Convert Pelican Dive light to LED?

sw629

Newly Enlightened
Joined
Jan 3, 2010
Messages
8
Good day all.

I've got a Pelican Laser Pro 6000 (4 D with dual filament #5003 Bulb) that I would like to convert to an LED. Tried se3arching here and google but to no avail. The bulbs don't last very long anymore (or maybe it's me that's not lasting as long anymore) This is a great light and I don't really want to shelve it. Any help would be greatly appreciated.

Cheers, Will

NB I've also got a couple of Pelican King Lite 4000 lights (8 D #4003 bulb) that I have already retired because bulbs are getting harder to find and I have newer lights that replaced them, but maybe an LED option for them as well?
 
Last edited:

alpg88

Flashlight Enthusiast
Joined
Apr 19, 2005
Messages
5,282
not really. you can't dissipate more than a watt or two inside sealed housing. and being it underwater doesn't help if heatsink is not touching the water directly. can't do much with them, unless you rebuild entire head, but that would make it more expensive than getting new light. i have pelican 8d too, good light but outdated.
 

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
I joined CF forums because of a similar problem and looking for lamp info. The Pelican King Lite is a very rugged waterproof housing and with 8D's fitted (or 2x1800mAh 3S LiPo in mine now) they would run a LED head for days. The two position switch could be used for two LED arrays or switching modes if the driver had the smarts.

The original twin filament lamps pull around 12 Watts anyway, so heat from a much more efficient LED replacement shouldn't be a big problem I'd have thought (say 3W to 6W ).

I've already started looking at a simple solution and one that jumps right at me is using 12V LED down lights. The flood brightness is actually quite similar to the King Lite flood lens and lamp.

The part I need info with is finding a cheaper or dead Pelican 4003 bulb to rat the bayonet base out of. If I have this I can make the LED head interchangeable with the spare reflector and old Xenon lamp.

Does anyone know what the Pelican 4003 bayonet type is called ?
The bayonet is 15mm diameter with 6.2mm spaced twin lead contacts on the base. It's very close to a car brake lamp but the locking lugs are not in the same location as those.

Thanks,
Martin
 

alpg88

Flashlight Enthusiast
Joined
Apr 19, 2005
Messages
5,282
The original twin filament lamps pull around 12 Watts anyway, so heat from a much more efficient LED replacement shouldn't be a big problem I'd have thought (say 3W to 6W ).


Martin

you can't compare the two.

bulbs don't need the heat to be moved away. bulbs don't care if they are hot, leds do.
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
It's true you can't compare the two. But if you can get a heatsink inside the housing, with enough square inches of surface, you could dissipate several watts inside the plastic. I'd say you want two square inches of surface per watt. That's counting both sides if they are both exposed. Thus a 2 inch diameter heatsink with one inch diameter covered on both sides would give a bit over 4 square inches, enough for about 2W of LED.
 

PCC

Flashlight Enthusiast
Joined
Oct 28, 2007
Messages
2,326
Location
Sitting' on the dock o' The Bay...
It doesn't work that way. Incandescent lights produce the heat in the light itself. LEDs produce heat as a byproduct of the way it works and the actual LED itself gets hot. You can have a thermal runaway situation powering an LED with just two watts if you don't use a heatsink. The maximum you can drive an LED with minimal heatsinking is about 400mA, slightly more than one watt. Placing a huge heatsink inside a plastic housing with no way to dissipate the heat is about the same as minimal heatsinking and should be treated the same way. In other words, you can convert your light to LED, but, if you don't do some major work to get the heat out you're limited to about 400mA (about 130-140 lumens) max or you'll be going through LEDs faster than you were going through bulbs.
 

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
Interesting debate, I agree that incandescent lamps don't care about some heat. Given that most of the energy from the incandescent lamp is lost in heat, the fitting still has to bare it and dissipate it I think (or melt). So I'll not give up hope just yet as there is a bit of metal in the Pelican lamp head base :)

What if the LED fitting that I use is already designed to be run in a sealed up plastic housing (instead of a home down light), and is close enough to the original bayonet that a washer might make it compatible ?

