapore Post, it's their fault... :_(
Ordered on 10/28, I could probably order another batch of the same stuff from RMM and it'd get here sooner.
#2 & 5 correct!
#1 at the left is a bit of an unknown. PN is XTEAWT-00-0000-00000LCE7, I was able to find it's Q4 bin & 3000k, but not the actual tint.
I really want a Nichia triple but am concerned about light loss.
Comfy, how much light loss are you guestimating from the triple optics not being fully exposed?
Any chance for a head on shot of the front end? I want to see how much optic is blocked by front end.
It’s hard to tell from the angled shots as some optics look more blocked than others.
How many lumens otf from the triple Nichia at 3A?
When you say ‘icky brown’ I’m confused as the one sample of Nichia 219 I have is not icky brown.
The spectrum of tint I’ve heard described about the 219 ranges from yellowish tint to rose tint.
The one I have I would describe as slightly yellow tint but prefer rose tinted 219.
I have some samples of 119 that are my favorite tint.
Could you make a triple with Nichia 119? :bigsmile:
Thanks and great job as always!!