The battery carrier mod that Vinh does is called a “Wick Mod”. It is a little copper wire inside and soldered from the bottom to the top of each spring. Supposedly helps with lights that demand a lot of current.
My TK61vn, TK75vn KT and K40vn all have the “Wick Mods”.
Can you share a pic of your tk61vn driver?
+1
I asked Fenix if its possible to buy a replacement driver for the TK61..
"
:Sp
I'm gonna try 5.3A with two R075 resistors (5.3A to LED according to the math). Pulled the MCPCB, nice heavy copper (not al), and will replace the emitter with a XM-L2 U2 1A, and de-dome it.
K- thanx! We've done plenty of those - I go back and forth between copper braid and wires. On these regulated drivers, not sure exactly how it benefits - may a little, maybe get longer life from the cells?
It’s possible Tom’s LX1330B is reading a little high, who knows for sure. I do know in the past his meter read very close to my LX-1330B meter.
Also whats interesting to note though is that my much more expensive Extech EA31 luxmeter tends to read slightly HIGHER than my LX1330B. Just about 1% though, so they read very close and consistent. And also note that Self Built also uses an Extech like mine. So who knows.
Irregardless, my way of thinking is that the most important thing with any of us guys taking measurements, regardless of me, Tom, Self-Built, etc. - is to be consistent. And to MAINLY use these measurements for comparisons for one light we have tested vs another.
I feel VERY confident saying that if we get a reading of 200k on one light and 240k on the next light - that second light is truly 20% stronger in throw than the first. So if a reader already has the first light and is considering buying the 2nd light, they can be assured if they do, it will give them appx 20% more throw than the light they already own. Obviously you have LED and general manufacturing variances, but I think the point is valid.
Basically nobody on any of these forums own a true integrated sphere or professional lux meters (as far as I know), so all of our testing numbers are going to be slightly off here or there. Regardless of it being Selfbuilt, Tom E, and myself included. We all do the best we can with the best equipment we can make or afford to buy.
With all this said, I am definitely having some doubts. Since I only have one meter, I'm thinking I may have intermittent issues with this one now -- just not sure. Ideally, I should have a light put aside that outputs consistent and re-test it with every measurement, and also have a backup meter. I measured a TM26 rated at 3800 lumens to be 3,100 lumens in my lightbox the other day as another example.
Also, the 2nd TK61 I measured (another unit) tested to be only 177 kcd - much closer to selfbuilt's #.
I'll continue to measure and post, and hopefully I'll find time to check calibration #'s, batteries in the meter, etc. - get a bit more confidence in it. I'd still rather publish #'s than not - pics just can't do justice if there's no long distance measurements included, and/or enough lights to compare with.
Anyone got dimensions of the enormous reflector…?
Thought I had pics of the TN31 and TK61 reflectors side by side - TK61 is 96mm head diam, and with their lightweight, plastic reflector design, the reflector is a high percentage of that diameter. Biggest reflector out there I believe.
Tom E, so vinhn said that 4A is the limit? thats very low! do you notice any problems or heat with your mod? you said it went up to 4.7A after you modded it, right? i’m afraid to let my friend mod it now when i read that 4A is the limit without modding the driver for heat?
I was hoping someone knows the depth of the reflector, so I can get an idea of the width / depth ratio.
Don't think I ever said that bout 4A ?? Someone else did maybe?? Sorry, losing track here... Again, we get tidbits, breadcrumbs... Didn't get to do runtime testing on the TK61 I have -- I have it 2.75A tailcap using two R075's, 5.5A effective, but did not actually measure amps to the LED.
If that's true about what he said about the 4A, it may mean he or someone looked up the specs on the parts and saw a rating of 4A max, in which case it could be true, and anything higher is at risk. But of course we all know CREE's specs call for 3A max 
, but with FET's, it may really be a risk. The knucklhead project may have the same issue - using a FET rated at 4A max. FET's are blowing but may be related to amps, may not, don't think they know yet.
