Uh-Oh, I have my first 2D LED Maglite... FAIL!

The first LED Maglite I have had to play with, gifted to me by another kind member here.

I have always wanted to see if one could use a high powered XM-L LED in one of these things and still retain something like the stock focus adjustment, with the stock reflector and center pedestal.

I guess I will play around with it some... In case some of you haven't taken apart one of these, I have a couple photos - Naturally Wink

magled1

Which emitter is that? Luxeon?

magled2

Side view

magled3

Screw removed and sub-assembly is out.

magled4

Pulled the spring off and there's a cup inside, which does not fall out.

magled5

It doesn't come out the top either, so just set it down and push on the center LED assembly, with two thumbs (only one showing here, had to use the other hand to run the camera).

magled6

LED assembly (pill) and the spring cup.

magled7

Now what can I do to replace the plastic tower? Looks like a copper coupling would be about the right diameter and a piece of copper pipe would slide into the coupling in place of that steel one over on the right. Hmmmm, it might just work. Fill that copper pipe so it's a solid "pill" and the Nichia would go on top, slot the copper coupling so it works like the plastic tower. Put the screw hole in the filled copper pipe and it will work like the original steel insert.

But I need more mass,

magled8

So, how about using a slice of Aluminum that fits in the body and attach the copper coupling to that so it has a base plate and it gives mass and it will contact the body for more heat transfer?

But, what about a driver? Fit a 7135 driver up in a hollow in the Aluminum and it sits right above the switch contact. Interestinger and interestinger...

And what about transferring heat from the copper center "pill" that moves, to the copper tower? That's an easy one, but to make it more interesting, let's use that copper "pill"/tower as the negative contact. Instead of thermal grease between the two, let's use some grease that conducts electricity.

I've got it all in my head, now I just have to do it and work out all the bugs, but you got the gist of it now. Maybe?

More to come............

----------------------------------------------------------------------------------

08-11-2012

The new tower/heat sink is partially made, so here's a photo:

copmaghs1

I ended up using copper to make all of it. When I got to looking at it, I knew I wanted to keep dimensions as much as possible and I wanted it to be fastened to the top of the switch housing. That meant, it would have to sit against the "C" ring in the body, so I made it out of four pieces of copper, plus the coupling sticking up.

I cut a 1" copper coupling and heated it just about red hot (to soften the copper), then I flattened it and cut out three circles. I also cut out a circle (a little larger), out of 24 gauge copper sheet. I soldered them all together and used a drill and dremel to cut out the center hole, where the coupling goes in. Then I drove the coupling in and soldered it. Now it's a one piece tower and heat sink.

It's not going to be a fantastic heat sink, but it should support the Nichia led I am going to use. I still have to cut the slot in the tower and make the solid pill, as well as a lot of clean up, so there's plenty left to do.

That's all for the moment.

-----------------------------------------------------------------------------------------------------------

08-12-2012

I'm not one to say I'm proud of something, cause usually I'm not satisfied with the results enough, but there's always a first. I am proud of how this tower came out. I never thought I would be able to make it and have the tolerances come out well enough to have it work.

hstower1

The tower itself is composed of a 1/2" coupling with a slot cut in it. Inside is a 1/2" pipe, filled with copper shot and solder. On top is a 3/8" copper coupling filled with copper shot and solder. The top and bottom are covered with copper sheet.

hstower2

Here's how it will set on the gutted switch housing. I will use the original spring, but it will be modded for less resistance. I happened to have a 3mm tap & die and that was what I used to make a threaded hole for the stock guide screw, in the slot. It all works as it should and it is very smooth movement and no wobble.

hstower3

I thought at first, that this thing was cocked a little, but I took this shot across the top and it looks like it's pretty straight and level. I really did not think it would come out this good and I'm betting I couldn't do it again if I had to.

That's all for now. I still have to dril for a wire and mount the led on top, as well as many other little things.

-------------------------------------------------------------------------------------------------

08-13-2012

Well, just to show the rest of the process, here's how the assembly went together (up to a point).

sw1

In order to keep the Positive isolated, I had to wire directly to the switch, so the spring would not be "live".

sw2

Instead of the metal cup, I put a plastic ring down there, so the wire could go thru.

sw3

I soldered the other wire to the negative terminal and fed it up thru.

sw4

I put AA in the center to seal it and isolate it, then (no photos), I put the rest together and wiored it up.

No photos, because I didn't take any before testing and it was a FAIL. I will explain in a minute.

bstd

Beam shot standard, the bedroom during the day.

b1sec

1 Second shutter, IOS 200, Aperture F:2.7 Daylight white balance.

b1-30sec

1/30 second shutter, everything else the same.

bprogram

Program Mode - Shows just about what I saw color wise and a little less bright than what my eye saw. (Our walls are not white, they are some kind of "pale peach", (according to the wife). Not condusive to good beam shots.

OK, here's the SCOOP on the Poop.

