[DIY] Driver for 635nm diodes, 200mA~ Cheap! And my PCB making tutorial, Easy!

[Rafa]

Hey guys!

Today I was bored, and I decided to design some CHEAP driver for low powered diodes. It is suposed to run in one Li-Ion battery, and it works ONLY for 635-780nm diodes. This means, 635 diodes, Senkat diodes, LOC diodes and CD Burner diodes. I can give up to 700mA but if you pass 200mA it will porduce a lot of heat, and we do not want that.
So, here I'm going to explain a little how it is, how it works, and how to make it.

It probably don't fit an aixiz module, but it's not designed specifically to fit in there. It is suposed to fit in this DX host:
DealExtreme: $2.50 Mini LED Keychain Flashlight (1*AAA)
And the second version of the pcb (the one in the tutorial) fits this host:
DealExtreme: $2.90 Mini 30-Lumen LED Flashlight with Carabiner Clip (1*AAA)

These hosts are awesome. A friend gave me one PHR in one of this hosts and its my favourite laser. But this one was built with some switching regulator. Check it out in my sig.

Well so, it's a linear regulator, if you want to make one you'll ned the following components:

LM1117-ADJ SOT-223 Package
4.7uF (or more) 0805 caps (two)
One 0805 resistor.

The value of this resistor depends on the current you want.
For some common currents here is some values. I give you the target current and the most accurate current you can get by using commercially available resistors.
40mA (635nm): 31 Ohm-> Use standard 30 Ohm -> 42mA
150mA (Senkat): 8.3 Ohm -> Use standard 8.2 Ohm -> 153mA
180mA (Overdriven Senkat): 6.9 Ohm -> Use standard 6.8 Ohm -> 183mA
200mA (LOC, 52x CD Burner): 6.3 Ohm -> Use standard 6.2 Ohm -> 203mA
For > 200mA Diodes like LOCs and 52Xs you can use an AMC7135, wich give constant 350mA. But I'm not covering this here.
Use the correct board version for each regulator.

You got scammed if you paid more than $5 for all of this.

You can calculate a custom resistor for a custom current by using this formula:
R1 = 1.25 / (Iout)

So if you want 40mA: R1 = 1.25 / (0.040) = 31 Ohm

Next step is to make it.
So, if you don't know how to make PCB's use my tutorial on how to make them. It's on the next post in this thread.

Here I give you the PCB files. They can be opened with ExpressPCB. I attached too a XPS file wich can be opened with windows to print the PCB.
Note that for printing you must enable the flip option in printing options, as long as this is going to be on the upperside of our PCB.

Next is to mount the PCB board in the aixiz head, and the press it into the flaslight.

[Unfinished]

 

[Rafa]

Re: [DIY Plans and guide] Driver for 635-650nm diodes, up to 200mA~ Cheap! [Unfinishe

So, you want to learn how to etch pcb's right? You came into the right post.:beer:

For this you need:
New copper board
Sand paper
Acetone
A toner printer (photocopys in photographic paper will work just fine.)
Ferric chloride
Duct tape/electrical tape, etc
A hand saw
A vice (or something to hold the board)
Double-cut file.
An iron.
A drill.
A 0.8mm drill bit.

I will post how I made the second version of the driver. The circular one.

First, we sand the copper.

We take the photocopy, and...

...we put it in the board like this.

We turn on our iron, max power.

And then we PRESS HARD into the copper board with the photocpy over it, and under the iron, sadwithcing it.

Make all the pressure you can with the tip of the iron, all arround the photocopy.

Let it get to its normal temperature. Should look like this.

Cold water

Put it under it.

Remove the paper with the finger, be careful no to remove too much or some toner will say goodbye too.

Dry it

Remove the paper. Wow! The design is there!

TIME FOR SOME SAW!!! Oh yeah. I don't think i have too much to explain here. Just be careful not to cut trhu the toner.

Now you should have this. Time for etching it. Put it under the Ferric Chloride for some time, depends on the chlorides quality. Stir the chloride while it's submerged in there.

Once its finished and dry it looks like this.

Check the board with some light. If there is any fail it should be seen here.

Now put some acetone in a cotton cloth.

Remove all the toner and paper left in the board.

Now the board is clear! And best of all, ready to solder something on it!!... But hey!, wait, the board is round. Oh yeah, round it with the file!

That's how it looks once rounded.

Time to make some holes! You better change that bit dude!

This 0.8mm bit looks better.

Drilling finished.

And here is it, with half of the components soldered in!

Hope you find this tutorial useful!

 

[comradmax]

Re: [DIY Plans and guide] Driver for 635-650nm diodes, up to 200mA~ Cheap! [Unfinishe

What's the dropout voltage you are expecting from this? I read the datasheet and it looks like 0.9-1.1V at 200mA output. I've not made a laser yet, so I am probably missing something here How would a single 3.7V battery power a 3.0V diode with this driver?

