Luxeon Star in Mag Solitaire !

I hope this time I can hold the "smallest mod" title a little longer

First of all, pay another visit to the nearby university biology department and sweet talk with a nice grad student---This time in addition to a 0.5mL Eppendolf tube
also ask for a plastic cap for a 1dr-sized glass vial. You will need to cut a spacer out of the cap.


Next, bore/file the end of Solitaire tube until the opening is the same ID as the switch (ie, OD if the Eppendolf tube)



I used the standard Luxeon Star driving circuit powered by
ZXSC300 and cramed the whole thing into the section cut out from the Eppendolf tube.

A small disk with diameter of the battery tube was cut from a tin can. LS-emitter (Q2H from Lambda) was glued and soldered on one side and the circuitry unit was glued on the other. When finished, the metal disk will sit on the rim of the battery tube, function both as a negative contact and a heat-sink.





The arrow points to the metal disk.

Now put the lense and the plastic spacer and the head on

And you are all set---Note that now the ON action is to tighten the head, and the OFF position is to loosen the head---opposite to the original MagLite.



A little beam shot between Solitaire LS Mod (left), Solitaire 5mm Nicha Mod (right) and Photon I (the dim blue spot on the top)



I couldn't find suitable optics to produce a good focused beam. I tried to cut out a section from the NX05 lense and polish up the surface. It did produce a focused beam, but only about half the light was actually reflected (by eyeballing the beam brightness with the original NX05) so I just opt for a flood light. If anybody can send me a better lense, I'd be glad to try it on.

The current draw is barely 200mA (due to the thin wires used, you just can't push more current through, even if you totally ground the feedback pin. Although later I thought if I'd put less turns on the inductor, I might be able to push the current a little higher) so the LS is only driven at about 70-80mA. But the high quality LS (Thank you Lambda!) produced a lot more light than a 5mm Nicha overdriven at 70mA. For a flashlight as small as Solitaire driven on a single AAA, the light output is really more than enough to get you around the camp or read in the tent---and I am more than happy for the expected 2+ hours battery life.

Nice job! I reckon you have earned your name! Did you have any help from nanoroids??

- Don

Very nice--congratulations!

Cheers,

Victor

for sale?

Nice mod - wanna one too

Try a NiMh instead of Alkaline to see if you get more juice going

Klaus

Mr. LEDmASAP Very Nice!

I wonder if someone makes an acrylic ball small enough to make a lens from......

You certainly are pushing the size limits, and getting away with it nicely. I like!

Mr. LEDmASAP
I have some small lucite balls of 1/4 and 3/8 inch if you want to try them sent me a PM and your address and I'll send them to you.
Chris

WOW, very impressive!

Now, make 100 more and sell them back to us

Someone needs to find a medical supply outlet to get some Eppendorf tubes and vial caps.

There's nothing left to say...
WOW!

http://cmgm.stanford.edu/~vesicle/karen.html

Eppendorf tubes.

Out of curiosity, did you try the one-transistor LED driver circuit you used in your previous mod?

Probably that circuit without a resistor on the transistor base would drive the LED to the limits of what the battery allowed. (Most likely regulation wouldn't be an issue, as the battery's internal resistance would prevent you from overdriving the LED.)

BTW, thanks for that link to Eppendorf tubes.

My other question to LED Mods - Where did you get the ferrite bead used for your inductor in the first mod?

suggestion for a very important thread to come:

the big sale of lots of solitaire-1W-LS mods



bernhard

I know Target has most of their Mag's on sale right now. I just picked up a 2D for $13 and change.

LED,

Nicely done! Solitaires can be a rewarding challenge, can't they? Now try this mod with a HD or SE in a Sear's 1 AAA tool light [Dorcy?]. They cost about $3 US [at least back at Xmas], have a focusable reflector and a clickie tail switch.

