I am looking for a light I can take to shoot video in underwater caves. My max depth will be around 100 feet. I am looking for a light that will be somewhere near 100,000 lumens, have a spread of around 120-180 degrees and will burn for at least 2 hours. Heat dissipation is not too important, as the light will be submerged in 68-72 degree water. I am handy with electronics, so I can even make my own light if you can suggest components. I want the color temperature to be warmer so maybe around 2700-3000 K?
If a cheap enough pre-existing flashlight can be found, then it doesn’t even matter to me if it is not that waterproof and breaks after 10 uses as I can just buy a bunch of them.
The brightest light I have found so far that looks like it might work for me is the Imalent DX80 which is 32,000 Lumens but I am not clear how long it will burn at that brightness and it is only waterproof to 2 meters and pretty expensive.
The really expensive option would be the BigBlue VL33000P which is 33,000 lumens, waterproof to 100 meters and burns for 2 hours at that brightness. But it costs $1500. Can I find a cheaper, brighter light than this?
Well, that is strictly not possible with current technology.
You would need to dissapate 1000W of heat, and a 2kWh battery, and that would not be possible to carry by anybody except if you had a huge backpack, and would heavy.
It would also cost a ton of money just in batteries. Even at 200$/kWh, it would cost 400$ in batteries alone, and you would need to connected them together, which would cost even more.
So, 30 000 lumens though, the only way would be to put together a light with 12x XHP70.2s 4000k and under drive them. You would just need a big metal plate to hook them up, and a way to waterproof them. Or you could boost the output even more, but that would be overkill.
I suspect dissipating even 1000W of heat wouldn’t be a problem for a dive light that’s strictly being used underwater. The water should have no problem removing the heat.
A bigger problem might be finding a light with a casing strong enough to withstand 100’ depth, and big enough to hold enough LEDs and cells to power that many lumens for that long. Might be better to pick a large dive light so you know how much space you have to work with, then see what can be squeezed in.
Even then I suspect 100,000 lumens for 2 hours isn’t possible with today’s battery technology. As the previous poster mentioned, you would probably need a backpack to hold all the batteries.
Alright well, the Big Blue VL33000P does 33,000 Lumens for 2 hours down to 100 meters and is about the size of a soup can. I was just hoping there was a less expensive / brighter light that you folks might know about.
Or just get a G700 “tactical light” offa Amazon. If you believe the spex, that’s gotta be at least 100,000lm and waterproof, and runs off a trio of AAA alkaleaks. <snort>
Don’t be rude guys. The man asks a simple enough question.
I don’t think we have a big diving population around here, so you might not get much info on existing lights.
As you said, heat shouldn’t be an issue under water, and the components are available to build your own, the waterproofing / host would really be the tough part.
The brighter you get, the more your run into diminishing returns. Apparent brightness is on a logarithmic scale, so even though 100k is 3x brighter and would require 3x the power, it would only look marginally brighter.
Bluesword’s reply in post #2 tells you what you need as far as lighting components, but I can’t be much help in terms of the waterproof housing. It won’t be cheap, but it will surely be way less than $1500. Will take quite a bit of work though.
Just a thought, it might be cumbersome, but how about putting a light bar into a polycarbonate pipe?
Of course heat dissipation would be the challenge, would need solid bars leading to external heat sinks, and an external battery pack. Only good for setting up in a static position as well, although strapped to a scuba tank if you are moving forward in a cave, the ceiling bounce would be diffuse and interesting at very high lumens.
If it helps, the specs on that Big Blue VL33000P don’t add up.
Ignore the red XPE LEDs and focus on the 45x XML LEDs. Driven at 3A/10W with proper cooling, these are capable of ~1000lm each, so, 45 of them could deliver 33,000lm, but they’d need ~330W to do so.
Thus, it would take 330Wh to sustain that output for 1h, and 660Wh to sustain it for 2h.
And, so we are clear, as we can see above, their claim isn’t that it can output 33,000lm OR run for two hours. They are saying that it can do both, at the same time.
If you guessed that this would be made up of 8x 26650 cells, it seems you are right. It looks like it contains 8 cells, and the specified weight looks about right for 8x26650.
A high capactity 26650 holds ~ about 20Wh. 8 of them then would hold ~160Wh. In a listing for the pack on BHPhoto, it is described as having “29.6V of output power, 5000mAh capacity, and 148Wh energy consumption rate.” That doersn’t exactly make sense, as stated, but 8s LiIon batteries would have a nominal voltage of 29.6v. 5000mAh is a reasonable number for 26650 cells, and such a pack would carry a nominal 148Wh.
Ah, but remember, we estimated it takes more than twice that much energy to push 33,000 lumens for one hour, and they are claiming this light can do it for two.
So, as I said, their claims don’t make sense. As we’ve seen, their claims are out of wack by about 4x.
Oh, wait. They, are probably using the FL1 standard to measure runtime. No wonder their claims don’t make sense. FL1 describes the runtime from the light’s output 30s after being switched on until it drops to 10% of that level. So, what’s actually being claimed is that, on its highest mode, the Big Blue VL33000P will put out 33,000lm with a charged battery, 30s after being turned on, and that 2h later, the output will have dropped to 3,300lm.
I haven’t done that math, so that actually sounds plausible, if one assumes a gradual decline. It isn’t, however what a lot of people think output and runtime specs mean. Including me, I didn’t think about it until I was considering legitimate reasons the claims were so far out of wack.
It would seem to me though that a light intended for shooting video should sustain a nice steady output over the course of its standard runtime, not a gradual decline to 1/10th the original level. Thats about 3.33 stops, the difference between a nice clean ISO100 and a noisy ISO1000.
Building a light that is basically 3x a Big Blue VL33000P is a bit more plausible than it would be using the assumption that 100,000lm would have to be sustained over its 2hr runtime. It’d still be a challenge. If I were doing it, I’d probably shoot for a much lower tight, or at least narrowly, regulated output over the targeted runtime.
First, lets use a bigger battery pack. The BigBlue pack looks like it weighs about 1Kg. If use 18650s, 1kg works out to be about 20 cells. If we use 3,400mAh cells, that’ll get us 20x3.7v*3,400mAh = 250Wh. If we bump it up a little bit, to 24 such cells, we have 300Wh, about 2x what the BigBlue light has. Suitable cells probably run ~$7-8 each. I’m not sure what they would cost in pack form, but, for now, let’s estimate $240.
Ok, so if we have 300Wh, and our target is a regulated runtime of 2h, that gives us 150W to work with. That isn’t bad. With great cooling, that could be ~15,000 lumens, though we’ll probably have to give some of that up for warm color temps and good CRI. Even if we end up giving up 33%, that’s still 10,000 lumen.
To make cooling easier, let’s spread that over, say, I dunno, 24-36 3v emitters, each outputting ~350-400lm. I’m not sure what emitters to use, but lets spitball at $10/emitter including whatever MCPCB is necessary for an array of them. That’s $240-360.
To keep currents and resulting resistive losses down, let’s run them as parallel chains of 3-4 emitters (or more) in series.
I really don’t know what driver(s) to use, either, but whatever it is needs to be regulated, at least in the lower 2/3rds of its range, and adjustable. It should be possible to run at up to, say 600W. I really have no idea what to use, but it seems like $100-200 should be enough.
That leaves the housing. I have no idea what that might cost.
Also remember that emitters tend to be more efficient the less power you run them at. Having 100 emitters run at 1.5 amps should be more efficient than running 50 emitters at 3 amps.