To that end I've bought such a LED lamp online, a 3W (allegedly) automotive white LED tail lamp, with a BA15S bayonet base. Car tail light housings don't have much air movement and in fact also get hammered by heating from the sun as well while in operation (daytime running or brake lights). I'm thinking this might work in the Pelican as a flooder. So I'll give it a try and report back, and if it works well I take some photos of the modded reflector (will need a 15mm hole cut in it). From the specs I've managed to find, a BA15S bayonet LED lamp is the same diameter, and the locking pins look visually to be in the same position as the Pelican #4003.

Cheers,
Martin
 

PCC

Flashlight Enthusiast
Joined
Oct 28, 2007
Messages
2,326
Location
Sitting' on the dock o' The Bay...
Car LED brake and reverse lights (the brighter ones) don't operate all the time unless you're in bumper to bumper traffic so they don't heat up that badly. If you do run an aftermarket LED bulb in a housing designed for an incandescent bulb in LA traffic in the summer and are caught in a traffic jam then I'd say that you will be replacing tail lamp assemblies.

The simple fact remains that you need to get rid of the heat somehow and not doing so will cause components to fail. Sure, you can load up that plastic light with 2lbs of copper and run an SST-90 at 9A for a few minutes at a time, but, how useful would that be? What happens when you need to run that light for more than ten minutes? The answer is you'll blow the emitter.
 

alpg88

Flashlight Enthusiast
Joined
Apr 19, 2005
Messages
5,282
not to mention those led tailbulbs fail more often than regular bulbs,
these bulbs are very bad example.

but hey, don't take our word for it, do it than tell us how it went.
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
The plastic housing isn't a perfect insulator. If it was, you couldn't run anything in it without it overheating. The incan bulb will radiate some of the wasted energy out the front as IR, but some of it soaks inside as heat. If the housing were a perfect insulator, this would eventually melt things inside the light.

So if you can get the heat from the LED into the air inside the light, and the air can circulate by convection, you can get a little bit of heat out through the wall of the light without overheating.

The question is how much is a little bit. As PCC says, a watt is almost nothing. I'd bet you could do two watts without jumping through major hoops. That could get you 200 lm OTF, which is probably way more than the original incan could ever do, without sacrificing too badly in the LED lifetime.
 

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
I've completed step one of the test of a LED conversion. I've bought a LED 'tail light' from DX and had a play at fitting it to the King Pelican. Turns out to be relatively easy as the base diameter and locking lug positions are a match.
A mod is needed to the lamps base contact, and the main reflector needs a 15mm bore out from the original size. The internal spring is tossed out, and the lamp is used as a flooder (forget any throw ability). Actually I've found that a flood pattern is more useful when immersed for snorkelling or spear fishing at night, so I care not for throw.
I have both the diffused and diffused reflector for the Pelican, and I find the diffused one works best as the LED tail lamp is segmented into five facets. The facets cause a slight flower pattern to the light output.

When I get time I'll turn the light on for 20 minutes and see if if does overheat, but so far it doesn't seem to get very warm.
After that I have a more tricky thing to sort out. The DX LED tail lamp is not current controlled, it just uses limiting resistors. Performance plummets when the LiPo's get below 11 volts, so I think I need to try to fit a small current driver OR somebody can let me know if any of the DX LED tail lights (white) have a real current driver module inside them. If so I'll save the effort and just buy one of those.

I have a bunch of photos if any cares for more info, I'll post two now anyway if I can sort out how...

Cheers,
Martin

P1081443.JPG

P1081446.JPG


Edit: Tried sharing images from Google+ and DropBox, fail...
 
Last edited:

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
I would enjoy seeing the photos if you figure out a way to post them.

There is another possible way to approach the heat transfer problem that I have considered for my old 8 D cell Ikelite. Consider to mount the LED "facing" the reflector.

If the LED thermal pad is mounted on a copper or Al rod that passes through the front light lens, it will provide a considerable thermal path to the water. The downside of this is likely the beam will be less perfect, and of course there is some risk to drilling through the faceplate, but at least it gives you another option. Perhaps if you put a "step" in the rod, you can use it to form a seal to the front face.

Two drivers that are easy to use - taskled CC1A or LedSupply buckpuck each can supply 1 amp for quite a discharge range of the battery setup you have for around $ 15. I have used both with no problems. That combination and quite a number of LEDs options can produce decent light for a long time.

