Well, I enlarged the area where the contact board sits, it’s pretty much press fit for 21.5mm now. I also reduced the size of all my crappy solder points from the first attempt. Reflector no longer contacts the blobs.
It still shorts out though. Could anything on the driver be shorting against the inside of the pill? It’s not as snug as before, but it is still slightly slanted at about a 30 degree angle.
Either power lead could be shorting out from the driver to the light body. They solder to the board very close to the outer edge. Check those and insulate them with shrink wrap tubing if you can.
Do you have any larger shrink wrap tubing? Like for a battery? If you do, you can cut a disk to cover the pcb with a hole in the middle for the emitter, try to keep the hole tight to the centering ring or the emitter base itself and cut the outer diameter to completely cover the star. This will help ensure nothing touches the top of those connections as sometimes that’s a very tough place to protect.
I’m having a tough time visualizing this. Did you mean star instead of the pcb? Otherwise I’m lost.
The star is the pcb. Well the pcb is stuck on top of star. I think DBCstm is applying you need to make something like this to cover the led pcb connections.
http://www.fasttech.com/products/1616/10001923/1194800-16mm-insulation-gasket-for-cree-q5-led-emitter-10-
You could also cut the center out in the shape of a square to fit perfectly over the led, like this.
http://www.intl-outdoor.com/cree-xml-insulator-gasket-10-pcs-p-463.html
Sorry, yes I meant the star, been doing too many different things and got sidetracked lol. Basically, cover everything on top of the pill that’s not going to show through the hole in the reflector.
ah, ok. I see now, thanks. I’ll heatshrink the wires and see what I can do about making an insulation gasket.
Keep in mind that it may very well get over 200º F at the pill, perhaps hotter. Maybe even close to 300º. So unless you want something melted all into your pill, it would be pertinent to keep the heat factor in mind.
That being said, I’ve been using a piece of heat shrink tubing and haven’t seen any signs of it shrinking down from heat so far. And I’ve used it in some lights that are run pretty hard, like the K3 with MT-G2 at over 3000 lumens.
Alright so I went ahead and disassembled it all again. I got to looking at the underside of the driver and noticed I did a pretty crappy job and may have put the negative wire in the wrong place. I’ll upload some pics asap.
I know the Positive wire goes into the center blob… does the negative go on the lower left hand blob near the edge of the board?
After
the negative lead joins the left leg of the chip opposite the positive lead. The positive lead joins the incoming resistor? Polarity sensor?
Edit: Deleted irrelevant pic
Red +, Blue -
Is this right? I had it this way, but the solder blob for the negative managed to cling onto 2 legs. I was wondering if that was the cause of the issues?
I don’t think that is whats wrong here. You have the right points ploted in the pic, but my driver does have the two legs of the chip soldered together with the ground wire.
Dang I was hoping that was the whole issue.
Well, back to the driver as a whole, instead of electrical tape, would a section of heat shrink tubing over the driver pcb and toroid prevent shorting inside the pill without causing any issues?
Sorry about my totally irrelevant picture, I had it in my head you were using a 105c board for some strange reason. The brain get’s old, the feet get cold, we walk straight into danger, ever so bold.
I wouldn’t think both legs on that mcu should be soldered, but it’s gonna take someone with more electronics saavy than me to figure it out. Sorry.
you see the two indents near the croc clips?
thats where I solder my negative on that board, either one will do, just pick the one that works best for you.
you were saying you made the pill press fit for your contact board, have you used a meter to confirm continuity between the pill and the ground ring of the contact board?
I have not checked continuity… Set DMM to Ohms, and… positive lead to spring, negative lead to pill? Would just the contact board inside the pill be able to verify continuity?
Edit: Reassembled and still getting the quick flash of light and then nothing (with the head and reflector off). There is a reading of .665 roughly when checking continuity between the contact board spring and the pill. Bigger heat shrink should be here in a few days for the driver.
You should be able to touch the ground ring on the driver with one probe, and a clean non-anodized surface in the head with the other and get beeped at by your DMM.
When I use an existing driver board for a contact board, I like to de-populate the old board. Take everything off. Then make sure with the continuity setting on the DMM that the pad where I’m going to solder the spring is hot through the board to the place I intend to solder the red wire and attach the new driver. Same goes for the ground, make sure the ground on the inside of the contact board is connected to the ground ring on the new driver. A short wire soldered between them will suffice. 2 short wires soldered between them is even better. This makes absolutely certain that when the contact board is locked into the lights head and grounded, that ground is carried through to the new board. Then the black/red wires from the new driver go to the led as usual.
Double check these points to be sure your new driver is grounded while floating inside the pill. If it’s able to move at all, your ground would be intermittent if just left floating in there. Solder the wires between the 2 boards ground rings and you’ll be certain it’s grounded.
I get a small reading from the ground ring to the threads (no other non anodized location in the head).
So the board with the toroid on it, up inside the pill needs to be immobile? I have the 2 semi-pliable wires connected from the contact board to the above locations on the bottom of the driver board marked in the pic. There is a ground ring on the driver pcb as well?
I’m using this full stock driver. Only I took just the spring off the original driver and placed it on the aftermarket’s contact board.
As long as the contact board is grounded and both wires are connected between the contact board and toroid board you should be fine. But with the driver in the pill you should get a loud beep from the DMM when touching the ground ring and the threads in the head. If there’s no beep, you’re probably not getting a solid ground.
My DMM almost instantly beeps with a 017 reading on the display. I have the stock contact board but am running direct drive. The components that came on the board are stripped. With a computer laptop pull it’s getting 3A. I’ve got better cells in better lights. 
Okay so I put together my little test rig; hooked everything up and touched positive wire from the driver spring to the positive terminal on the 2x 18650 in series.
Same thing as in the assembled pill/flashlight.
Flash of light and it’s gone.
So there’s something up with the driver. :weary:
Bottom of driver pcb, front side
Bottom of driver pcb, back side
