rotating mirror system for shifting a beam?

[wosoka]

I am looking for some mirror setup where rotation one or more elements in the mirror setup will allow to shift the beam along its axis.

Two mirrors in a galvanometer scanner type setup work well for controlling the angle of the beam.

Rotating risley prisms can be used to only offset the beam without changing its angle but the possible positions are limited.

Dove prisms are great for rotating the beam along itself or rotate in a circular fashion if prism center is shifted.

But as for offsetting (shifting) a beam along its axis from a (-1,-1) position to (1,1) without changing the angle of the beam, I can't find any setup which involves a rotating mirror rather than shifting mirror. But I feel there may be something missing from my limited optics knowledge.

[Frank]

Hi Wosoka,

Maybe combine with a parabolic mirror to make the beam parallel once its angle is changed?

 

[wosoka]

Hi, that will introduce distortion to the beam. ...right?

Edited March 29, 2018 by wosoka

[Frank]

1 hour ago, wosoka said:

Hi, that will introduce distortion to the beam. ...right?

Yes, I would expect so, as do lenses, but perhaps minor enough for the application (which we know nothing about).

[wosoka]

Oh where do I even begin.
I can say this is for pico video projection beam and I've tried laser scanners and while they are cheap (the "slow" ones) and easy to control, the distortion of the beam caused by angles of the scanner mirrors is hard to correct by undistorting the pixels in software and results in loss of resolution. Something which would be perfect would be to be able to offset the tiny projection beam without affecting its angle. I had some luck with a gantry system and stepper motors as well like in the CNCs but it was just not fast enough, although close.
 

9 hours ago, Frank said:

Yes, I would expect so, as do lenses, but perhaps minor enough for the application (which we know nothing about).

Well, the barrel or pincusion distortion may be corrected by a corresponding distortion lens before the scanners. But with this solution you seem to be not addressing the distortion of the beam which has bounced off of two scanning mirrors at an angle, I think.

Edited March 30, 2018 by wosoka

[swansont]

If you want to shift (displace parallel to original) the beam, you need two mirrors, and they are in a fixed location the displacement is limited to the size of the mirror. If it's steering, you are limited by how far you can tilt the mirror — automated systems I'm familiar with don't typically have a large "throw" for their actuator. I think laser light show systems (and laser printers) use acousto-optic modulators since they're faster, but the angle difference over a scan is still somewhat small (several milliradians)

[wosoka]

You can't shift in two axis with just two rotating mirrors, you'll need four and the mirrors for the last two would have to get pretty big, for a diverging beam too big to be practical.

Laser light show projectors all exclusively use galvo scanners, I've built some myself.

For video projectors there's this but shifting amount is very limited and costs thousands. https://www.optotune.com/products/beam-shifting

Edited March 30, 2018 by wosoka

[swansont]

1 hour ago, wosoka said:

You can't shift in two axis with just two rotating mirrors, you'll need four and the mirrors for the last two would have to get pretty big, for a diverging beam too big to be practical.

That's interesting, because somehow we do that in our lab with two.

 

[wosoka]

4 hours ago, swansont said:

That's interesting, because somehow we do that in our lab with two.

 

Do what? What is the mirror setup? With two rotating mirrors you can either rotate the beam in 2D or shift the beam in 1D. But shifting the beam in 2D with 2 mirrors? That's new to me, if this is what you do then please explain.

Or do you mean a mirror attached to a device that can rotate in two axis rather than one?

Edited March 30, 2018 by wosoka

[swansont]

54 minutes ago, wosoka said:

Do what? What is the mirror setup? With two rotating mirrors you can either rotate the beam in 2D or shift the beam in 1D. But shifting the beam in 2D with 2 mirrors? That's new to me, if this is what you do then please explain.

Or do you mean a mirror attached to a device that can rotate in two axis rather than one?

It's not clear to me how you are using "rotate". Do you mean the mirror is an axle, and is rotating? Because that doesn't seem consistent with your explanation. I was assuming you mean you can change the orientation of the mirror with respect to some coordinate system.

Mirrors in kinematic mounts will move along two axes. left-right, and up-down. The motion is independent. 

 

Look at the first two consecutive mirrors after the laser (solid black components, after the EOM) in this setup. 

I can translate the beam left-right, and up-down. The first mirror changes the angle, and the second one returns it to a path parallel to the original, but displaced. (and you could do this so the second mirror sends the beam away from the laser, instead of back toward it)

 

[Frank]

8 hours ago, wosoka said:

Oh where do I even begin.
I can say this is for pico video projection beam and I've tried laser scanners and while they are cheap (the "slow" ones) and easy to control, the distortion of the beam caused by angles of the scanner mirrors is hard to correct by undistorting the pixels in software and results in loss of resolution. Something which would be perfect would be to be able to offset the tiny projection beam without affecting its angle. I had some luck with a gantry system and stepper motors as well like in the CNCs but it was just not fast enough, although close.
 

Well, the barrel or pincusion distortion may be corrected by a corresponding distortion lens before the scanners. But with this solution you seem to be not addressing the distortion of the beam which has bounced off of two scanning mirrors at an angle, I think.

If I can't create an image without distortion, I'm going to say we're left with arrays of lasers and/or fibre-optic guiding.  Fresnel magnifier might fit into the mix, advantage being they're relatively cheap and large, but introduce distortion.  Just some ideas...  Maybe arrays of pico projectors and a Fresnel magnifier array screen.

 

[wosoka]

9 hours ago, swansont said:

It's not clear to me how you are using "rotate". Do you mean the mirror is an axle, and is rotating? Because that doesn't seem consistent with your explanation. I was assuming you mean you can change the orientation of the mirror with respect to some coordinate system.

As mentioned before with an actuator such as galvo scanner or pixel shifting actuator, not a mirror holder, by hand...

Edited March 30, 2018 by wosoka

[swansont]

10 hours ago, wosoka said:

As mentioned before with an actuator such as galvo scanner or pixel shifting actuator, not a mirror holder, by hand...

Mirror mounts can be driven remotely.

[wosoka]

not really, not without DIY or decent speeds

[swansont]

8 hours ago, wosoka said:

not really, not without DIY or decent speeds

Those weren't conditions of your OP.

[wosoka]

In all fairness you can't blame me for not mentioning something in my original post in a post after many other posts I made where I mentioned and linked to fast motors.