RGB LASER / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

RGB LASER

1 2 3 4 next

Move Thread

LAN_403

...

Sun Mar 25 2007, 10:57PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

EDIT- scroll down to find revision 3 built and working!

For my 1000th post...

I was looking at my available laser parts, and decided to to make a red+green laser, and hopefully add in a 405nm blue diode from a blu-ray/hd-dvd player sometime to make it a full color red+green+blue laser.

In any case, the first step was to package the red diode, since it was just a bare diode.
A little more looking turned up a little heatsink, and some thermally conductive epoxy, which led to this

Then, I dug up a collimator that was given to me in a different laser order and found that it needed to be held about .2" away from the heatsink, and held there to within ~+/-.002". A little more head scratching turned up some nylon washers, so I glued one to each side of the assembly. Then I took some 300grit sand paper, and slowly ground down one of the washers until the beam was colminated. I forgot to take a pic of the finished laser, but you can see it pretty clearly in the latter pics.

The second step was to get some green, and after a few quick decisions it was determined that I needed about 3mw of 532nm. The huge difference in power (80mw red vs 3mw green) is because the 650nm diode I used is actually almost IR rather than red. This is good because it gives a very deep/rich red, but bad because you need a ton of power to get a decent perceived brightness. Then, I though about using a Nd:YVO4/KTP crystal and a 1w 808nm laser diode (a piece of one of my 25w laser diode bars) to DIY the green laser, but I decided that since I needed soo little power I would just use the guts of a green pointer that I had handy, which really simplified things. To had diy'ed it I would have had to do the alignment for the diode, the crystal, and a 5x or preferably 10x beam expander to get a mildly collimated beam, and it would have ended up costing way more than the $30 for a complete module with an integrated sense photodiode and even a driver board (but I scraped that since I need analog modulation for later)

Then I needed to dicrotic mirror that would reflect as much of the red beam, and as little of the green beam as possible, and I found a turning mirror out of a dvd-rom to do the job quite well. (it passes about 5% of the red, and 80% of the green at 45degrees IIRC).

The final step (well, until I cough up the $$ for a blue diode) was to mount the optics in a stable package. Now, this would be impossibly hard for many people, as to get the beams to look collinear at and distance you need to keep the beams within +/-.2mRAD (.01degrees), and aligned to within +/-.01" in x/y. Luckily I happened to have a ton of nice optical stages around, so I just made up a little 5 axis aligner, and did the alignment.

To make sure the beams were perfectly collinear I put a little bounce mirror at the far side of the lab (aka, bedroom) and send the beam back toads me. Then, I used a squirtbottle of water set for a fine mist to create a dense fog to see the beams. From there, I could easily get them aligned to within .01" at 20'.

But then I ran into trouble, keeping the stuff in place. The 1.0 revision was done with a ton of CA glue, but the combiner mirror was up on a big block of plastic and likewise was really loose (gently pushing on it would send the beams out of alignment like .5")

So I decided that if I was going to pull it all apart I would make room for the blue laser, and neaten up the beams a little, so I cut a piece of aluminum heatsink and used that for a base. To attach the optics I used some UV cure epoxy, and all was going well until I tried the alignment. First, I tried with the UV epoxy, I got everything aligned, hit it with 10 seconds of UV and then noticed that the beam had walked a solid .5" from where it started, because as the epoxy cured it expanded slightly. So I decided to try some normal 5 minute epoxy. I mixed up a batch, dribbled some on the 2 parts I was aligning (since I was using a 90degree turning mirror for the green I had to align both the mirror and the laser at the same time) and dialed in the alignment. But, then as I waited for the epoxy to cure I noticed the beam slowly walking away (over an hour it would drift about .1"). So I tried touching the parts I had already mounted, and of course even the slightest pressure sent the beams out of alignment, and they were slowly drifting around.

So now I am on the hunt for some decent UV cure epoxy that doesn't expand/shrink much, and most importantly, cures 100% solid...

