[PSUBS-MAILIST] Light Experiments

Sun Apr 16 21:14:25 EDT 2017

Alan, your light looks like it is coming together nicely.  Good job. I was
looking at the pictures of the unit both the assembled unit in the sink and
the disassembled unit to see what you could possibly do to reduce the array
temperature.  A couple of observations:  Acrylic is a good thermal
insulator and off course aluminum is a great thermal conductor so as you
note, the back if the enclosure has got to do the bulk of the work in
conducting the heat away from the array.  You could increase the heat
transfer rate away from the array several ways.  The first is that you
should flip the light in the sink so that the light comes out horizontally
rather than vertically.  You want the fins to be oriented vertically.  Since
the water is essentially static in the sink, you have to rely on natural
convection to pull heat away from the fins.  The concept being that as the
water around and between the fins heats up, its density drops and it
becomes more buoyant relative to the colder waters so it wants to go up.
This sets up a natural circulation pattern. Having the fins vertically
promote this flow.  The second observation is your heat transfer rate would
increase if you had more surface area in the fins. If you look at most of
the commercial LED lights, they have more fins and they are thinner than
yours.  I am assuming you are using thermal grease between the array and
base. IF not this would be a problem. One other point; Borosilicate glass
is about nine times more thermally conducive than acrylic.  This would aid
in transferring more heat out the front lens.  These three changes, testing
with light to optimize natural convection, increasing the number of fins
and reducing their thickness and switching to Borosilicate glass should
drop the temperature of the array significantly.

Cliff

On Sat, Apr 15, 2017 at 11:33 PM, Alan via Personal_Submersibles <
personal_submersibles at psubs.org> wrote:

> I put a temperature probe inside the housing of my 80W LED light.
> The probe was up against the acrylic lens, as I was concerned about
> what temperature the lens was seeing on the inside. The lens is 10mm
> away from the LED.
> The LED manufacturer told me that on normal LEDs the temperature
> out the front of the LED can be 338F (170C) but my flip chips should
> run cooler.
> Briefly; when run at 30W in water they stabilised at 257F after 5 minutes.
> At 45W they stabilised at 280F after 5 minutes. I stopped upping the amps
> at this stage.
> Next experiment, I filled the housing with silicone oil.
> At 45W it hit 116F in 4 minutes & temperature stabilised.
> At 76W it hit 139F in 6 minutes & stabilised.
> So oil has massive advantages in lessening temperature on the lens,
> maybe because it is a buffer from the radiant heat & also transfers the
> heat
> out through the housing quicker.
> Greg your thoughts would be appreciated on this; I am thinking that
> without the
> oil the acrylic lens would see temperatures at which it is formed
> especially
> if I went up to 80W. Although the outside of the lens was cold there would
> be a
> temperature transition across the thickness of the lens & it would be
> considerably
> weakened. I am building to 500ft, so it would potentially see 250 psi in
> operation.
>    At this stage I haven't made up my mind whether to go with a
> borosilicate lens
> or oil fill with an acrylic lens. I had a large bubble & it would be hard
> to eliminate
> all bubbles in the oil filling process.
> Will try & attach photos.
> Alan
>
>
>
>
>
>
>
> Sent from my iPad
> _______________________________________________
> Personal_Submersibles mailing list
> Personal_Submersibles at psubs.org
> http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20170416/c5b32bc2/attachment.html>