Planned IR Handheld Build: Driver?

[Wolfman29]

Hey there people.

So I recently received my MXDL host from DX (or DinoDirect? I always get those two confused) and I like it. I made an aluminum heatsink for it, but I just realized that it only has room for a single 14500 cell... and 1/2 AAs are rare enough as it is, even without them needing to be li-ions.

So, I decided that I could stick an IR diode in it! I know it's dangerous, but this isn't my first laser, I have built quite a few (most of which failed, but two are working properly, a third soon), and I will get a pair of safety glasses for this before I complete the build, of course.

However, now I have to get to the question:

I know that 808nm diodes are case positive. Considering a case positive build, and considering that I wanted to do DIY driver build (I can make an LM1117 driver that will drop out little enough voltage to run an IR), how do I go about doing this? I can't have the driver have a constant positive rail... so could I possibly use an LM1117 circuit as a current sink instead of a current source? Just reverse it and put it in the negative side of the circuit?

Thanks!

 

[rhd]

I've pondered a similar scenario myself.

You have two options, as far as I can tell:

1) Use a NEGATIVE regulating alternative to the LM1117 and basically build the same circuit, but build it to regulate the negative rail. Potential problems include the fact that I'm not aware of a negative regulating alternative to the LM1117. The negative version of the 317 is the 337 (I believe), but that's not low dropout. Maybe some searching will yield a negative regulating LDO IC. Give it a shot.

2) Bore out the aixiz hole in your heatsink enough that you can wrap (isolate) your aixiz module with kapton tape. Then tap the side of the host near the driver for the negative feed from the battery, and basically do everything else as normal. With your aixis isolated, you can build a typical DIY driver.

 

[Wolfman29]

So no using it as a current sink? I threw together an LM1117 driver and stuck it in a current sink mode and it *seems* to be regulating to 195mA... but I forgot what mA I set it to with the set resistor (I covered it in heatshrink, lol), and I will continue to do some testing... but I don't like the idea of having to isolate the heatsink... especially because I have no idea what you mean by tapping the side of the host near the driver =p

Alright, so that didn't work. Took it apart, turns out I set it to 125mA. I plugged it into two 18650s, started smoking

EDIT again: Turns out that the only negative regulator that is LDO that I can find is the LM1175... and it requires a minimum voltage of -4.3V (minimum in magnitude, not absolute minimum)... so no single-cell builds there.

It seems that there is no way to go about building a DIY handheld single cell li-ion IR.

 

[rhd]

Isolating is the only route then (although there must, somewhere, be an LDO neg reg)

To isolate, essentially:

- Kapton tape the aixiz, so that the aixiz module and diode don't get any electrical connection to the host

- With no case pin to rely on, you need the DIY driver to be able to get the (-) case feed some other way. You can try to get creative. By "tap" the case, I really just meant think of a way to get the negative battery feed from the wall of the host. Sometimes a coil of stiff copper wound inside the host 2 or 3 wraps will have enough tension to maintain a clean connection. Sometimes you can drill a hole into the thread of the head connection, and then feed a bit of exposed wire through it. When you screw the head on, the wire will pinch, and you'll get a good connection to the host. The challenge is that aluminum is hard (nearly impossible) to solder to well, so you can't just solder a lead onto the inner wall of the host.

It's probably a lot easier in practice than it sounds.

 

[Wolfman29]

OH... I see what you mean. Hmm... I am trying to decide how I could do that.

Fortunately, there is a little nook already in the driver pill where I suppose I could stick a negative lead to the host... just need to find some sort of conductive adhesive that is really cheap... anyway, bed time!

 

[rhd]

As opposed to adhesive, if you can find a way to use natural tension, I think the results will generally be better. That's why I like the idea of coiled copper wire (thick-ish), because it holds tension, and there's no way for it not to be in contact with the host wall multiple times throughout the 360 degree wrap.

 

[Wolfman29]

The only problem is that there is no room in this host for a coil... it's all very tightly fitting.

 

[CDDZ]

The only problem is that there is no room in this host for a coil... it's all very tightly fitting.

What is the ID of the host , is it possible to press a piece of copper tubing in there? a 14500 will fit inside of 1/2 copper pipe. You could solder a wire to that for a good ground.

Chad

 

[Wolfman29]

It's this one:

Portable 1W LED Flashlight Light Lamp Torch Black(1*AA), Europe - DinoDirect.com

However, no - I couldn't even fit copper foil in there. The fit for the 14500 is *really* tight. About a mil off of press fit is my best guess.

 

[anselm]

You could use the driver off a 6$ ebay green pointer, those will run off a single LiIon,
it is a sink (negative side regulator) and it will give you up to 400-500mA.

Heck, an IR laser not very different from a greenie that got it's crystals knocked off.

 

[Wolfman29]

That *is* a good idea, I just wanted to use a DIY driver for this one (I got so many damn LM1117s in all types of packages, a crap ton of SMD resistors, and 50 SMD 10uF caps! Nerd rage!)

 

[chefla]

As I said in my other post, try the LT3015. This might be the right driver for your application.

 

[Wolfman29]

Yup! Sounds good.

 

[Cyparagon]

If you already have the parts, there's no reason not to use them. Just put the regulator after the diode instead of before. It doesn't care.

 

[Wolfman29]

I think I have tried that... didn't work, as far as I know.

And just to be clear, you mean just connecting the negative pin of the diode to the input on the LM1117 and then the end of the LM1117 circuit to the negative of the battery?

 

[Cyparagon]

No. This (simplified)