Dive light build recommendations

I’ve been thinking about making a dive light for a few years now but always had to much on the go, now my work has slowed down and the major reno’s to the house are done so I figure I will take the next several months to configure, design (and possibly sell if all goes well) some underwater ONLY compact 1X 26650 wide angle video lights for small wide angle cameras like the Hero cam. Unfortunately I have ran into a couple issues….

The light would utilize a TIR optic to get a 100 degree beam that has soft edges, this cannot have an abrupt cut off line as this light is for video. Problem is I cant find a 100 degree TIR, hell most of the time they don’t even give the angle the lens produces. I know an 80 degree exists but cannot find it either.

One XM-L2 U2 @ 5000K Putting out 1000 ANSI lumens sustained for an entire hour… Now I was told by a guy who knows his lights that 1000 ANSI lumens for an hour would be slightly optimistic with only 1X 26650 but I really really want to keep that number with out cheating… Any suggestions on how this can be done?

As for the driver, I would like it regulated, reverse polarity protection running at 3 amp? maybe a buck boost driver
2 modes, Always start up at 100% next mode at 50% or 60% or 65?? I hear there is not much difference between 100 and 75% and the run time would almost be the same. Any idea on which driver I could cram into this small light? What would the best % be for the second mode?

The switch would be a magnetic momentary switch (with stainless steel spring) located in the tail cap. Not many dive lights have this option but I don’t like the idea of a twisty on a dive light.

There you have it, any ideas or suggestions will be greatly appreciated :davie:

100 degree is very large, if you need a floody 100 degree without rings or hotspot for even video lighting, I would give a try to an emitter installed behind the lense without any reflector at all. An XM-L if memory serves right should be emitting light at 120 degree.

Well at least you should give it a try, sure it's an inexpensive test!

Can't help for the driver, mainly because you want to use a momentary magnetic switch and I have no idea which driver can be made compatible with that.

Ther project sounds interesting.

Rockspider, that is a good point. And that would certainly give a beam profile with a gradual fade around the edges.

Thanx for the replies guys,
Using a Hero2 camera a person will see the cut off beam from an 80 degree lens which is why I want to use a 100 degree… unless you switch the POV which for diving should be on wide. Looks weird above water but looks great once you get in. The thing that worries me about a “mule” is a lot of the light might be wasted by to wide of a beam… Unless I recess the LED into the light a bit less then I need… Interesting, will have to think about this one. On paper the TIR beam would be brighter but I wonder if it would be noticeable to the human eye…

Yes but the TIR works reflecting the light in a similar manner to a metal reflector, e.g. it will project a hotspot with corona and spill. The videocamera will "see" a bright white hotspot in the middle. You want something that has much more even distribution. Don't think you'll be happy with the TIR, no matter the angle.

Another option could be a very OP reflector, very orange peel, to smooth the transition between hotspot and spill.

I would still try the emitter alone...

Ever seen a UF-H2 or H3?

Definitely don’t want a reflector and I can’t find any info on super wide angle TIR’s, asphirics have bad cutoff’s so that leaves me with a mule.
I just looked up the UF’s, they have that cut off that I want to stay away from.
On the plus side making it a mule means it could be even smaller then I thought :bigsmile:

About the XM-L2 from the Cree website….

“Delivering breakthrough lumen output and efficacy in the XM package, the XLamp XM-L2 LED is the highest-performing, commercially available, single-die LED. Built on the SC³ Technology Platform, the XLamp XM-L2 LED delivers up to 20% more lumens and lumens-per-watt and double the lumens-per-dollar of the original XM-L. The XM-L2 LED offers the unique combination of high efficacy and high lumen output at high drive currents, delivering an unprecedented 1198 lumens at 116 lumens-per-watt efficacy at 3 A, 25°C.

Now there are many flashlights that say they can run a XM-L T6 regulated at 800 Lumens for an hour and a half so does this statement above not say that one should get up to 20% more lumens in the same time frame? This means I should be able to get 1000 Lumens for an hour and half no?

From a single 26650? King Kongs are 4000mAh, at 3A draw the max theoretical duration is 80', but the cell discharging graph must be checked carefully. One hour at max output is a more realistic figure, the remaining time will probably show a low voltage, which will trip the driver on non regulated area, and risk of overdischarging.

Hmmm interesting. So the run times for these other flashlights using the XM-L T6 are false?