Fenix TK75 Driver Mod Help

So after searching all over I have found almost no breakdown pictures of the Fenix TK75. I got a great deal on a used XM-L2 version and of course I had to take it apart. I am up in the air wether to change the LED’s but I want to increase the current. I am no driver genius like many of you guys here so I’ll post pictures of the driver. Any help would be greatly appreciated.

You don’t need to mess with the glued head to get to the driver, but of course I did! I fought with this for a while trying many different things but this is what finally worked. Place the head in a vice wrapped in some protective material, I used old jeans. Then take a heat gun and heat up the bezel evenly, I tried to keep as much heat as I could off the lens while doing this. Then take a rubber strap wrench and twist it off, you will need this!

Now for the driver. These are components I am not use to seeing. If anyone see a way to modify please let me know. I am wondering if Vinh uses the existing driver or replaces it with a new one. Thanks guys








I cant help with the modding of the driver, but i can tell you that Vinh uses the original driver. I had asked him to mod my XM-L version and he said he wouldn’t be able to push it as far due to different drivers between the XM-L and XM-L2.

Well that is good news! Thank you

Very nice. Here's the danger, it's a buck driver like the DRY so if you go too crazy you can possibly burn out the LEDs. Proceed with caution! You know from firsthand experience how that can go, wouldn't want you to ruin the TK75, it's not so easily replaced.

That big 6R8 is an inductor...yellow capacitors....then a bunch of feedback loops and resistors that must all serve a purpose but may be over my head.

You know what, I bet if you email Vinh and ask him he will just tell you.

He’s a great guy like that.

Nice layout on that driver from Fenix, seems they've been using it in other flashlight since that one is made on 2012.11.06

I figured it was a buck driver… as for burning the LED’s, I know a company that sells bare LED’s at great prices :wink:

I figured most of that stuff is for the modes and the fancy two button system it uses. Can that FET driver handle 3 MT-G2’s in parallel lol they fit in the reflector holes.

I have heard nothing but great things about Vinh and considered doing that but I want this info to be out there for the public. It would be wrong for me to ask and then post what he has worked hard on.

Sense resistors:

R050 = 0.05 ohms

Not knowing how far it can be pushed before it falls over and zaps the LEDs, I'd start safe by stacking something relatively high, like 1 ohm or 0.5 ohm. Only after you see how much it increases with each reduction will you get an idea of what it will take to get where you want it.

Awesome thank you!

I have a trimpot from RMM that i’ll throw on there and see what it does

+1 comfy. Those R050's are the resistors you need to work on. The TK61 has the same exact resistors: two R050's. I added an R100, tested it, and got a nice bump, then added an R082. So for the 2S2P cell setup, I got battery amp reading of 1.51A stock(3.02A effective), then after the 2 resistors added, got 2.38A. So this is 4.76A effective.

If you do the parallel resistor math:

R050 = 0.050 ohms

2 * 0.050 = 0.025

.100 + .082 + .050 + .050 = 0.016078

So the ratio of 0.0161 to 0.025 should result in roughly the bump I got in amps.

Thank you for the info!

This is the setup I used to test it. I directly soldered 18awg wire to my harbor freight red DMM and then soldered it inline after the driver. Then I soldered the driver to the battery pack.

The stock results I got with 4 x Efest Purple 25A 18650’s
0.05
0.36
1.01
2.84

Then i stacked two R50 resistors and my results were
0.05
0.38
1.07
3.00

You really need to be measuring the output current directly, not measuring the input and then using that to estimate the output.

edit: Wait... you DO have the meter on the output? So where is the '2.84*2' coming from? The output is the output, no multiplication by anything.

Those are basically “tailcap” readings, imo!

Yeah I wasn’t thinking when I wrote that. When I was comparing to tom’s readings I multiplied mine to but he measured his differently.

3 LEDs are wired in series, which would need 10-10.5 volts? And the battery config is 2S2P, or 8.4v? If so, that's not possible with a buck driver and those measurements make even less sense than before. :~

If I understand you correctly, you have reduced sense resistance from .025ohm to .0238ohm, so your drive current should theoretically be 2.98A after mod. Close enaugh, I think! BTW the driver is a boost driver boosting from 2s2p battery voltage to 3s emitter voltage.

Well, the measurements make sense, but not the multiplying by the number of cells thing. If the meter is inline with the output to the LEDs, and all the LEDs are wired in series, what the meter shows and nothing else is what's going to the LEDs.

Ok this is the new setup I did with the stock resistors

Low - 0.05A VM1=8.11 VM2=7.90
Med - 0.36A VM1=8.07 VM2=8.44
High - 1.02A VM1=8.00 VM2=9.08
Turbo - 2.86A VM1=7.87 VM2=10.21

Sorry to confuse things - the TK61 is a different setup (1 LED). Think your first wiring drawing was good, so you are measuring on the LED side, better than how I did it... Got no time right now, but you got a little bump for resistor values that would only give you a little bump, so, sounds good! Now to step it up .

Use this: http://www.1728.org/resistrs.htm for calcs.

Any progress on this? I got a TK75 I just started working on - see the same thing - the two R050's. Thinking of going for the R100 and R082 mods, or there-abouts. Only other thing to do is upgrade the emitters from XML to XM-L2 U2 1A's, and for that, need to crack open the bezel and it seems glued pretty good - need to spend some time and muscle on it. Did you open or try to open the bezel?