If your goal is to "see" rather than "photograph", consider to experiment with a shorter wavelength monochromatic LED. I have been told that light in the 488nm range is quite good for penetration in seawater but I have not tried it yet. Obtaining LEDs in that wavelength is not always so easy.
 
Last edited:

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
My posting permissions show 'You may not post attachments' so perhaps I'm just forbidden. No doubt this is to reduce spam or something, but it's not helpful right now. :)

See if this works, I just pasted the links below rather than used the link tool in the message editor...
Let me know if they are visible and I can add some more.

https://www.dropbox.com/s/dj11vcx1v81cjoc/P1081443.JPG

https://www.dropbox.com/s/tc70ys2dwtqp0up/P1081446.JPG

Actually my goal is a bit vague. I think it's more to reuse the nicely made light rather than throw it away. It will get used for seeing rather than UW photography though, but I do have an UW camera so I guess I could do that one day.
 
Last edited:

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
Thank you for the pictures, personally I think it came out pretty nice. The heat transfer aspect is dicey, but you never know. As long as you have a backup light of some kind it probably does not matter.

I think that once your post count increases to 10 or something they let you post pictures but the drop box links worked.

Your project has inspired me to think about a few mods I could do on some old lights around the house. Now I just need to finish a few hundred projects in front of it so my wife doesn't freak out.
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
This forum doesn't allow attachments. You have to "insert an image" using the appropriate icon in the toolbar (mouseover, and you'll see it).

You also have to edit the link you get from DB. Change the "www.dropbox" portion to "dl.dropboxusercontent".

Your links in post 13 didn't work for me, but when I changed the URL as above, I was able to see the pictures. Nice, by the way.
 

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
Inserting a couple more (I'll try the insert image tool again). EDIT: Tried and failed again, so used Image Shack links.
Still not done much about adding a driver, but I have found a tiny one meant for 12V down lights in my parts box. It also came from DX some time back for another LED lighting project (Quadcopter).








 
Last edited:

mr.sneezy

Newly Enlightened
Joined
Dec 22, 2013
Messages
9
I've ordered another type of BA15S style white LED lamp, supposedly it has more output and has a 12V-24V driver (yes I know more output is more heat). I just figure that if I am to play around with the LED retrofit idea I may as well push it a bit further.
The lamp is this one
http://dx.com/p/1156-ba15s-p21w-7-5...g-light-tail-light-signal-light-12-24v-277989

I suspect it's going to be a hot runner, but I have a cunning plan.
If I can fit a thin thermocouple probe to the base or alloy LED holder (I have such probes), and run it out through the King lights switch hole (remove switch lever), I can monitor heat easily enough.
If it runs too hot on the LED holder case (somebody suggest a MAX allowable temp for me please) I'll open the lamp and tweak the driver for lower current, until heat is under control (done that before).
Hopefully there will be some usable light at that point. If not then I've only spent $20 total and learned a bit more about LED's :)
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
Max temp is a very difficult question to answer without great detail. It depends on how long you want the light to last, how good the LEDs are, and how good the thermal transfer is from the LED die to the point you measure temp.

The closer you can get the probe to the LEDs, the better.

The LEDs are probably crap, and you probably want them to last a hundred hours or so, so I'd say keep the die temp below 120C. At that point the substrate under the LEDs is probably 110C or thereabouts. Looking at that lamp, it looks like it has an aluminum body. If so, then mount the probe on the body as close to the LEDs as possible and shoot for <100C.

Of course, having none of the necessary information to make proper calculations, this is only guesswork. If you are bold, get a few and run them for a couple days at various temps and see what happens. That would be excellent information to have. A cheap luxmeter would be an awesome addition to the test setup.

On the other hand, these are designed to work in an automobile lamp assembly, which will have no heatsinking, and ambient temps up to 50C-ish. If you can keep the ambient temp around the bulb below 50C, you should be okay.

By the way, even though these are designed for use in cars, don't put them in your car. Put on your flame suit before mentioning bulbs like these in the automotive forum. Your idea, however, sounds like a very appropriate use for them.
 

DrafterDan

Flashlight Enthusiast
Joined
Jul 28, 2013
Messages
1,317
Location
Phoenix, AZ
considering that the original poster asked the question in 2013, and then was heard from no further, I'd say no. Seems it firmly falls into the "more trouble than it's worth" category.
 
Top