Bascially any resistor mod is risky -- if I knew enough to reverse engineer a driver/circuit and compute the max amps it's capable of, I'd know enough to design my own 
.
So how exactly did you boost your light? only resistors which gave you like 400 lumen + you said went to 4.7A effective afaik, now you got it to 5.5A?
Vihn told me my TK75vn was pushed to 4.5A
Yes i know, but is that with resistors only? or he modded the driver?
What about the TK 61? do you have any info about max amp without modding the driver? thanks.
I didn't mod the driver. Sorry, no time to test much yet... All I did was de-dome and resistor mod.
What did rdrfronty measure in lumens? Though it was like 1,700? If so, no way, no how is it as 4.5A -- more like 6A+, because the TN31 was at 6.5A supposedly, and also around 1,700 lumens. I can understand a little bump in lumens for a bigger reflector, but not a lot. Sorry - no time right now....
Thank you! actually i only care about some general safe limit, i know you said its a risky thing to do and i should understand that, and i do….but i guess 4.5A should be fine for short runs on turbo ~ 1min? i never use it more then few mins on turbo, any light i got, only when i was testing my TK 75 runtime.
If my friend go to 4.5A from 3.1 stock, should i get like 30% + output generally?
OOPS!!! I forgot this is the TK61 thread!! I also have the TK61vn . Factory is 3.1A Vihn pushes it to 5.75A!
Ohh, ok. The TK61 reflector is about 81mm I.D., while a TN31 is about 66 mm (just measured both).
Crap - ok, so I got a tailcap of 2.75A, theoretical 5.5A to the LED and measuring only ~480 kcd at 5 m. I'd expect more if vinh is doing 5.75A - dunno if calculated from measuring at the batteries or measured at the LED though, but could be getting some loss's somewhere, maybe focusing, maybe loss of power to the LED. Thinking vinh measured at the LED -- I know he's done that before.
If I knew what resistor values he used, I'd know more perhaps. But of course if he swapped out FET's, that may effect the resistors compared to the output to the LED.
Converting tailcap amps to emitter amps in combination with a buck driver is like doing lux measurements by holding a light meter in one hand, a thrower in the other, stretching out your hands to roughly 1 meter between them and take a measurement.
Its not accurate, its not "effectively" or "theoretically" the same as doing it precisely from 10meters.
You dont know the efficiency of the buck driver so there is no way to know if its accurate. Doubling tailcap current in two cell series lights just happens to often be quite accurate on many typical budget lights driven to 2-3 amps and with freshly charged cells. But its not always the case. With a very efficient driver (like LD-29) emitter amps could even be higher.
Lets look at some examples.
Example 1:
Emitter current 3A on Noctigon. Vf is 3,4V (Vf number by djozz)
Energy at emitter =10,2W
Tailcap current 1,5A x2 cells at 4V (with load)= 12W Energy input to driver
Efficiency is 85%.
Example 2:
Efficiency is 85%
Emitter at 6A and emitter Vf at 3,9 (VF number by djozz)
Energy at emitter= 23,9W (notice that required energy at 6A is more than double of 3A due to the higher Vf)
23,9W/0,85= 28,11W input is needed if driver efficiency is 85%.
3,6A tailcap curent * 3,9045V (with load) * 2 batteries=28,11W input.
(I assumed voltage sag under load. Numbers above may not be 100% accurate, but should give a good picture.)
-1,5A tailcap current from two cells equals about 3A at the emitter with 85% efficient driver
(Simplified, saying 1,5A *2 cells= 3A emitter current is spot on in my example)
-3,6A tailcap current from two cells equals about 6A at the emitter with 85% efficient driver
(Simplified, saying 3,6A *2 cells= 7,2A emitter current is way off)
I never assume emitter amps with buck drivers, I doubt vinh does either. Especially not when he specifies emitter amps.
Always measure at the emitter at this level instead of using simplified assumptions. Efficiency of buck drivers often range from around 70-95%. And even if you assume 85% or 90% or whatever efficiency, its not as simple as to just double tailcap amps to get emitter amps. It only works in certain situations.
Okay