Kinda-sorta related question... TK75 has optional extensions for another 4 cells, so input voltage goes up to 12.6 (2S4P) 16.8 (4S2P). That means this has to be a buck/boost combo driver, right?
not too mention Fenix recommended not more than 3 extensions is used at 1 time, although Fenix engineers tested up to 9 extensions :bigsmile:
So that's what the '75' in TK75 means! :D
No - I tested back to back carriers - the voltage goes from 8.29v (one) to 8.35v (two), consistently. So, the carriers are in parallel, 2S4P. These are probably "smart" carriers, it looks like. I didn't take one apart yet, but they appear to have PCB boards on the ends, vers/date stamped.
Ah, so it does both + and - connections at the front of the carrier? And the contacts at the rear of the carrier are only used if there's another carrier behind it?
Anyone can give me and general idea how much in % i will gain with doing the resistor mods? what else can be done which doesn’t require many hours of work?
Yes - definitely.
Well, not sure how familiar you are with the relationship of amps to output. Doubling amps will not double output (lumens), it will do less, but typically will make a substantial difference. The triangular copper star in the Fenix lights is pretty impressive, but is begg'n for XM-L2's if your light is the older generation XML U2 one. I know there has been a couple of revs on the Fenix drivers, older one supposed to be weaker, but not sure when the transition was (corresponds to XML to XM-L2 transition or not).
Didn't do any measurements, but there's dribs and drabs of info on the TK75 around. Think's the LED's were driven at about 3A stock with the two R050's, so should safely be able to boost it to ~4.5A. I measured a stock TK75 last night at 3100 lumens (XML U2) and a resistor modded XM-L2 (domed) at about 4,400 lumens.
WOW, so you are saying, just with resistor mod only i can gain around 1000 lumens (asuming that domed adds + 400?)
I got the new version 2900lm. How much does the exchange parts cost? it it time consuming? i got a friend who is uber pro with this stuff.
Well... It's not really apples-to-apples. Not sure, but maybe the XM-L2 (XML U2 to XM-L2 U2) would bump 10-12% (same copper star before/after). Roughly on good copper, a 3A to 4.5A should bump output at maybe 25-35%?? Rough guess's but think it's in the ballpark, and would make the difference I'm seeing.
So let me get this straight. I got TK 72 XM-L2 version.
With everything stock, only resistors swapped, i could gain lets say at least 20% ? where do you guys get the swap parts?
Is this mod safe for the leds/batteries?
"Is It Safe?" one of my favorite movie quotes - Running Man I think... I resistor modded and tested a TK61 and feel confident it's safe. Of course it's higher amps, more amps being pulled from the cells, more heat, etc. You want to play, you gotta pay... Fenix had some great lights - 18 gauge wires in bothe the TK75 and TK61 -- biggest I've ever seen in a stock light, super efficient, max heat mgt for min weight, etc... If any light on the market can take higher amps, it's these Fenix lights. There are always risks going to higher amps on every light though. If you are that concerned about it and want 0% risk, don't do it. Of course there's been dozens of TK75's modded the same way and haven't heard of any failures (of course hundred's/thousands high amp mod lights out there). The more moderate the resistor mod is, the lower the risk, the lower the output.
Good info here guys. Can someone link me to a thread which discusses what happens when a buck driver is pushed too far?
TK75 is using a boost driver. 8.4v in, 10.whatever out.
But with buck drivers, some of them don't care and you can short straight across the resistors and it keeps right on going. The ones that don't (Solarforce S1100/2200, the new 32mm DRY driver, probably others) if you short or go too low on the total resistance it connects the full input voltage to the output, which in a buck setup can easily ZAP! the LEDs, depending on the voltages involved..
I'd stay with resistors resulting in 4.5 to 5A max on these - AND DO NOT BRIDGE the resistors. Think vinh is doing resistor additions on his Fenix mods, but he bridges the TN31's and K40's. Lessons learned I'm sure, and he does dozens of them.
Thanks Comfy. So for the affected buck drivers this appears as a cliff? The driver stays in regulation and behaves nicely as you increase current until suddenly it’s DD? I’d be satisfied with the “what” of the situation but I’d really love to read about the “how” of the situation. My understanding of a buck converter is very limited, so I may not be able to understand the “how”. Is the inductor getting saturated?
Thanks alot guys! my new (not here yet) TK 75 will be modded to at least 20% more output i hope 
I ordered some R015 and R010 resistors, which just came in. I didn’t have anything that low. I will post the results later…… Also if you bridge the sense resistor it disrupts the UI 
Those sound really low, much lower than I would use I'm think'n. Did you do the parallel resistor math? The TK61 I mod'ed had the same two R050's, and I added one R100 and one R082 and got about 4.5A out of it, I think.
Here's the info on the TK61:
Ok – stock amps at the batteries is 1.51A (2S, 2P cells), so 3.02A effective. Added a R082 and got 1.99A at the tail (3.98A effective), then added an R100 and got 2.38A (4.76A effective). It appears to be reliable so far – mode switching is working well, but need to do more testing.
Results:
lumens: 1,622 at start, 1,608 at 30 secs, 266 kcd throw.
So OTF lumens went up from 1,217 to 1,608, and throw went up from 195 kcd to 266 kcd.
R050 + R050 + R100 + R082 = 0.016078, Tom's ~4.5A
R050 + R050 + R015 = 0.009375, too low!
Even if you remove one of the original R050s you get:
R050 + R015 = 0.011538, still too low!
Removing both R050s and using a single R015 would be sorta close. But you have to wonder, why did they use two resistors in parallel when they could have just used a single R025 instead? Using more resistors to get to the same effective total resistance is more accurate, it spreads the work between them and they run cooler, and heat affects the resistance.
If you are even considering messing with sense resistors you MUST BOOKMARK THESE TWO PAGES: