DIY RGB questions.

[bloompyle]

I'm looking into building a DIY RGB projector, and by projector I mean it'll just shoot out a laser like a normal labby, I just want 7 colors from it. I know I could drop $600 on a white fusion thing, but I really like the DIY aspect, and this project intrigues me. SO! Onto the questions.

What's a good ration for 635/532/455? I've been told 1:1:1 is a good start, but just wanted to get perhaps more input on the subject.

Is there any easy way to get 400mW of green for under $150?

Will the standard RGB mirror units from o-like still function well with 635 and 455?

Do I need correction optics for the 635 and 455, or will it work fine with them being multimode and much wider beams compared to 532?

What glasses do you recommend for viewing ~1W of "white"? Also for the cyan, yellow, and magenta, do I need specific glasses for those too? Or just the ones that I use for red, green, and blue?

As you can see, I have a lot to learn, this is something that yes, I want to do, but I am willing to take the time to do it right. Thank you in advance for any and all tips =]

Also, side-question, any tips on knife edging? I'm thinking of making a watt of 635 or 650, and was looking for tips on doing so. Thanks again!

 

[ElektroFreak]

In my experience 3 parts red, 2 parts blue and 1 part green makes a great white color. The individual wavelengths do come into play a little, but red requires the most power, blue is in the middle and green requires the least.

You should correct the 635nm diodes yes, since their beams are too wide for much of anything without at least using a single prism to correct. The 450nm diodes are fine as they are IMO. Of course this would be referring to correction beyond that given by your collimation lenses. Others are more picky and will tell you to correct those too, but I've put mine in front of quite a few sets of eyes and I've never had a single person say the blue beams were too fat.

 

[Sigurthr]

Aye, I'm quite fond of 4 parts 658, 2 parts 450, and 1 part 532. If I had a stable and suitable surface to use as an optical breadboard I'd whip up a nice white with a couple of dichros and cheap modules. IMO 200mW red, 100mW blue, and 50mW green is very doable inexpensively and you get to stick to all single mode/TEM00 outputs.

 

[bloompyle]

Aye, I'm quite fond of 4 parts 658, 2 parts 450, and 1 part 532. If I had a stable and suitable surface to use as an optical breadboard I'd whip up a nice white with a couple of dichros and cheap modules. IMO 200mW red, 100mW blue, and 50mW green is very doable inexpensively and you get to stick to all single mode/TEM00 outputs.

Any figures for 635 use? And how low must you drive the 445 to get such low output from it?

 

[Sigurthr]

There are singlemode 450nm diodes which top out at 150mW, just use one of them. (I don't know a source for them atm, but DTR is planning on stocking them iirc). I've no experience with 635, sorry!

 

[Hemlock_Mike]

Just limit the 445 current to get the desired color. All the diodes are different in the low power range and you will need to turn the pot - so to speak.
The ratios provided are dependant on the dicros you use and in the beam path order.
Enjoy the experimentation !!! That's why this is a hobby. Learn and share.
HMike

My RGB project still needs to be tuned but the colors are good just using the eyeball color method of power adjustment.

 

[bloompyle]

Just limit the 445 current to get the desired color. All the diodes are different in the low power range and you will need to turn the pot - so to speak.
The ratios provided are dependant on the dicros you use and in the beam path order.
Enjoy the experimentation !!! That's why this is a hobby. Learn and share.
HMike

My RGB project still needs to be tuned but the colors are good just using the eyeball color method of power adjustment.

I know how to turn it down, I just know that these are very high outputting diodes, can they even be "pushed" so low?

And yeah, I expected a lot of trial and error, and I was actually looking forward to that part, just wanted to know a good place to start, to hit the ground running to to speak.

 

[Bionic-Badger]

Interesting mix of 3:1:2. For photopic (light adapted) vision, blue should have a much lower lm/W conversion than red, especially at 635nm (~20 vs. ~150). However with scotopic vision (dark adapted), blue has a much higher lm/W conversion versus 635nm (~600 vs. ~4). The L-86 projector I bought has a mix around 2:1:4, but I find them too bluish. I suspect the final mix is somewhat of a mix of the two.

One thing Andy_con mentioned is: it doesn't really matter what your white-point mix is. Having analog control is the key. One reason for this is that you'll be essentially "setting" your white point yourself, and just changing the colors with respect to that. With no other reference point the audience will assume that white is white. I know that the 405nm, 635nm and 532nm laser mixing on this cheapo Reke RGV projector I got as store credit from Dino Direct does look very white at times, even moreso than the other mix with 445nm.

 

[Things]

He is exactly right too. You could tune your laser at a venue for the best looking white, then when you take it home, the white looks wayy off again. Just be generally within the ratio and it really doesn't matter.

