Doing a little work on my RGB...
... so of course I took some photos.
http://picasaweb.google.com/closeeno...icalThreshold#
I need more red. This thing started out as a Neo-Laser NLSP250RGB. Then the blue failed (horrible thing) so I decided to swap it out for one of Dragon Lasers 200mW blue modules. That got me far improved performance. These projectors have a 635 nm red with horrible beam geometry so I think that's going to be the next thing I switch.
Other things I've done to it include fitting new galvos; LaserShowParts sold me a nice set of ScanPros which I fitted. I found that for some reason the X-galvo tends to fail, possibly because we tend to run wide for eight hours at a time. As a consequence I've fitted a mixed pair now! The lovely accurate and fast Y galvo is a ScanPro and the sloppy but wide-angle X-galvo is a cheap Chinese thing.
I also modified the 5v PSU to make it cleaner, since that was crashing the FB3, which due to problems with the FB3's USB driver was crashing the Mac I ran it off when I tried to restart it. (I also had a problem restarting the FB3 while I was working on it, but reformatting and reinstalling the FB3's memory card fixed that- I guess the FB3's firmware is at least partly stored on the card?) Replacing the blue also helped that a lot. I guess the old blue was not as well decoupled as the Dragon Lasers one.
The optical table in these units is not very well designed; it's clearly designed to be as small as possible, with dichros in aluminium castings screwed and epoxied to the plate. It also uses a dichro even for the red, which does not need a dichro- an HR would be better. Weird. The optical table is mounted in a metal box, which helps to keep things clean, but it has an acrylic output window which is not very optically pure. I have removed it on mine (with pliers and great care, since it is close to the pass-blue dichro) and obtained a good improvement in collimation performance.