ENEDED

Wondering out loud here, wasn’t it said there was 10mm of driver bay area? So, with a single sided 17mm FET driver being around 3mm thick, isn’t there a LOT of room to add a heat sink press fit into the driver bay for the emitter shelf to sit on? Couldn’t this heat sink go all the way to the current pcb, (remove all components on the driver, place Katpton tape over all contacts or sand it smooth) This 10mm thick heat sink could have a 17mm driver bay cut into it, or simply a 17mm hole bored through, if done in the middle it could actually replace the current contact board and the battery positive could press into the 17mm driver itself, with battery tube negative pressing into the heat sink.

Effectively, a 70mm dia 10mm thick heat sink being introduced in place of the current driver. That would be sufficient for virtually anything we could stuff in, right?

How is the driver currently being held in place? Is it a press fit? Could I machine aluminum inserts to have a wider lip on top, sized to fit the walls of the driver bay, such that it would replace the driver and have a 17mm driver bay in the center, effectively acting like any of our smaller lights? I could add the 4mm emitter shelf onto that for a 14mm solid piece, then it wouldn’t be an issue for the two piece to have good mated contact.

I don’t know how many lights will sell in a group buy here, or how many of those would need this heavier heat sink, but I could try making one and see how it goes and if I’m able to do it successfully I could offer to do it for those wanting to push this light harder. Not sure how many I could do, and of course I’d have to make sure I could do it at all, but I’m game…. I work cheap, by the way, :wink: (well, for this lol)

Tin foil isn’t optimum, a direct contact between metals is the best, tin foil gets wrinkled and allows air space, that extra sheet of metal doubles the available air space with more surfaces. Best to have the 2 pieces in contact directly and not need a go-between.

As I understand it, the emitter shelf “plate” is already sort of a press fit, so it should be good.

Looks like a few mm's of space in there from post #124 - enough for a piggyback FET driver. At a min with no metal work, could sand down the shelf the pill top sits on, and use a good thermal grease - it would help I'd think, not great. Just the shear size of the pill top would shed some heat.

Those tailcap springs look awful - definitely need bypass wires.

The whole anodized surface issues - I'm not sure - better or worse thermally. Could also drill and tap holes and screw it down w/brass screws, if there's enough thickness in that area of the head. 12 brass screws around it - would look cool anyways...

Of course a custom copper substitute of the whole piece from Dale would be totally awesome, and secured down.

the driver can go anywhere in the space beneath the reflector, this would give us more space for the heatsink and would be extremely easy to mod and fw reflash since all the components are on top of the shelf.

this light is begging for a direct drive from 4x26650 without any electronics between emitter and cells since there is so little space inside. Just for running one mode on and off. Like those flashlights we had when we were kids. A solution to heatsinking could be using a small fan right under the pill or a copper plate attached to the pill from beneath

Got it! Packing wasn't ideal, but it seems to be in great condition so far look'n:

Full Content of the box shown below. Thin bubble wrap, 2 little cushions were at the bezel end. Would definitely prefer to see foam in the box. The box itself was ok - could be thicker, but it's thicker than the usual budget light boxes. The foam around the box wasn't covering the box ends, and the DHL envelope was the thin kind - no padding/bubbles.

(I'm more anal now about packing, after my recent experiences with thin bubble wrapping)

Condition of the light looks pristine:

Just for size comparison, the pic shows it about right (zoomed in from a higher distance):

Threads between the head and battery tube (connection for battery replacement). I think the threads are so pointed, I couldn't get them in focus:

Hmm not as big as I have thought!
Not that it is small but I just thought it would be bigger… :stuck_out_tongue:

Want some #'s? K- this is totally stock as-is. Threads came bone dry and they are not square, so I applied some Nyogel - only thing not stock, and couldn't stand the squeaky sound.

There are 4 modes: lo-med-hi-variable strobe, in that order. Appears to reset to lo in a couple secs.

Tail amps on one EFEST 4200 40/60 cell - measured with direct contact on the driver (14 AWG DMM wires):

Lo: 0.54A Med: 1.40A Hi: 2.4-2.5A (readings are moving around, generally go up over the first few secs)

Tail amps on four EFEST 4200 40/60 cells - measured by jumping the switch with the DMM leads:

Lo: 0.77A Med: 2.10A Hi: 3.58A (readings appear to be steadier)

On the one cell with no battery housing or tail, pretty loud buzzing. On 4 cells all closed up, but with the tailcap removed to expose the switch, the lower modes seemed silent.

Lumens on hi: 1020 @start, 962 @30 secs (my PVC lightbox)

Throw on Hi - measured indoors at 5 meters, as I always do:

163 kcd (807 meters)

Well, seems like a lot of potential. Who knows what XM-L2 LED it is, could be T6 or U2, and on a cheap alum MCPCB. Again all stock for now...