Keep up the good work,
Dan

Good point PercaDan,
The Sear's (Craftsmen) 1 AA tool light with the tail switch and the regulated LS is something I would dream to have. Get the regulation to 100mA and a little ball to throw something and I will beat a path to your door!

Wow, I didn't expect such a flood in responce to my post

This mod is just a home-brew prototype, it haven't been thoroughly tested yet---like dropping it to the ground, bumping it around inside a toolbox, light it in the rain, etc etc, and it looks really amateur. So I don't have plan to sell it to anyone just yet. If I can find a place to make a few small PCB's and make the unit more professional looking, I may offer to sell a few of these drop-in plugs.

Another thing is the small inductor, it took me a good while to make one (see the details below). I am not happy with the reliability as is.

To answer Entropy's question, yes I have tried the single-transistor circuit, and it worked. But the converter efficiency was no where near the IC-driven unit. I pulled over 250mA, and the transistor got really really hot---so hot that it actually had a thermo-breakdown---the circiut would stop oscillating unless I hold the transistor between my fingers to keep the temperature down, and a burning sense developped within seconds.

I guess in a Satcure circuit, the transistor was driven by a triangular wave rather than the square wave in IC-based circuits. The transistor would see high current and high voltage at the same time; whereas IC-driven transistors see either high voltage/low current, or high current/low voltage. So eventually I settled down using the Zetex driver.

The ferrite bead was still the choke core (JW Miller Magnetics, model number FB73-422, 11mm in length and 5mm in outer diameter, ordered from Digi-Key, part number M2308). I cut out a small section with a fine steel file---it's actually very easy, just file a few shallow grooves along a circle, and a piece of the ferrite will break off during the filing. Then smooth out the edges on a piece of sand paper.

As I mentioned in an [URL http://www.candlepowerforums.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=3;t=003872#000000] earlier post [/URL], this core is conductive. So I coated it with plastic---heat it up with a soldering iron and press against a glue stick, then wipe off excess by some tissue. It worked ok, but I believe if I could find some non-conductive cores the efficiency of the converter will be better---the few hundred ohms on the core does waste some energy.

For the LS-HighDome's, I don't know if it's just my bad luck or it's general, but anyway the only one sample I got from Lambda was not as bright as the LD's (when selling those HD's, Lambda did point out they have no BIN's, so they may not have been the highest grade out there.) and currently I don't have any SideEmitters. I will see if I can get any without having to flattern my wallet---Canadian Dollars are falling constantly, and everytime I bought something from the US I got heavy taxes

But anyway I plan to make more mods and I will keep you posted...

Originally posted by LED mods As Small As Possible:
Wow, I didn't expect such a flood in responce to my post

To answer Entropy's question, yes I have tried the single-transistor circuit, and it worked. But the converter efficiency was no where near the IC-driven unit. I pulled over 250mA, and the transistor got really really hot---so hot that it actually had a thermo-breakdown---the circiut would stop oscillating unless I hold the transistor between my fingers to keep the temperature down, and a burning sense developped within seconds.

I guess in a Satcure circuit, the transistor was driven by a triangular wave rather than the square wave in IC-based circuits. The transistor would see high current and high voltage at the same time; whereas IC-driven transistors see either high voltage/low current, or high current/low voltage. So eventually I settled down using the Zetex driver.

Click to expand...

<font size="2" face="Verdana, Arial">I wonder if this is "normal" operation for the Satcure circuit or if it was being overloaded and hence not running in the most efficient manner.

Have you monitored a Satcure circuit powering a lower-current LED with an oscilliscope to see what kind of waveform it has? Maybe at these current levels a transistor with higher gain is needed. (Maybe a variant of a Darlington pair. Have a small transistor driven from the inductor, which is used to drive the higher base current requirements of a larger transistor that is used to perform the switching.)

Maybe I'll need to build up a Satcure circuit on a proto board and pull out the old 'scope in the basement. Don't think I could get away with borrowing one of the nice Infiniium digital scopes from work.