In any case, it did stay aligned long enough to try to take a few pics of the beam, although laser beams ae hard beasts to photograph.

edit:
Blue laser diode acquired. Well, technically it is way closer to UV at 405nm, which is roughly as bright as far as lumens/watt as 730nm near IR light, at about 1/2,000 as visible as 555nm in bright light. However, when your eyes are dark adjusted they are more sensitive to UV, and it becomes about as bright as 620nm light, which is quite visible at about 1/150 as bright as 500nm. Also, keep in mind that eyes have a nonlinear response curve to brightness, so in reality it takes about 5x as much 405nm to be as bright as a 532nm yag laser.
However, the fact that it is UV works out great, as so long as you use a fluorescing target (white paper, or dust) it fluoresces at that electric blue that bleached paper does under a blacklight

In any case, I hooked him up with a culminating lens, and now I have a beautiful beam running around. Admittedly, there is a TON of scatter (I am not sure exactly why, I think it could be related to using a very short focal length lens for collimation (like .2"), or the fact that I used a plastic lense coated for who knows what with a UV laser diode, but in any case there is more power going into the scatter than the beam

Bit of trivia, 405nm laser diodes (at least the one I have) are actually glass clear! Unlike the 'normal' GaAs laser diodes which are pitch black, you can see right through this guy. I am not sure if the laser is actually grown in the clear substrate, more likely it was grown on top, however it was quite a surprise when I took a look at him under the microscope and was met with a water clear laser diode. I just wish I could see a full wafer of those!
peter

...

Sat Apr 14 2007, 09:27AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

wehe! now I can say that I have built a white light laser

In any case, turns out the problem with the 405nm diode was that there was a tiny spot of glue right centered on the window of the package which was screwing with my beam, so after removing that I had a decent culminated UV laser.

That is the laser I carefully removed (see the pr0n

), I just glued a Fl 0.2" lens on iit, much like I did for the red one.

Then it was a 'simple' matter of gluing everything down to a plate of aluminum. I will spare you the details, and leave it at that I still need to rip out the blue diode and redo the alignment for their red before I can call this finished...

That got me here:

the optics assembly

without a flash

all lasers running (powers adjusted to look white, but the camera didn't pick the green as it should, and with just the blue/green lasers

just blue and red/blue

red/green as just green

just red, and the blue/green lasers powered up with a little water vapor

ragnar

Sat Apr 14 2007, 10:34AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

UNbelievable, Peter! This is outstanding.
They look quite well aligned -- collimation is quite good?

Electroholic

Sat Apr 14 2007, 10:56AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

holy shit thats quite a setup you got there.
so what kind of glue did you use at last?

...

Sat Apr 14 2007, 05:21PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I ended up using a bottle of 'gap filling' (really thick) CA (superglue) and some accelerator (some really nasty stuff that you spray on it that makes it cure at like 100x). That stuff is the most brittle that I could find, but it still is too flexible.

I am aiming to keep the beams collinear, so I have to keep them aligned better than the divergence of the beams (about .8mrad or so), which I don't think is possible the way I have the mirrors mounted.... As it is, gently pushing on them can walk the beam about 5mrad

My plan now is to pile a ton of glue on the lasers (you see, the lasers are round, and I don't have the means to machine a matching piece of aluminum to mount them to, so I am using a big glob of glue, which sorta works but I can still walk the beam about 3mrad with mild pressure on them) and then add some aluminum plates on the sides of the mirror to help stabilize them.

I am also worried about my uv laser, as it has way more divergence than the other ones, so at long distances there is a yellow spot surrounded by a purple glow more than there is a white spot. But even more worrisome is that the beam is most definitely not a Gaussian profile, more of an eye shape with a big vertical plume. I am thinking of ditching my homebrew collimator and putting it in a small commercial one that had a red diode in it, and seeing if that helps.

Oh well, it works and the proof of concept is there. Now I just need to get it boxed up and with a decent beam

Sam

Sat Apr 14 2007, 06:24PM

Registered Member #227 Joined: Mon Feb 20 2006, 10:47PM
Location: Cambridge Ontario, Canada!!
Posts: 127

machine a 1/2" hole or what ever the size of the laser body is into a large piece of maybe 1" by 1" Al. the use set screws though the block to hold/line up the lasers.

my $0.02 CND

Conundrum

Sat Apr 14 2007, 08:22PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4062

Low melting point (59C) alloy works well. If you have some indium you can DIY the formula.

Low 136:- 49% bismuth, 18% lead, 12% tin, 21% indium
(or the appropriate quantity of 60/40 with added tin to achieve correct ratios)
Doesen't change shape much as it cools, in fact this is the main selling point of that particular alloy mixture.

Hope this helps.
-A
Who bought $100 worth of indium for making chip removing alloy and still hasn't got around to actually making some more!

...

Sat Apr 14 2007, 11:57PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

hmm, never thought to solder the lasers down...

I have a ton (ok, like 10g

) of indium, some of the low melting point alloy you were selling on the old 4hv, and a few different types of indium solder, (from my laser diode mounting projects) and all sorts of flux, I bet I can make something work.
Although the red diode is made of aluminum (as opposed to brass for the other 2 atm) so I might run into trouble there...

I put the uv diode in a 'real' colminator, and the beam is a LOT better. So that solves one of my problems. Now to get that mirror alignment problem done....

...

Tue Apr 17 2007, 06:07AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

Ok, the soldering the lasers down looks like it is going to work well

The first problem I had was that the red laser was solid aluminum, so I had to find a way to attach that without soldering to it. I started by grinding a flat onto it with a dremel tool, which sorta worked, but then I was thinking that it would just be way easier to find a collimator made out of brass. So I started looking, found an old 5mw red laser diode module that had a nice small brass collimator. So I stuck the red diode in, and I was pretty happy.

But then I started thinking about how crappy the blue beam was (let me tell you, it was pretty bad) and figured that just about any collimator would be better than the lens I was using. So I kept searching, and found an old laser pointer, but after dremeling it apart I found out that it was sooo old that is 1mw 650mn diode was in a 9mm can, instead of the 5.6mm one I had. So then I kept looking, and found an old laser level that was soo dim that it was useless (well, when you are only allowed 5mw of power, even with a 630mn diode you aren't going to make very much of a line regardless of how lossy the optics are) so I cracked it open. And to my surprise, there was not 1 laser in there, but 2 identical brass collminators

So in an attempt to try to make the laser look like it wasn't built by a 16 y/o, I decided to use the matching collminatos. So I pulled them apart so that I just had the brass outer part, broke out Andre's solder that has been sitting in a pile for the last 2 years and tinned up a nice big blob on them. If you are insane enough to try this at home, make sure you use flux, lots and lots of flux, and that there are no plastic optics in there...

Then, I needed something to solder them to, so I took a hint from the The first lasers I soldered down and smashed up some burned out to-247 fets (no shortage of those here

) and dremeld them square.

Then it was a simple matter of applying some heat, and I ended up not even using a hotplate and just using a 50w solder iron (temp controlled weller) to do the dirty work.

So now I have a well collimated beam, and it has a nice fat plate to glue down (but needs some spacers made up to get them all the same height).

Then there was the problem of the optics, and how I don't have a decent way to attach them to anything. So thinking back to the lightwave 142 laser I got 2nd hand (some moron ripped the highly customised yag and ktp crystals out of it

) I remembered that they were glued to a glass post, instead of how I had them attached. Figuring that I have a bunch of glass blocks on hand, I decided to go that route.

So here is the new proposed beam path:

I also added in a power sense diode for the red beam, as the one I am using doesn't have one for some odd reason...

Couple of updates (but I hate to make a triple post...)
1. The glass blocks work great! I glued up the red/uv diodes and even with mild pressure the beams stayed aligned

A quick back of the napkin calculation says that I have locked the mirrors down to about 2microns

2. I pulled the plate off the mounting setup and the beam moved about about a mrad (I am trying to keep them aligned to 1/4mrad). Pissed off, I decided to apply a little pressure to the plate, and found that even with a few ounces of prssure I could walk the beam .5mrad.

So, now I am going to have a plate made out of 1/4" thick Al (compared to the .064" pate with 1/4" fins on it) and hope that solves it.

In the mean time, it is time to make some drawings...

Dalus

Wed Apr 25 2007, 08:21PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Looks quite good. What kind of galvo's are you using to steer the beam around. Just two motors with a mirror mounted at a angle?

1 2 3 4 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob