[Solved] qlite driver shorting?

I have a qlite driver that seems to be shorting the battery + to - when it’s in a light.

Outside a light, it seems to be fine.

Has anyone seen something like this?


EDIT: It turned out that the problem was that the insulation on the positive emitter lead got cut, so that was shorting to the star.

I wonder if the positive lead solder blob is contacting the flashlight body when in place? That’s what I would look for. The positive is always hot, the driver switches the negative lead on and off.

I just had a similar problem with a 3A Qlite yesterday, it always goes into direct drive mode, in or out of host, and with or without the third star grounded. Hank is kindly sending a new one with the next order. Is yours going to direct drive in the host and all modes outside the host? Has to be a short to ground somewhere, try kapton tape between the edge and the host where you suspect there might be…

Because the positive lead is so close to the ground some solder can actually cause the short.

Thanks for that. I had a look at it and noticed the positive lead seems to be bent over so far it may be touching the contact of the chip next to it, so I suppose that could be the cause?

Also on a side note, I noticed that these drivers have much better quality wire leads than the similar Nanjg 105c drivers from FT (I also had one of them dead out of the pack yesterday as well, not going to bother with trying to RMA it). From now on I plan to get all my 105c drivers from IOS.

The inner pad of the diode and the LED+ pad are on the same trace, just separated by the green masking. Scrape the green off and you'll see. The output from the diode to feed the MCU is the diode pad at the opposite end, closest to the ground ring.

Thanks for that, will save me some effort: guess no need to go to the trouble of moving the lead then.


Nice :slight_smile: do the other side to :wink:

Nothing on the battery side to look at, it's essentially a 4x7135 slave board for the goodies on the side shown. The capacitor is fed by the via in the trace that goes to MCU pin 8, the MCU's power input.

Star #1 connects to ground, but has no via through to the MCU. You can see MCU pin 4 is grounded on its own trace, same as star #1 is on the battery side, so star #1 does nothing because it goes nowhere. Star #2 goes to MCU pin 5, star #3 to MCU pin 3, star #4 to pin 2.

I've built a single-sided 8x7135 before, move 4 chips to the MCU side, place the capacitor between MCU pin 8 and the middle ground pin on the nearest 7135. The battery side ends up completely naked. :p


I may’ve not explained clearly.

When the driver is in the pill, the light is not going to high (DD) mode. It’s not lighting at all, and I smell burning odor.

It’s not that the emitter - is being grounded, but I think that the battery + is shorting to the battery - (all on the battery side).

The battery also got hot, and I pulled it out as quickly as I could and set it aside somewhere “safe” (away from the house), and it stayed hot for awhile.

Scared the heck out of me :(, and it was late last night, so I decided to put it aside for awhile, lest I do something even more stupid while in a rush.

Actually, now that I’ve had a day away from it, I’m not sure if it’s a qlite. It’s a blue PCB, and I thought that it was one I got from IOS awhile ago. Aren’t the Qlites from IOS blue PCBs?

If I have time tonight, I’ll try to do some review… carefully…

If the ID of the step the driver sits on is too wide, it can touch things it's not supposed to. Some C8 pills are really thick walled and have to be opened up for clearance. Compare a C8 pill to a P60 pill, the difference is huge and there's not a lot of wiggle room to start with.

When you say 'outside a light, it's fine', what does that mean? When the pill is not in a light, or when the driver is not in the pill, or what?

The picture that I posted is of a recent Qlite from IOS

What I meant was that I had tested with the following configurations, which worked ok:

1) Driver with emitter wires soldered to the emitter but with the driver not pushed into the pill. I had the + lead from a power supply connected to the spring using a clip, and the - lead from a power supply connected to the edge/ground ring of the driver using a clip - This worked.

2) Driver with emitter wires soldered to the emitter and with the driver not pushed into the pill. I had the + lead from a power supply connected to the spring using a clip, and the - lead from a power supply touching the pill - This worked.


So something is shorting the LED+ or something else connected to the same to the inside of the pill. BAT+ is always connected to LED+ at all times, any short anywhere on that part will give your symptoms.

Is this a secret project, some reason why you can't say what light/pill it is or show pictures of the driver compartment?

Another thing that might be a cause is if you use a bigger spring (inverted), when the spring is compressed the bigger coils can short out to the legs of the adjacent 7135s.


Sorry, no, it’s not a secret project… I was just being lazy :)…

I was building another one like this:

but using a triple with 3 Nichia 119s that I got from texaspro:

Actually, I just found the problem. It looks like the insulation of the emitter + lead got cut, so it was shorting to the star. I’ve fixed that, and the light is now working.


Thanks comfy. That was one of the things that I thought might’ve been the potential problem. In this case, it turned out it wasn’t, but you’re right that the spring coils get awfully close to the 7135 pins!!

Qlite are not blue because they are based on Nanjg 105C driver actually, I assume you have that driver from FT that is blue with 3 stars only with a blue and red leads which is not a Nanjg 105C layout.