[PSUBS-MAILIST] LED lights

Alan via Personal_Submersibles personal_submersibles at psubs.org
Sun May 13 15:47:54 EDT 2018


Rick,
it doesn't sound like an electrical issue if your friend can hang them
in a bucket of water for long periods of time. 
Are they screwed down hard to an aluminium heat sink with heat sink 
compound in between? There are plenty of videos on this.
The other issue could be the pwm that you say you have between the
buck / boost driver & the led. The led may not like this; I presume this is
for dimming.
You could just try dropping the voltage  till they run at 50W & see if they
last. Then of course you could have a bunch of useless leds, but if your
friend can hang them in water with success this might not be so!
Let me know if I can help. 
Alec has had success with the led light that Cliff designed.
There is a good book for learning electronics " Make electronics" by Charles Platt.
It is geared toward practical people like ourselves & reduces theory to a 
minimum.
Cheers Alan

Sent from my iPad

> On 14/05/2018, at 5:47 AM, Rick Patton via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
> 
> Thanks for your input Alan! Since I DON'T understand anything about electricity at all, this has been a real frustrating process and I have been very tempted to just go back to the plans for the K-350 and use the tried and true incandescent bulbs in a one atmosphere housing but I haven't given up quite yet.
> We just had a new fissure pop up last night, I think number 17 after a couple days of all fissures just just off gassing with no lava. Wish I had the sub done for when it reaches the ocean and it will.
> Rick
> 
>> On Sat, May 12, 2018 at 7:23 PM, Alan via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>> Rick,
>> is your buck converter constant current, constant voltage?
>> Needs to be constant current as when the LEDs heat up they draw more
>> current & can burn themselves out. 
>>    At a certain voltage the current can be kept under control but I've seen
>> graphs where just 1 more volt will cause the current to really climb.
>>    You need to screw the LEDs down hard on to a heat sink with a thin layer
>> of heat sink paste between. They will burn out if this is not done properly.
>>    I have been told that people drive the LEDs too hard & that competing
>> manufacturers tend to over rate their lights; so it might pay to drive them at
>> 80W rather than 100  W.
>>    I have glass lenses & no compensating oil. When I tried an acrylic lens it
>> trapped the heat in! You can feel the heat pour out the front with the glass
>> (pressure & heat resistant borosilicate) lens.
>>    I am using buck boost constant current drivers with dimming that I had made
>> up for me in China. They work well but there are a few things I want changed.
>> One of the components needs heat sinking as it gets too hot when the LED
>> is on high power; the on off switch is on the pcb board & I need it to be remote.
>> It is really fiddly taking off this switch & soldering an extension wire. Also
>> the default mode for a disconnected / broken wire on both the dimmer & on off 
>> switch is that the light stays on & this would kill the light if a wires soldering 
>> broke when the light was out of the water.
>> Will get them to change these things some time. If you get desperate & want
>> some of these drivers I will get on to it & order a few more. Mine are 12-54V
>> buck boost constant current with dimming.
>> Glad you haven't had a volcano pop up in your garden. Have been watching
>> the news footage & it looks pretty spectacular. 
>> Cheers Alan 
>> 
>> 
>> 
>> Sent from my iPad
>> 
>> > On 13/05/2018, at 8:51 AM, Rick Patton via Personal_Submersibles <personal_submersibles at psubs.org> wrote:
>> > 
>> > I am still trying to come up with a viable working LED light system and I think I have it boiled down to cheap arrays? I have a friend helping me that understands electricity a lot better than I do and he seems to think it's a heat issue, not dispersing it quick enough.
>> >  
>> > I am using a 10,000 lumen array that is about 1 1/2" square and costs $2.75 each and made in china. Well, you've herd the saying, you get what you pay for? These are rated for about 100 watts and like to see around 34 volts so were running them threw a buck booster? and dialing them down to that from about 38 volts and then running them threw a PWM and then to the array.
>> > 
>> >  I at first was putting them in a housing that I had fabricated for them using a strain relief fitting to run the power cord threw as the housing is full of mineral oil then setting the housing in a 5 gal bucket of cool water and they are still burning out so I tried just hanging an array by it's wires into a container of mineral oil and still only lasting a couple of minutes so I got a thin cold pack from the freezer (without telling my wife of course!) and sandwiched the array in it thinking that that would eliminate any questions of the heat dissipation theory and it still only lasted a minute!
>> >  
>> > My friend is still insistent that its a heat problem and that the thin layer of plastic that holds the frozen stuff in is enough to keep the array from seeking the relief that it needs before melting down which I don't agree with. He says he can hang an array by it's wires in a bucket of tap water for a number of hours with no issues and that water disperses heat much faster than oil.
>> > 
>> > Sorry for the long post but I felt it necessary to let you guys know exactly what I am doing for better input. I want to put this behind me so I can get on to the next thing! It would be helpful to hear from those who are using a small high luman LED array as to what you are using and where did you buy it. PS: I snuck the cold pack back into the freezer unnoticed!😂
>> > Rick
>> > The fissures stopped spewing lava two days ago so will hope for the best but looters are having a field day!
>> > _______________________________________________
>> > Personal_Submersibles mailing list
>> > Personal_Submersibles at psubs.org
>> > http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>> 
>> 
>> _______________________________________________
>> Personal_Submersibles mailing list
>> Personal_Submersibles at psubs.org
>> http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
> 
> _______________________________________________
> 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/20180514/08ed9010/attachment-0001.html>


More information about the Personal_Submersibles mailing list