P3 Flexmod

[buffo]

that doesn't work while shows are playing ;(

It does if you're running a decent controller. The shutter should close any time you don't have output, and not just at the end of the show.

Back before I switched to the GM-20, I used a stepper motor from one-stop-laser-shop (their STP-8) as a shutter in one of my projectors. It make an audible "thunk" ever time it switched. (GM-20's are silent) When playing a show you could hear that think clunking open and closed multiple times during the show playback. (The annoying sound of the STP-8 is the main reason I upgraded to the GM-20 in the first place.)

All of my diode reds emit a faint glow when they are below threshold, and so do my 445 blues. But you can't see that on the wall because I have a shutter. (Well, with one exception - I do have one projector that doesn't have a shutter installed yet, and on that one you can sometimes see a faint spot now and then.)

Adam

[kecked]

All I was saying is that the balance pot might need adjustment. The rest is interesting but totally unrelated to what I was saying. My issue was that with zero bias I still had a glow. That should not be. Yes of course we want to set a bias for thresholds. If you don't you really mess up color and modulation. Beam suppression should be real easy with diodes since they can modulate in the MHz range.

[swamidog]

let me rephrase...

i'd have to use one for each diode and that would be a huge drag. i don't want any glow unless a diode is supposed to be on. remember, i also use my projectors for photography projects and diode glow is a a spoiler.

It does if you're running a decent controller. The shutter should close any time you don't have output, and not just at the end of the show.

Back before I switched to the GM-20, I used a stepper motor from one-stop-laser-shop (their STP-8) as a shutter in one of my projectors. It make an audible "thunk" ever time it switched. (GM-20's are silent) When playing a show you could hear that think clunking open and closed multiple times during the show playback. (The annoying sound of the STP-8 is the main reason I upgraded to the GM-20 in the first place.)

All of my diode reds emit a faint glow when they are below threshold, and so do my 445 blues. But you can't see that on the wall because I have a shutter. (Well, with one exception - I do have one projector that doesn't have a shutter installed yet, and on that one you can sometimes see a faint spot now and then.)

Adam

[The_Doctor]

All I was saying is that the balance pot might need adjustment. The rest is interesting but totally unrelated to what I was saying. My issue was that with zero bias I still had a glow. That should not be. Yes of course we want to set a bias for thresholds. If you don't you really mess up color and modulation. Beam suppression should be real easy with diodes since they can modulate in the MHz range.

What is the threshold current and maximum rated current for that green diode? With reds there's a ratio that is pretty much similar regardless of power, but I've wondered if greens differ a lot, as they already do in Vf. Specifically, is the threshold unusually low for a diode of its power, compared to red (or blue)?

Also, does your Flexmod have beam suppression? If not, like many other current sense mechanisms, if the beam is to be totally extinguished by zero current, then the sense mechanism may sense nothing and try to push current until there IS something to sense! Many systems depend on this tiny residual current just to have stable function, and if your green diode has a very low threshold, it may be visible even at this low drive.

What I said just there relates to theory, rather than Flexmod specifics, but I doubt the Flexmod is all that far removed from theory of current control.

[buffo]

i also use my projectors for photography projects and diode glow is a a spoiler

You're saying you have enough glow that with one of the other lasers on, you can see the color shift because the one that is supposed to be off is glowing slightly, and this throws off your photography?

Either you've got the most color-sensitive camera on the planet, or you're doing something wrong. Because as long as at least one laser is on, the brightness of the "glow" from the idling one(s) should be many orders of magnitude dimmer and thus should have no effect. Add to that the fact that the scanners are moving the beam around, and I find it nearly impossible to believe that you can perceive the color shift.

I think what you are describing is a case when all the lasers are supposed to be off and you still see some faint glow. And as long as the mirrors aren't moving, I agree that this faint glow is noticeable, even with your eyes. (With the mirrors in motion, it should be far too dim to see with the human eye, and even a camera would have a hard time picking it up.) Nonetheless, this is exactly the case where a shutter will solve the problem, PROVIDED YOUR SOFTWARE CLOSES THE SHUTTER WHEN IT SHOULD.

Adam

[The_Doctor]

I agree with Adam on the photo thing, at least in the context I've done it. It's really easy to stretch a histogram or change gamma, or even just brightness and contrast to drive the idle glow below the noise floor, right to zero. It needs a bit of care if you want to preserve dynamic range and linearity, but it works. Tuning out by eye is the tougher thing to manage, which is why I mentioned various ways to cue the brain into ignoring it. (The way I saw it, is this is similar to various problems in audio compression and noise reduction, and a visual equivalent of psychoacoustics might work well where more obvious ideas fail).

If some hardcore practical scheme is needed for each diode, at high speed, under full operator control, I think my scheme in post 18 of this thread could cover it very efficiently.

EDIT: I totally second what Adam just said about scanning the blanked beam, a slow wide scan, a big circle perhaps, as a 'parking orbit' should render any blanked-light beam invisible by eye. Only a long exposure camera shot could pick it up easily.

[swamidog]

i think you're underestimating the amount of glow that's present after you adjust the drivers so all the diodes have the same low end characteristics.

granted i don't do things in the traditional way.. but (points to 8" cubed laser projector on coffee table) this little satellite of love has three single mode diodes and three flexmods in it. i use this projector for programming and watching laser shows (at ~300mW it's great my living room), watching ilda wave shows, and art projects.

if i was just using this projector for LSX shows, i really wouldn't care about bias setting on the flexmods, i would do it all in software and that would solve the glow problem. however, that's not really practical if you're watching an ilda wave show. if you adjust the bias settings for best performance for a sound card dac (in my case a modified echo layla), you will have a surprising amount of diode glow.

maybe i'm being too picky, but i spend a *lot* of time working to reduce (at least greatly minimize) color shift during fades and cycles. if this was a big, bright projector the glow wouldn't be an issue.
this absolutely enough to be visible in my photo projects. especially the green diode.

You're saying you have enough glow that with one of the other lasers on, you can see the color shift because the one that is supposed to be off is glowing slightly, and this throws off your photography?

Either you've got the most color-sensitive camera on the planet, or you're doing something wrong. Because as long as at least one laser is on, the brightness of the "glow" from the idling one(s) should be many orders of magnitude dimmer and thus should have no effect. Add to that the fact that the scanners are moving the beam around, and I find it nearly impossible to believe that you can perceive the color shift.

I think what you are describing is a case when all the lasers are supposed to be off and you still see some faint glow. And as long as the mirrors aren't moving, I agree that this faint glow is noticeable, even with your eyes. (With the mirrors in motion, it should be far too dim to see with the human eye, and even a camera would have a hard time picking it up.) Nonetheless, this is exactly the case where a shutter will solve the problem, PROVIDED YOUR SOFTWARE CLOSES THE SHUTTER WHEN IT SHOULD.

Adam

[SaltyRobot]

Swami I know what you are talking about with the glow. I recently took my chi brick and switched the red and blue diodes to flexmods. Now I am experiencing the glow you are talking about. Especially annoying when I power on because I am letting light out of the projector before the computer is even linked to the Etherdream. But I have also noticed it in the middle of shows too when I have a green pattern scanning and the red and blue are supposed to be blanked. The Red and Blue will emit enough light that I can see the small point in the center of the screen. I run my projector with full lights on in my basement while watching TV. My 400mW projector is bright enough that I don't need to turn off any lights, but I can still see the diode emitting when they are technically supposed to be off. Maybe I am doing things wrong as well.

Adam would the GM-20 correct this type of situation? Or would it only be useful when all three diodes are supposed to be off. I guess what I am asking is how does the GM-20 allow a pattern of one color while diverting the other two colors. Not possible right?

[The_Doctor]

...if you adjust the bias settings for best performance for a sound card dac (in my case a modified echo layla), you will have a surprising amount of diode glow.

In this case, any chance you could run your digital signals through noise gates first to increase the low-level difference to improve contrast?

[buffo]

If you are seeing a spot in the middle of the scan field, that means the galvos aren't drawing anything at that moment. In that case, a shutter will help - assuming that the software you are using is smart enough to close the shutter in that case like it's supposed to. (Pangolin does this, but not all software does.)

The reason you see the rest of the pattern (away from the center spot) in the other color(s) is due to persistence of vision. But if the spot is in the center of the scan field, the scanners have to be at their neutral position, and this is when the shutter would normally be closed - even if only for a split second.

Now, if you see color banding on the lines as they are being drawn (that is, the line is supposed to be green, but you see a hint of red that adds a yellow hue to it), then no, a GM-20 won't help you there. But if that's your problem, then you probably have the threshold current set too high on the drivers. (I've never seen this myself.)

Likewise, if you see color on a blanking trace line (that is, when the lasers are off while the scanners are moving to a new location, you see faint "trails"), then that is another situation where a shutter won't help, and in that case again it's because you don't have your driver(s) adjusted correctly.

I never see the spots you're talking about in my projectors with GM-20 shutters, nor do I have any visible blanking trails on any of them - even the one without a shutter. This is even on my low-power RGB's, which run from ~ 400 mw to 600 mw.

However, on the one RGB projector that does not have a shutter (roughly 1.2 watts), I do have the spot in the center that you're talking about. And yeah, a shutter will fix that, I just need to get around to installing it.

Adam