Do protected cells fit?

It fits tight with the unprotected ones, but the springs are long. It feels awkward tightening up with cells loaded - you can feel the cells brushing over the driver bumps. I'm think'n not to even try protected - no point to ever use them anyway - let me try...

163kcd at only 962lumens from a domed XM-L2 does indeed sound promising. How is the beam quality? any ugly rings?

Thanks for the info Tom E. I'd consider this a whopper. a shame i cant catch fish the size of this light. I wonder how many 18650's weight wise is comparable to a 26650?

The Trustfire protected 26650 cells fit fine, so probably all protected will work.

I measured 9.45 mm vertical clearance in the pill, between the driver/contact board and the pill top plate.

The head is a true 100 mm in diameter - it is huge.

The driver retaining ring is super thin aluminum - maybe 1 to 1 1/2 threads on it (spots seem to have 1 thread, but other spots maybe 2). Also, why they have a loose alum ring under the driver, I have no clue - guess they thought it was needed for better contact of the cells to the driver, so it raises the driver. Probably an after thought/fix. Would have been much better eliminating it, increasing the thickness of the driver PCB, and making a thicker retaining ring. The retaining ring scares me - too much fiddling I can see this thing falling apart. There are 6 holes for loosening/tightening it, and one I already tore one side out from trying to get it out - used a punch and light hammer taps to get it loose. I have no tool than can span 90 mm or so.

For the driver - all cell contact points (5 including the center) go thru a big bad diode each before the LED+ wire. Also there's fairly long traces on each side of the PCB for each cell. Probably explains some of the voltage drops, resulting amp drops - diodes and long traces.

For piggybacking a driver, those diodes can be removed and replaced with wires - that would certainly help. Might be able to jumper some of those traces as well.

Reupping the hype! :)

Looks very promising. Bit of a shame they cheaped on the threads.

edit: dangit.. Driver situation is a bummer

Well, think bout it - this light does stock about what a TK61 stock does, and the TK61 has an AR lens, copper MCPCB, maybe the TK61 is a little lower in amps. For the LED, dunno - TK61 is a U2 but not sure what this is.

My 163 kcd could be low - I measure only at 5m. I've seen as much as 10% bump measuring at 10m+.

if it stays cheap, it will be an interesting host - it has a lot of potential, but just as many issues that need improvement. I cant wait to see what an overdriven dedomed xp-g2 can do in this light.

Beam spill is wider than a HD2010, which is pretty wide.

Took out the pill top/shelf thing. On mine, it's just sitting there - no tight fit at all. The MCPCB is epoxied in - I got it off ok.

Stock MCPCB is 1.52mm thick, Noctigons are 1.55 mm - this is good.

Only got bout 2.6A on a single charged SANYO GA on the batt+ brass pad with shortest routes, diode jumpered, LED- wired direct to ground connection (no MCU). Not sure where the loss's are from, though could be the LED/MCPCB, or driver PCB design itself somewhere - maybe the board traces are thin, but they look pretty wide. This is a simple, crude driver for sure.

Looks easy to piggyback a driver in though.

The aluminum LED/MCPCB shelf is 4.0 mm at the thickest, 2.0 mm on the edge, 3.0mm thick in the middle area, and about 2.1mm thick in the center, under the MCPCB. Of course the thinnest parts are under the MCPCB and the edge that makes contact to the host/body.

Just to address these issues so far with my opinions on it:

  • Too thin. -> Well, it's thin but really, really wide, so lots of metal, but I'd like to see it twice as thick
  • Insufficient contact with the wall. -> Again maybe, but the radius is so darn big, there's lots of 2.0 mm shelf overlap material. Screwed down would be a good solution
  • Non screw on design. -> Threaded maybe is better, all depends on how tight, how many threads, etc.
  • Not enough head room to piggy-back driver. -> it's like 9.5 mm open vertical area - fine for typical FET drivers

That shelf is pretty vast. I mean, we’re only trying to slough off the heat from a single emitter. It’s not like we’re trying to cool down a hard driven triple.

BOSS1 does look a bit smaller than we though in your coke pic, but body is still too thick to hand-held steady grip according to wkhchin81.

There is probably nothing much to improve for most buyers on for the pill, except to strip off the coating on the contact surface and cement it to the wall.

9mm head room should be sufficient but what it people want to convert into 2S2P?

Palight is one of the few companies that keeps improving there flagship models. According to their site, their current X960 model is already a 6th generation design. I am pretty sure they will work on next gen of BOSS1 very soon.