Need Advise: Diving Lamp For Video Recording?

remuen

Enlightened
Joined
Nov 26, 2001
Messages
600
Location
a place surrounded by Europe
One of my fellows has built his own diving lamp with 19 Nichia LEDs driven with about 20mA and he uses this diving lamp for underwater video recording.

Btw, this diving lamp is a beautyful piece of work as he is a graduate mechanical engineer working in a institute of technology where he has every possibility to make all these mechanical parts. Unfortunatly he lives not really in my neighborhood ....

He now wants to make a brighter diving lamp and he asked me for help for the electical part. I have two possible solutions in mind:
- using about 30 white Nichias (about the max. number because of the amount of space available)
- or using two or three white LS w/O

My fellow told my that every color temperature above 4500k will be ok so this is no problem with both LED types.

He only has 4 AA NiMh cells in this diving lamp and wants to have a runtime of 90 minutes so this would be possible with both solutions. I also would be able to make a DC-DC converter for him. Heat dissipation also would be no problem as we could use the metal housing for this (which is in the cold water).

My main concern is the distribution angle with the LS as it is smaller than the one of the Nichias.

My questions are:
- Has someone already made some experience with the LS in a diving lamp for video recording?
- Which of these solution is better and why?
- What else should be considered?

Thank you for every tip and advise!
 

Hemingray

Enlightened
Joined
Jul 2, 2002
Messages
380
Location
New Hampshire
Tek-Tite makes a 60 LED dive light, also a 40 LED
unit is in the works. Not inexpensive, around $300 for it. From what I have read in Craig's LED Museum review on this light, it is 6 cells, in a series-parallel combination,and uses a couple of silicon diodes to cut the 4.5 Volts down to 3.9V or so, thus not beating the LEDs to death so quickly. So, I suppose that one could be made up,
with the 30, 40 or how many LEDs in a parallel arrangement and use resistors or diodes as needed to keep the overdriving down to a safe and sane level. I would imagine that designing a decent waterproof housing will be the major challenge here. Each LED would not need its own resistor, groups of 3 to 6 can be paralelled and a suitable
resistor can be used (figure 20 to 30 mA per LED and 1V drop for 4.5 Volts or 2.5 V drop for 6 Volts DC power. The less power wasted in heating up resistors, the better.

/ed brown in NH
cool.gif
 

remuen

Enlightened
Joined
Nov 26, 2001
Messages
600
Location
a place surrounded by Europe
Just to point it out better:
The mechanical side of this diving light is no problem as my friend is some kind of an expert and his prototype already is tested. IMHO for me it's also not a problem on how to connect Nichias or LS to the the batteries
grin.gif
or making an boost converter for this light.
As I'm not an expert for underwater lightening (water is too wet
grin.gif
) my concerns are: As far as I remember I've never seen an LS in a diving light. Is there a certain reason for this? Is the distribution angle of a LS w/O too low for such a diving light for video recording? Has anyone made some experiences with an LS in a diving light or has even compared Nichias and LS in such an application?
 

The_LED_Museum

*Retired*
Joined
Aug 12, 2000
Messages
19,414
Location
Federal Way WA. USA
The Light Cannon 100 HID comes with a pair of textured transluscent diffusers that can be inserted inside the bezel to change its hot, narrow beam into a medium flood and wide flood, depending on whether you use one or both diffusers. These diffusers were specifically designed to allow the LC to be used as a submarine video light.
 

remuen

Enlightened
Joined
Nov 26, 2001
Messages
600
Location
a place surrounded by Europe
The Light Cannon is definitly not the light my friend can use. The main drawback are the batteries as he wants to use AA cells which are available almost everywhere in the world and are easy to take with you.

But the diffusers could be of interest. Has anybody an idea where one can buy such things or what else could be used to diffuse the light of an LS without having a big loss of light?

Btw, I just thought about this: Could one file the collimator down to get a wider angle of light distribution as one can do with the 5mm Nichias?
 

Hemingray

Enlightened
Joined
Jul 2, 2002
Messages
380
Location
New Hampshire
The best AA NiMH cells are about 1800 mA/H, so if you need a runtime of 90 minutes with 4 in series, the maximum total current draw can be 1200
mA (1. amps), probably should aim for no more than 1 Amp, just to have a little safety factor.
So, that would be 50 LEDs at 20 mA each, proportionally less if they are driven harder.
That would allow 3 Luxeons at 350 mA each, or 4
if they are lowered to 250 mA each, 6 @ 165 mA each, etc...

