Flicker

[Bradfo69]

Now that I have a (couple) decent projectors, I find I'm becoming a little more critical of output and studying things a lot more closely. Is flicker basically a problem inherent to the quality of galvos? When watching graphics shows, obviously some of the more complex things are displayed with flickering of the image. For example, in "Recipe For Love", when the cookbook floats across and the word "Recipe" is displayed, it flickers. Now, I'm using DT40's (both pros and wides). If I bumped up to Eye Magics or Cambridge, would I reasonably expect to see that flicker disappear? I know it's that whole 5% improvement for a bazillion dollars thing and nobody but laserists can appreciate it but, the itch is starting!

[allthatwhichis]

I have to admit I even get flicker on my CT6215s in my Kvant; they are the 30k set. I think it is more the speed of the galvos over the quality. If you were to go to the 60k CTs you would get a good deal less flicker.

[flecom]

the speed of the galvos over the quality

correct

a set of 30k galvos drawing a complicated image will have noticeably more flicker than a set of 60k scanners for example

its all about speed

[Bradfo69]

I'm sure it is...at least in most things!

60k CT's. Humm... I'll get right on that! Maybe next year. I need to replenish my accounts from SELEM first!

[polishedball]

I myself am waiting to see what Pangolin delivers sounded like it will be very nice, as I recall it was given a christmas to market date, but think i heard that for QS working on the QM last year about the same time.

I'm sure it is...at least in most things!

60k CT's. Humm... I'll get right on that! Maybe next year. I need to replenish my accounts from SELEM first!

[Jem]

Hmmm... I have high speed CT6215h's in my projector and still get flicker on some images. Problem is I don't have any other scanners to compare them with so it may be that the flicker is much less than with standard galvo's.

However, I have had my CT's set for 30k scanning within the LD2000 software and watched a show. I have then upped the max scan rate to 60k (in LD2000 software) and re-watched the show... not that much difference to be honest. I think it may be something to do with what scan rate the show was originally designed for that is more relevant.

One thing I can say though is that you can increase the scan angle on the CT's and not overdrive them, also complex abstracts look magnificent

Cheers

Jem

[araugh]

Now, I'm using DT40's (both pros and wides). If I bumped up to Eye Magics or Cambridge, would I reasonably expect to see that flicker disappear?

Like the other guys said, speed is the key, but remember that galvo speed is proportional to scan angle. When you increase your scan angle you decrease the speed at which your scanners can move because now they have to swing over a wider range. 60k's may do 60k great at 8 degrees but don't try to push them that hard at 45 degrees or you'll end up with some problems.

[masterpj]

To get rid of this flicker best is by increasing your projection distance a lot.
it will get rid of all that flicker..well to a certain range that is.... for example that part of the cookbook will not flicker if you scan from a pretty big distance.
faster scanners do work though.

[heroic]

Let's correct some misconceptions.

1. Galvos have a maximum speed.

In fact, galvos don't have a maximum speed, except in so far as they have a maximum deflection, and a maximum acceleration; therefore their maximum speed is limited. What slows a galvo down is that it has mass, the coils that drive them have a current limit, and therefore a maximum flux density, and therefore a maximum acceleration. The scanner amplifiers contain a control loop that makes the galvo seek towards a position defined by the input voltage, using a position detector for feedback. If the mirror doesn't make it to the input position by the time the voltage changes, you get scanner undershoot; this is actually mandated in the ILDA projector standard, and is the purpose of the circle- it should appear symmetrical, circular and entirely inside the box when the scanners are being run at their proper angle and speed, even though the actual points in the frame are outside the box.

Practical upshot: simple frames, small angles, and frames composed mostly of curves, can be scanned more quickly since the available acceleration is limited.

2. Eyes work like cameras

Actually, no. The human eye is very different to a camera- a camera has a shutter (and, in fact, CCD video cameras have an electronic shutter) that is open only for a very short period of time. Each frame of video only shows that portion of a laser trace pattern that occurs during the interval when the shutter is open. If the scene is dimmer, or the aperture is smaller, the automatic exposure control in the camera will keep the shutter open longer. The video will then show less flicker.

On the other hand, the eye is photosensitive on a continuous basis, and the relationship between persistence is a function of intensity and wavelength. Brighter projections appear less flickery. For photopic vision- the usual kind- it's necessary to scan an entire frame within approximately one twentieth of a second for the frame to appear continuous. For scotopic vision, the interval is somewhat shorter, but it depends on the person.

3. Point rate is all you need for flicker fixing

There are actually three different quantities that relate to how fast you can scan things- the physical rate at which the scanners can move, the rate at which the DAC streams samples to the scanners (point rate), and the rate at which the point stream changes (frame rate). On top of that, there are the parameters of the device you're using to view it (eyes or camera).

All in all, it's a complicated mess that could be fixed by sorting out the appalling 1980s-era hack that is ILDA projector standards. For example, if the DAC could read the galvo position detector output, and if it took an idealized version of the frame then it could optimize it intelligently... I mean, font renderers have been doing that since the late 70s.

More practically, you can reduce your flicker by setting your video camera to 24 frame mode, small aperture if you have it, and by reducing the scan angle. Simplifying the frame will help, and faster scanners will help too.

[j4cbo]

More practically, you can reduce your flicker by setting your video camera to 24 frame mode, small aperture if you have it, and by reducing the scan angle. Simplifying the frame will help, and faster scanners will help too.

To clarify -
If the issue is flicker in video recordings, that will help.

If the issue is flicker as seen by the eye, the only thing you can do is increase your frame rate. "Flicker" is "the frame rate is too low". In order to be able to turn up the frame rate without problems, you may need to simplify the data, get better galvos and turn up the point rate, etc - but the frame rate is what you see.