 

[bloompyle]

I'm not going to be at a venue, this is just for personal use to have a decent DIY project, and get really cool results from it (4 more colors.) And 3:1:2 is that red:green:blue? I might just use 650 to keep it cheaper, then when I want to make something more powerful I'll pick up 200mW of green somewhere, and the new 500mW 635's, and then I'll worry about correcting it. So for now it's 4:2:1 red:blue:green, and I'll look for those single modes. I've seen them here and there, I think O-like has them, but I'm not sure.

EDIT: Talking about these? They seem to be TO-38 so that'll be an issue :/ Anybody on LPF make 3.8mm modules?
http://www.o-like.com/index.php?main_page=product_info&cPath=9&products_id=148

 

[Bionic-Badger]

The problem with going with 655nm is that 655nm is much less bright to the human eye than 635nm. You'll have to up the power significantly, and with power output topping 300mW for 655nm it may not be bright enough compared to your green and blue. You may consider just investing in the 300mW 635nm red for now ($50), and then you can work from there with whatever else you have. The 445nm blues already have a decent dot (as far as they go) for projectors.

There are some single mode 445nm/635nm diodes, but they're lower power, and more expensive. Other than for pointers, I wouldn't use them, and especially not for a projector.

 

[bloompyle]

The problem with going with 655nm is that 655nm is much less bright to the human eye than 635nm. You'll have to up the power significantly, and with power output topping 300mW for 655nm it may not be bright enough compared to your green and blue. You may consider just investing in the 300mW 635nm red for now ($50), and then you can work from there with whatever else you have. The 445nm blues already have a decent dot (as far as they go) for projectors.

There are some single mode 445nm/635nm diodes, but they're lower power, and more expensive. Other than for pointers, I wouldn't use them, and especially not for a projector.

Right, that's why I was originally planning for 635, but then the correction optics and such was iffy for me because I'm very new to lenses. My first laser was an LPC-815 =] I'll see what I can do for the 445's, I've noticed the single modes are about $80 a pop, AND TO-38 so I'd need to purchase a specially machined housing too, so those diodes would drop me about $100, whereas correction optics, maybe around $25, so we'll see =] Also, perhaps the word "projector" was a little misleading, the most projecting this will do is perhaps a liquid sky affect in my basement haha, this won't be for shows, I won't get any software to make images and stuff, just a DIY white laser if you will. However, that being said, I do have a question. Is there any cheap method to "split" the colors? Like this:

 

[ElektroFreak]

No unfortunately the only way to get an effect like that is to use galvo scanners and modulate the laser sources, which means you need a complete RGB show projector, but something similar can also be done with a spirograph setup instead of galvos. You can use diffraction gratings and mirrors to split a white beam into it's red green and blue parts, but each color will be confined to a single beam not a sheet effect like in your pic.

 

[bloompyle]

No unfortunately the only way to get an effect like that is to use galvo scanners and modulate the laser sources, which means you need a complete RGB show projector, but something similar can also be done with a spirograph setup instead of galvos. You can use diffraction gratings and mirrors to split a white beam into it's red green and blue parts, but each color will be confined to a single beam not a sheet effect like in your pic.

Galvos are the spinning mirrors in spirographs right? How does one modulate it, and how much does that cost?

 

[ElektroFreak]

Galvos are slightly more advanced than spirograph mirrors, true galvos are capable of full images and animations whereas a spirograph uses mirrors mounted to rotating motors, each being tilted slightly so that as they rotate they project a circle. Combining multiple motor/mirror assemblies allows for complex spiral patterns.

The laser drivers need to be capable of being modulated. Most hobbyists use the FlexMod from drlava (buy it through Flaminpyro here though as drlava isn't really up for business at this time). When you build your modules, you would use Flexmods instead of the usual Flexdrive or DDL (LM317) drivers. Then you can apply a modulation signal to each laser so as to produce different colors when combining them.

 

[Bionic-Badger]

Right, that's why I was originally planning for 635, but then the correction optics and such was iffy for me because I'm very new to lenses. My first laser was an LPC-815 =] I'll see what I can do for the 445's, I've noticed the single modes are about $80 a pop, AND TO-38 so I'd need to purchase a specially machined housing too, so those diodes would drop me about $100, whereas correction optics, maybe around $25, so we'll see =] Also, perhaps the word "projector" was a little misleading, the most projecting this will do is perhaps a liquid sky affect in my basement haha, this won't be for shows, I won't get any software to make images and stuff, just a DIY white laser if you will. However, that being said, I do have a question. Is there any cheap method to "split" the colors? Like this:

Don't worry about the correction too much. Yes, it looks like a bar, but you'll be moving the beam about quite a bit right? Also the size of the beam only become prominent at large distances. It doesn't have to be perfect, and you can even focus it to look somewhat like a dot at a particular distance.

For correction, you can also buy some cheap correction optics to help tame that laser. Also those optics are very useful for projector systems and such anyway, so think of the correction optics as a bonus!

For that laser effect, you might be able to achieve that with an RGB spirograph, timing out the TTL signals correctly along with your motor spin. It may help to have some positional feedback from the motors, like using a CDROM motor. I should look into making something like that as an experiment.