Luminus CFT-90 Testing - The Mother of all LEDs

I think this LED would make much more sense with your you recoil light than Synios.
You say that collar gives double lux and half lumens. You also say that your recoil light is meant to have very high optical efficiency.
But then you pick Synios which gives you 30% more lux for 95% less lumens (compared to CFT-90)….

Efficiency is how much is produced vs wasted.
I calculated slightly more than 85% for this recoil design.
It will remain 85% regardless of what LED I use because efficiency is defined as the output over the produced.

This is not the same as “what’s the most lumens I can get”

Sure. But when chasing records I view efficiency as nothing but one of tools used to get the desired effect.

i hope your right sir, or even close
ill be darn tickled if it lands in between your lux numbers! :wink:
2.65 Mcd, is where the gt beast should be happy at
and 4000 lumens in the spot.
oh my…. :open_mouth:

Well, gotta remember that it will heat up real fast so those numbers will drop quickly if it’s in a flashlight like the GT.
Active cooling and good CPU heatsink would be necessary to maintain the output.

Without knowing all the formulas involved I still I do not see that happening.

V54 got ~ 920Kcd w/ the TN42vn90. The GT getting 2.65Mcd??

TN42vn90 is underdriven. It got 900 kcd with 3800 lumens OTF. Scale that up to the full CFT-90 potential and you get like 5300 lumens OTF and 1.25 Mcd at cold start, with pessimistic calculation of losses. Vinh is looking after a buck driver, so he can power CFT-90 better.

I checked mouser and there are a lot of DC-DC converters that will take ~12v and bring it down to 3-4 for the LED.
They are all constant voltage though, not cc.
I don’t think this is an issue though, as long as the LED is cooled properly you can set it to constant voltage and the current won’t change.
They also have protection for overcurrent, so all you need to do is buy one that does 40A or 50A or whatever you want and it won’t exceed that value.

2.65mcd is possible
if enderman thinks its remotely possible, then it is
the gt is twice the light the 42 is
its at 1.3 mcd out of the box, under driven, without perfect tuning, and reflector is twice the surface area
should be very possible.
enderman knows more about this stuff in his sleep, than most here will know awake, in their lifetime.
i wouldnt question his judgement.

I calculated 2.3 not 2.65 btw :slight_smile:

ill take it.
when do you want me to send you my 2- gt’s?
and how quick will you have them back ? :open_mouth: lol ! :money_mouth_face:

:stuck_out_tongue: lol
That’s more of a job for someone like vinh who is a lot more experienced with flashlight modding.
I’m sure once he gets that 20A+ driver he’s working on finished he would be glad to mod the GT for you.

I don’t really do flashlight mods, more scratch builds :wink:

I do have plans to mod my R40 some day, I think I can get over 500kcd out of that tiny head!
Not enough time or money atm though.

Some nice beamshots from Wolfdog1226 can be found here.

Acebeam K60 with XHP70 in DD:

TN42 with CFT-90 at 18A:

Notice how the spill of the TN42 is not darker than the spill of the K60.

I’m dreaming about putting CFT-90 together with Texas Avenger driver in my MF-02.
Would that driver be OK choice for this LED? Buck would be better, but I reckon that with as low Vf as CFT-90 has it should be possible to get 90-95% of output (with fresh batteries, then dropping). Considering that TA is available and powerful-enough-buck, even of it existed, would cost significantly more it looks fair to me…
I remember seeing problems with 7135s blowing in high-current XHP70.2 builds, would that be a problem here too?
Is TA resistance low enough to enable 30-35A currents?

I would be willing to ship one of mine for this mod, Enderman.
If you are willing as well. :+1:

Haha thanks but no, I don’t do mods for people, mostly because of time restrictions.
I’m way too busy in university.
There are people like Vinh or others here on BLF that would be willing to do it for you and would probably do a much better job than me at it :slight_smile:
Taking the light apart without damaging it is also very hard, I would only take that risk with my own stuff, not other people’s.

Did anybody say, let’s built a new GT with this new LED? I’d be willing to start the research on doing something big with this LED. By research, I mean, donating money, so the guys with the talent and brains can do the research, and design for us. Is anybody else interested in doing a project?

If is true what guys say, about needing active cooling, I think we should look into that. How about building a 140mm reflector with this LED?

Vinh is modding several GTs with this LED.
They will likely be between $500-1000 each.

I don’t like working on other peoples things myself. I figured your response on the lines of what you mention, based on some of your comments on other threads.
Matter of fact, I noted your response prior to mine.

Maybe Vinh will come up with a nice recipe for the GT. If he does, I may get interested enough to follow through by sending him one of mine.

That last picture made me chubby. Any chance This won’t require me getting a second job to pay for it?