Maybe if the driver had a better thermal connection to the cell than to the led it wouldn’t be such a problem. I don’t think potting the driver while still mounted in the pill will change things unless it’s the driver heating rather than the pill heating the driver. Lowering the current is certainly one solution.
If the driver to cell contact relays excessive heat, the built in thermal protection of the cell will shut down and the battery will be dead until it cools down. I’ve had that happen before. So it comes back to having the driver heat sinking into the pill or a thermally conductive mass separated from the heat of the emitter. The Luminus is just so danged inefficient it creates excessive heat like crazy. I have an SBT-70 in an X6 that’s pulling high amps like this and only making 1500 lumens. 17.1A I think, so obviously the majority of the current is being dissipated as heat. I think the data sheet calls for 10A, so perhaps it would be smarter to use 9A as the control point.
If the heat us coming from the driver then yes but if the driver is suffering heat pollution from the led then no, the driver alone shouldn’t cause heat related issues with the cell. The pill needs to conduct heat away to the surface of the host rather than through the driver to the cell. If the entire light gets hit then there’s no place to hide.
Level 6 of 7 is making 7.53A and it still messes with the mode changes within about 30 seconds on this A6 driver.
So, LD-2 at 9A or 12A?
Zoomie’s are nuts in a light box. Fully zoomed on Turbo it only shows 489.9 lumens, this pulling 17.75A on this cell right after the light box reading. Zoomed all the way out, it shows 1462.8 lumens in the box. crazy.
Is that really the amount of light loss from being zoomed in or is it the light box not being able to read the output from the zoom correctly?
I don’t know. I’ve always seen that kind of separation even in little 14500 zoomies. Used to just figure it was the way of things, but at nearly 18A I’d certainly expect more from the big emitter.
Thrower lights with reflectors don’t show like that, beats me.
light penetrating the side walls of the box at an increased rate so that the meter isn’t reading as high as it should?
I read somewhere on Google that the flux capacitor on the zoomies is 55.3456% more efficient when the light is in flood mode. :+1:
BS and more BS, man y’all are deep into it tonight! 
Sch 80 PVC P Trap, no wall penetration. Flux Capacitor? You been watching old movies or what? The Delorean still sucks.
I have recently acquired a new soldering station with hot air re-flow. So I used that to pull the 6A resistor and it was so easy it should have made me suspicious. Put the 12A resistor on, no problemo. Put the moon components on, ready to go. So, now I’ve had nothing but issues. I’m getting the 12A just fine, but that’s all I’m getting. I can find lower modes out of the light, but not reliably. And in the light it’s 12A all the way, no modes.
I’ve replaced both leads, re heated the whole board with the hot air re-flow, can’t find any issue whatsoever. But nonetheless, Turbo only.
Some days you get the bear, some days the bear gets you.
I like the new upgrades
Ok, so, I pulled the problem child and started over. This time I used the iron and meticulously made the modifications. Everything went great. Moon components, soldered the “1” circle, replaced the 6A resistor with a 12A version. This time I can get all 4 modes, but it’s really tricky. Can’t just click the switch, gotta unscrew the tail cap til it almost touches then soft contact and I can get all modes, so what is going on here? Two in a row? The FET+1 was working fine, the 20mm Noctigon is glued in with the SBT-70 on it. All I’m doing is swapping in the driver. I just don’t get it.
Hmm, pvc isn’t totally reflective so there’s some absorption even if it doesn’t show through. Yeah, I’m bs’ing in that I don’t know how much and just wondering if it causes the meter readings to be non linear since only a simple linear multiplier is used. What would you call a car made of brass?
Pulled the second L2 and re-installed the FET+1. I figured if I used 20 ga leads instead of 18 ga it would at least be slightly less lumens. I was wrong again, it now pulls 18.75A from the Efest 26650.
From Luminus datasheet SBT-70 has ~3.8V Vf at 12Amps. Worst case scenario power dissipation oh highest mode with fresh cell would be :
Pdiss~(4.0V-3.8V)*12A=2.4W
so driver with some silicone cubes or similar potting stuff should work without problems,no need for other “heavy” thermal path modifications.
To me,that look like cell button top, or fully compresed spring is making a connection to moonlight resistor or something on driver’s spring side.
The LD-2 was piggybacked onto a 28mm contact board, a thick copper sheet was used as support and ground contact at the point of the large resistor on the LD-2. The moonlight resistor and component were in between the contact board, in the air as it were, while the main component side was pressed into a silicone cube under the emitter shelf.
Similar issues with 2 different LD-2 drivers, so the FET+1 went back in. Never did figure out what the issue was.
Those drivers worked ok outside of flashlight?
28mm is pretty large pcb,maybe there is slight possibility that pcb deformed under large pressure from battery/springs when you fully screw tailcap,and some kind of short occurred between big pcb +pad/+wire and bottom side of ld-2.
I tested the driver loose and it was seemingly working ok, tested again in the pill and got some random results that I thought might be from a weak touch of the cell to a spring, inside the light there are issues and it does seem to be related to the pressure of the cell on the contact board, which in and of itself makes no sense at all. The LD-2 was solidly soldered with heavy gauge copper sheet so there was no flex involved between the contact board and driver. But whatever was amiss, it was something I was doing with the piggyback or installation as both drivers acted quite similarly.
I didn’t test either driver fully loose from the contact board, in air if you will. Was trying to avoid that due to the amperage involved.
In both cases though, I was getting the 12A the driver was set up to deliver, or, well, around 11.55A or so as I recall from yesterday. So the driver was functioning without a short to ground, I was able to set up moon after the 2 minute wait on low, but the mode selection was a problem with both tries. (ie: I could get moon and it would not shift out of moon with the switch. I could play around and get the other modes by loosening the tail cap and playing with a slight touch of the cell to tail spring but again, nothing I did at the switch would change modes, it’d only turn it on/off.)
FWIW, I didn’t mount mine solid to the contact board when I built my Jax Z1. I put it all together and tested it just “floating” in the empty space in the pill - no issues at all. Once I was sure I had it all like I wanted it I potted it fully with Ultra Gray RTV. Yeah that makes it really hard to work on later but I’m okay with that. I’ve got it like I want it and I’m leaving it alone, plus the potting will give great thermal transfer for this driver.
I used the supplied Jaxman driver board, they sent it unpopulated to use as a contact board, and it had only a slim place on the outside for ground… my 18g connection leads wouldn’t fit between the 17mm LD-2 and the inner driver bay of the Z1. So I used a rectangle of heavy gauge copper sheet, curved slightly to match the 17mm driver and folded over to a 90º on the top edge to wrap around the driver and sit on the ground ring. This top section is/was solidly soldered to the driver first, then the bottom of the copper sheet was soldered to the blank driver board/contact board. This gave a very rigid support for the LD-2 and a nice robust ground. Testing with these 2 boards joined, outside the light, it seemed to work fine. Could be that my copper sheet was interfering with the contact board sitting square on the driver bay shelf and thus not being as grounded as I expected, but the FET+1 is mounted in exactly the same way and I’m seeing 18.75A at the tail with an 4200mAh Efest 26650.