There is no physical or electrical reason that Luxeons would not work, the water might aid in drawing off excess heat. If you use the Star/Os
(with Optics) the beam will be fairly narrow and sharp, if you use "naked" ones like the "Super 6" utilizes, the light will be more of a "flood" than a "spot". Also, the "pee green" color that some white Star/Os have does not seem to be as much of a factor when the plastic collimator optics are removed.

4 NiMH AAs in series should yield 4.8 to 5.0V,
so a 5 Ohm, 1 Watt resistor for EACH Luxeon should
give about 300 mA per LED, 6 Ohms will drop it to 250 mA, etc. (R = E/I) A switching regulator would increas the cost and complexity, but raise efficiency somewhat, and yield more runtime.

There are 9 A/H D Cell NiMHs available, about $10
each, in small quantities ( www.ccrane.com ), sure they are larger and heavier, but they will give you five times the "mileage" per charge.

My favorite "utility" battery pack is 8 AA NiMHs in one of the 8-AA cell carriers from Radio Shack.
9.6 to 10 VDC, and us really useful for most of my LED projects.
shocked.gif
winkie.jpg


/ed brown in NH
cool.gif
icon14.gif
 

remuen

Enlightened
Joined
Nov 26, 2001
Messages
600
Location
a place surrounded by Europe
Originally posted by Hemingray:

The best AA NiMH cells are about 1800 mA/H
<font size="2" face="Verdana, Arial">The best ones already do have 2000mAh (eg. Panasonic)
smile.gif


Originally posted by Hemingray:

... so if you need a runtime of 90 minutes with 4 in series, the maximum total current draw can be 1200 mA (1. amps), probably should aim for no more than 1 Amp, just to have a little safety factor. So, that would be 50 LEDs at 20 mA each, proportionally less if they are driven harder. That would allow 3 Luxeons at 350 mA each, or 4 if they are lowered to 250 mA each, 6 @ 165 mA each, etc...
<font size="2" face="Verdana, Arial">In your calculation you've forgotten, that if the NiMH's are switched in series we have to use either a resistor or a buck converter which both have some losses. Switching all 4 NiMHs in series would mean that the resistors would cause a loss of 20 to 25%!!! In this case a boost converter and 2 x 2 NiMhs (2.4 volts) would be a more efficient solution (somewhat around 15% losses).

And of course after these 90 minutes we should have a bit more than only moonlight. Therefore 30 Nichias driven with 20mA or 2 LS driven with 300 - 350 mA wil be the maximum we can use. The LS do have a better light conversion efficiency so they will be brighter. See here: Luxeon Star LEDs compared to Nichia LEDs

Originally posted by Hemingray:

There is no physical or electrical reason that Luxeons would not work, the water might aid in drawing off excess heat. If you use the Star/Os (with Optics) the beam will be fairly narrow and sharp, if you use "naked" ones like the "Super 6" utilizes, the light will be more of a "flood" than a "spot". Also, the "pee green" color that some white Star/Os have does not seem to be as much of a factor when the plastic collimator optics are removed.
<font size="2" face="Verdana, Arial">I never had any doubts the LS wouldn't work under water. My only concern is the beam means the light distribution angle as I have to use the collimator (I don't need a flood light). The pee green color is not really a problem as long as the main light has a color temperature > 4500K

Originally posted by Hemingray:

A switching regulator would increas the cost and complexity, but raise efficiency somewhat, and yield more runtime.
<font size="2" face="Verdana, Arial">Some of our best modders as ElektroLumens, Lambda or Daniel Ramsey wouldn't be happy reading your words
grin.gif
as in some cases a resistor is not only much simplier und cheaper but could even be more efficient.

The main advantage of a regulation is: We have a more constant output and of course we can boost the output voltage from eg. 2.4 volts (2 NiMhs in series) to eg. 3.4 volts.

Originally posted by Hemingray:

There are 9 A/H D Cell NiMHs available
<font size="2" face="Verdana, Arial">I also would prefer c or D cells but the 4 AA NiMHs are given and I must accept this ...
 
Top