 

[Rafa]

Re: [DIY Plans and guide] Driver for 635-650nm diodes, up to 200mA~ Cheap! [Unfinishe

What's the dropout voltage you are expecting from this? I read the datasheet and it looks like 0.9-1.1V at 200mA output. I've not made a laser yet, so I am probably missing something here How would a single 3.7V battery power a 3.0V diode with this driver?

635nm drop 2.2V~ and 660nm should drop even less so with the LM1117 eating 1.2V we get 2.4V for the diode. Enough.

 

[comradmax]

Re: [DIY Plans and guide] Driver for 635-650nm diodes, up to 200mA~ Cheap! [Unfinishe

I see, so one just needs to get over the LD dropout voltage in order to light it. I asked because I've only read about people running reds at 3.0V...

 

[gillza]

This is awesome! Very nice guide! Makes me want to try this myself I think I will soon

 

[Vaporizer]

Re: [DIY Plans and guide] Driver for 635-650nm diodes, up to 200mA~ Cheap! [Unfinishe

635nm drop 2.2V~ and 660nm should drop even less so with the LM1117 eating 1.2V we get 2.4V for the diode. Enough.

Rafa, don't you mean 3.4v? With the 3.6v Li-ions coming off the charger @ 4.15-4.2v...then there is enough.
Nice etching tutorial.

 

[Rafa]

^Yep there is enough. Anyway I'd prefer this for 635nm diodes, but it will work fine for LOCs and Senkats.

Your signature is... it's just true!

@Gillza: The tutorial is unfinished, you better wait for it to be finished if you dont want to get stuck on how to mount the Aixiz+driver into the flashlights haha.
Anyway you can buy the flashlighs, as the take a while to arrive.
Very soon the tutorial will be finished.

 

[Vaporizer]

The 635nm diodes I just got in from Modwerx are rated at 5-7mw. Current draw recommended 45-50ma. So, even though it looks like there isn't much voltage "overhead" available, you aren't pulling much and the battery is rechargeable. I think its a reasonable build.
Many of the <5mw just use a single resistor for current regulation. Setting it for 7mw with a 4.2v fresh charged battery it would do a natural falloff as it approached its min working voltage.
Rafa's example here gives you a good foothold on making an LM1117 driver and PCB suitable for higher power builds later on.

 

[Rafa]

For >200mA in portable builds I'd go for a swithing regulator.

And putting a resistor is.. saying goodbye to the diode.

 

[Vaporizer]

@Rafa,
Exactly, that's why I specified 5-7mw with a resistor for the 635nm I listed , but your PCB & LM1117 for higher builds with higher rated diodes if doing a home build driver.
I'm not sure where the actual cutoff should be, being a noob now here, but the less than 5mw a resistor seems to work. More ma's required than that you better plan on using a current regulator more than a resistor.
I hope that is a better clarification of my response.

 

[Rafa]

Personally I don't recommend LM317, and thats why I used LM1117 in this driver.
Having 1.2 volts being eaten by LM1117, and 2.15V being dropped across the LD, cutoff should be when the battery gets down below 3.3V.

 

[Vaporizer]

Well.I didnt catch that!
The LM317 has the 1.2v ref needed and as far as I can see would still work.
What is needed by the LM1117? I need to look at this........
If it requires less than 1.2v ref...better yet!

 

[gillza]

^Yep there is enough. Anyway I'd prefer this for 635nm diodes, but it will work fine for LOCs and Senkats.

Your signature is... it's just true!

@Gillza: The tutorial is unfinished, you better wait for it to be finished if you dont want to get stuck on how to mount the Aixiz+driver into the flashlights haha.
Anyway you can buy the flashlighs, as the take a while to arrive.
Very soon the tutorial will be finished.

I was actually referring more about the PBC making part of the tutorial But i got to admit that the design of the driver intrigues me as well. It would be good to understand how it functions

 

[Rafa]

Vref has nothing to see about the regulators drop-out, as far as I know.
Drop-out is how much the regulator needs to work, anyway I'm not sure how linear regulators work.

LM317 and LM1117 both have a 1.25Vref, but LM1117 has a 1.1V drop-out, and LM317 has a 2.5V~ dropout. What this means is that we waste less voltage in the regulator, and, we have enough voltage to power the diode, and we have even some "room" for th battery discharge.

 

[Rafa]

I was actually referring more about the PBC making part of the tutorial But i got to admit that the design of the driver intrigues me as well. It would be good to understand how it functions

BTW Gillza, It's just a SMD DDL driver but instead of a LM317 I used a LM1117, because there isn't enough voltage to run from a single Li-Ion battery using LM317