I assembled everything and put it all together. It worked for a minute, I mean the turning of the head moved the led up and down, then POP! The screw that the reflector touches against came out. The copper is too soft to support the strain. That's why Maglite uses Steel. I was concerned about this. It pulled the few threads right out of the copper. OH WELL, that's how you learn, by trial and Error. I see the error of my ways now. LOLEmbarassed

I went ahead and put in a 3mm by 10mm long screw in the hole and threaded it in as far as it would go and that works. The head can still be rotated and the led assembly will go up and down, but because it's just a screw instead of a roller, it is stiff and something will break, sooner or later. So..... now it's a garage light, till fall comes and I can strip it down and make a 1D out of it, to sell.

What about the beam shots? Ok, I put in 2D cells (Alkalines). It's plenty bright and brighter than the stock light, But.... DO NOT stipple the Reflector!! It turns it into a Mule. Well almost. It almost eliminates the entire spot all together and turns it into all spill and adjustment has no effect on it. Turn all I want and it's still all spill.

So, Jack, that answers your question. Do not stipple the reflector unless you like the beam in the photos.

Philips-Lumileds Luxeon Rebel emitter.

Careful with the 7135 drivers: they control the current on the negative side so using the tower as the ground could potentially bypass the circuit and give you direct-drive. also, how are you going to power it? 2D isn’t enough voltage to put the driver into regulation. 3C? Cut down to 1D and use 4AA? 1 X 18650 or 26650?

I had given the Rebel MagLED mod some thought before and my thought was to turn a piece of aluminum into a tower to replace the plastic one and the pill (as shown in your third picture from the top) would be replaced with a copper piece. I never did it, though.

Always interesting OL!
Man… you have golden fingers!

Yes, that is a definite issue to be looked at.

I’m not sure. I am thinking of using 3 Sub-C NiMHs that I have lying around. Just because I have them and I would like to use them up.

A 3C-NiMH setup in that host would be great. Those things pass a ton of sag-free current, and you’d be keeping Vin very close to Vf so the 7135 drivers will be running optimally and most efficiently. ( if you decide to go that route) PVC couplings can be modded and used as adapter tube.

The problem with Sub-C batteries is that good cells are getting harder and harder to find. You may have cells for your light now, but, what’ll you do when they wear out and you need to replace them?

Since you will probably end up sputtering the reflector of this light , would you mind sputtering it first and showing what it does to the beam in stock form ?

I hate the lack of spill in the current generation of LED Mags .

Thank you .

Well, I have Sub C NiMHs and I don’t have C NiMHs, so I will be testing this with Sub C NiMHS. :slight_smile:
Whoever gets it, can use C NiMHs, if they desire.
I can make it so C NiMHs will go in and still test it with Sub C NiMHs.

@jacktheclipper - I would love to do that, but unfortunately I am already under way with the mod and the plastic tower is gone forever. 0:)
If I can remember, I will do it on another LED Mag this fall. Sorry :frowning:

Has any one else sputtered a reflector in a current LED Maglight and di it improve the quality of the spill ?

I think that a better beam profile is possible over the ( IMHO ) throw-heavy profile of the stock reflector ...

For up close use the current version of LED mags is less than desirable .

It . Sucks .

Photo of the tower, in the making, updated in the OP.

Looks like this thing is going to be direct drive. A 7135 driver is going to be too big to fit in below the tower, so I will go with 3AA batteries and make it into a shorty, either that or just use 2D Alkalines, which will work direct drive on the Nichia (done ’em before) and let it go as a 2D. The Nichia will light up at 2.2 volt and at 2.8 to 3 volt, it will be brighter than the stock led.

The tower on a led light is much shorter than the tower on an incan. The slot comes all the way down to the top of the switch and even though the pill does not travel quite that far, it’s too little space to have a driver in there. The only other option would be to put the driver before the batteries, on the back side of the switch and I don’t even want to go there.

Has anybody tried putting the driver in the tail cap between the spring and the body? Just thinking out loud, sorry.

How would you run the positive terminal? I can see the negative there, but where would the positive go?

In case no one has done it, I can tell you that every type of reflector I have ever sputtered (and I do a lot of them), has smoothed out the beam and given more uniform spill. I can’t believe it would not do that in a stock maglite led reflector. Total output will go down, it always does with sputtered or OP reflector and with a stock light, that’s not necessarily a good thing, but there’s always give and take. For up close use, a little lower output would be less noticeable.

More photos and info added in the first post.

Gods man....that copper tower you made looks outstanding! Hells yes you should be proud! (I feel a little pride now just by my association with you :) )

What he said .

Orsm effort. love it.

that’s a truly beautiful piece of work. shame it’s going to be stuck inside a light where you can’t see it!

You need to get into the steam pump/ engine hobby, your work would fit right in!

FAILED PROJECT. Read all about it in the first post. Thanks for watching. :smiley: