After almost getting my first full color RGB laser working, I decided it was time to get serious about making a stable chassis to mount the optics to in the next revision of the laser. It took a few years before I got access to a machine shop to do the mechanical work in (thanks UCSB COE!), but at long last I have made a working RGB laser (or RGV if you will, blue laser diodes are still too expensive for my taste). For more information on the creation process of this and later revisions, check out the forum post I made on 4hv.org.
This overall design of this laser is exactly the same as the previous revision, but it is machined out of a block of aluminum so that the baseplate wouldn't flex anymore.
Here are are some renders of the model:
The overall laser and with the top transparent:
More cutaway views:
You can download the models that I am using, as well as drawings for this case and a similar one made out of 4 pieces instead of 2 below. The SolidWorks model has drawings for many of the parts, and an alternate case design, but requires at least SolidWorks 2006, or something capable of opening .sldpt/.sldasm/.slddrw files. If you don't have anything that can do that use the IGES model, which can be opened with just about anything
For the blue laser I decided to use a 'PHR-803t' 405nm diode (named after the sled that it can commonly be found in), which is good for 100-150mW of power, and can be had for under $20. For the red laser I decided to use a SLD1239JL 650nm diode, which is good for about 150mW before it starts to overheat. For the green laser I used a 50mw handheld laser pointer I bought from the Far East for $25, which was outputting about 65mw when I got it with 280ma of pump current (impressive!, although that did include a considerable IR component because the modules are not filtered). I used Aixiz collimator modules for the red/violet diodes because they are cheep (about $5), although they have a very large beam size (~5mm) which does not match the ~2mm beam size for the green module resulting in suboptimal color mixing. Additionally, there is about a 30% power loss with the 405nm laser diode in the acrylic lens, but this was worth it to not need to buy a 'real' collimator. (however if I rebuild the laser again i would at least find a shorter focal length lens to get a 2mm beam to match the green module).
I started out by machining the case in our machine shop out of some scrap aluminum:
I then bolted the case down to an optical bench, and used some rigjob aligners to align all of 6 optics to get the 3 beams coincident. Sadly, the UV cure epoxy I used to hold everything together was quite a ways past its expiration date, and when cured was about the consistency of old chewing gum--so the beams haven't stayed in perfect alignment (but are fairly close).
The end result:
A shot with the lasers running, and with the power supply I made for it (a simple lm317 adjustable constant current source):
And finally a picture of the laser burning some black plastic (just because) amnd what the beam looks like when reflected off of a DVD (note however that the camera used to take that picture is not very sensitive to UV, so the first order reflection looks yellow instead of white, and the violet beam is very dim).
This laser head ended up being used for a project class at UCSB, ECE94R, in a RGB Laser Projector, and has been working great. However, I have had issues with the beams walking out of alignment (due to expired UV cure glue), low output power (due to using salvaged dichroic mirrors out of DVD and HD-DVD drives), and unstable green power (due to using a cheep/low quality green laser module).
I can be contacted at firstname.lastname@example.org