Dead TK 75-HELP PLEASE!

shimey-both the positive, and negative, terminals are raised, and on the same end of the battery carrier. I don't see how turning around the carrier and screwing on a solid aluminum tailcap wouldn't short out the carrier. Tom, I don’t know if the LED’s are ok. I haven’t tried powering them from an external source yet. I believe the carrier is 2S-2P as it shows 8.37v with un-tripped cells, then after inserting the carrier and removing it shows 4.18v. Think Comfy is on to something here with the dead short between B+ and ground. Now how to fix this?

I also believe probably there is some short, and that explain why safety is triggered in batteries. one of the things I will not do in this point is to put unprotected batteries.
u have to find where the short is. Am not familiar how tk75 is build internally but first thing i have to check is battery compartment, if he can be pulled out do the short again happen between + and - ?
if yes problem is there, if no than continue to check the path the + goes till the driver and to the led. somewhere in this path must be a short circuit which tigers safety mechanism on batteries.
from what you say i think to things can happen when you put on bag. one is like you say the flash turn on but it heated up very much being modified and lacking fresh air to cold so led can go dead but i thing is very improbable that ALL three leds go dead in the same time, so probability driver have gone dead heating to much and probably is shorting something. or some wire unsolder-ed from heat and also make short somewhere.
or if u drooped it hard somewhere something in battery section have moved and is shorting from plus to ground, but this is probably less likely to happen.

two batteries at 3.94v seems to have worked some sec at high before something go puff and short inside…

If the short were on the output side (reflector touching MCPCB pads, or wire got hot and unsoldered itself, etc.), it wouldn't trip the protection until the light was switched on. It's not doing that, it trips as soon as the carrier is inserted & tail screwed on. It has to be something on the input side of the driver and not the output.

I've never been into one of these lights, anybody have good clear pictures of the driver?

Comfy, I don’t own a good camera, and the pictures I took with my phone are not good enough to post. Anyone else have pics?

Cells are 2s2p. 3 leds in series.


Thanks for the great pictures, Racer. Update- I desoldered the battery contact board and checked for a short there, and everything is fine on it. So, the short is certainly in the driver itself.

Check for shorts in the caps, the yellow blocks on the second photo and one on the first photo.

Thanks Ryan, how would I do that? Isn’t there always continuity across a capacitor?

Yup, you’d check for continuity with your DMM.

Capacitors will be shorted at the very instant you put your leads across it, but the resistance will rise to infinity as it charges and doesn’t pass electrons through.

Yes! Tailcap IS anodized, but I (forgot to mention) I cut a thin plastic disc that press fits in the tailcap for that very reason.

Ok Ryan, The cap, in the second picture, to the left of the “big square thing” is showing continuity, all of the time. The other caps are acting like you predicted they would.

Great, we have a lead. The cap is directly across the battery inputs, and the shorted cap may be tripping the battery protection or something else along the two paths might be the culprit. If you can desolder the cap on the left, that will verify that it’s the issue.

Edit: Upon closer inspection, does it show any signs of overheating or damage?

The cap does not show any signs of damage. I de-soldered it and now it acts like it should, and does not have continuity across it. The solder pads on either side are however still showing continuity.

I see, then it seems like something else is shorted out when it’s not supposed to be. How many Ohms is the short reading, 0.x?

Getting the same meter reading across the pads as when I touch my leads together

Hmm yeah it’s start to trace back I think. Unfortunately I don’t have a TK75 so I don’t think I can be of any more help, sorry!

Ok thanks for the help so far. I appreciate it. Hopefully Comfy will chime in after he reads these posts.

Pics rotated & more or less aligned, makes tracking down a spot from one side to where it goes on the other side.

(click for full size 1364x678)

I offer this without any suggestion that it is something you should try, only that it's something I have used many times when tracking down a dead short in an unknown location, when it's not possible to find it in other ways. It's a last resort kind of thing, but, most likely it will only break something that's already broken.

I was a professional auto tech for over 20 years and the only thing more difficult to find than a dead short where the fuse blows instantly every time is when the fuse only blows seemingly at random once a month. A true all-the-time dead short can be found rather quickly, but you have to have some balls to do it, especially when it's an expensive thing that doesn't belong to you. But, it's already broken, right? Whatever it is causing the short already is in need of repair, but you can't repair it if you can't find out where it is.

This is one of an assortment of 'not-fuses' I built over the years.

(do you see where this is going...? :) )

I never burned up anything that wasn't already damaged. The action starts closest to where the original short is, and if you keep the connection going in short bursts, only just long enough to spot where the first wisps of smoke come from, it's fine. You can then go to where the trouble is, and fix it. Fixing it is the goal. What happens along the way isn't all that important.

Good news is, since this is a dead short on the input side of the driver, you don't need to use the full original 8.4 volts. A single unprotected cell will do just fine. Connect it in short bursts, around 1/4 second (or less) every few seconds, until the problem spot makes itself clear. DO NOT simply attach a battery and let it sit there while you go fix a sandwich. If you can, mount or clamp the driver in such a way that both sides are visible, and have a helper watch one side while you watch the other. Wear safety glasses or use a shield - if you do it for short bursts nothing should explode, but there are no guarantees.

In no way should this affect any of the proprietary non-replaceable parts (like the MCU with the Fenix-specific firmware on it), all the stuff on the input side of the circuit should be fairly common commodity components that can be replaced. But you first need to find out which component needs replacing, if you're careful it should only damage the part that's already failed.

^ lol this is what i can call a very hard way. i must accept it should work, but its little bit to brutal to my tastes :bigsmile:

i think if we can arrange a live conference will be easy to find where the problems is. some experienced members that know their stuff and you with the torch in other side with a web camera and a multimeter and the willing to make a few measurements maybe can work :stuck_out_tongue:

edit : i forgot, i think it can be funn too, to mess around :stuck_out_tongue:

A lot of these components cannot be tested while they're still soldered onto the PCB, the other stuff connected as well makes it impossible. The dead component causing the short is already dead and already needs to be replaced, but you cannot replace it if you don't know which one it is.

If power is applied in short bursts just long enough to get a tiny amount of smoke from the problem component, what's the harm really? Is it really more extreme than removing components from the board so they can be tested on the bench? Plus, some of these parts aren't going to come off without a hot air gun. I'm going to guess he doesn't already have the tools to diagnose it 'properly' or else he would have already done so.

If it were mine, I would give it power just long enough to make sure nothing is likely to explode (tantalum caps can be pretty violent), and if it seemed good, I would cover as much of the driver as possible with my hand and see if I could find the general area that gets hot first, and then focus on that area until I had the exact part isolated. Then I'd remove that part and see if the dead short was gone. If so, find a replacement for whatever the part is and have a working flashlight again.