New experimental work on megahertz laser steering

[christo]

Interesting science read for anyone dreaming about way faster laser scanners in a post-galvo future.

The summary is that CMU have shown early results in reconfiguring liquid lenses using megahertz range modulation of ultrasonic stimulus. In other words, they rapidly change the shape of water using sound so that it refracts the laser to produce desired images and suggests a path towards high-resolution, high frequency rasterised image/video outputs. It's not ready to replace your galvos today, but maybe one day?:

https://imaging.cs.cmu.edu/ultrafast_steering/

[buffo]

If I'm following this correctly, this is analogous to a 2 dimensional AOM, except it uses water as the refractive element instead of a piezo crystal...

Being able to scan in both x and y using a single medium is really cool. I know that AOMs have been used for very high speed scanning in the past, but it always required a pair of AOMs. (Gary Stadler used this technique with an orange HeNe laser to create the lissajous pattern used for the "wormhole effect" visuals in Star Trek: The Motion Picture.) Even though the AOMs could manage a few degrees of deflection individually, because they had to be arranged sequentially in an X/Y configuration, the small aperture limited the total scan angle to something like .5 degrees or so. This design obviously would not have such limitations.

I didn't see any mention of the maximum deflection angle though. If it's in the low single digits, I'd say it's mostly a curiosity. But if it can manage even, say, 10 degrees, this suddenly becomes viable for all sorts of cool things. (Hi-res, raster-scanning laser TV on-the-cheap, anyone?)

Adam

[kecked]

If I'm following this correctly, this is analogous to a 2 dimensional AOM, except it uses water as the refractive element instead of a piezo crystal...

Being able to scan in both x and y using a single medium is really cool. I know that AOMs have been used for very high speed scanning in the past, but it always required a pair of AOMs. (Gary Stadler used this technique with an orange HeNe laser to create the lissajous pattern used for the "wormhole effect" visuals in Star Trek: The Motion Picture.) Even though the AOMs could manage a few degrees of deflection individually, because they had to be arranged sequentially in an X/Y configuration, the small aperture limited the total scan angle to something like .5 degrees or so. This design obviously would not have such limitations.

I didn't see any mention of the maximum deflection angle though. If it's in the low single digits, I'd say it's mostly a curiosity. But if it can manage even, say, 10 degrees, this suddenly becomes viable for all sorts of cool things. (Hi-res, raster-scanning laser TV on-the-cheap, anyone?)

Adam

in theory you could use high power lasers to steer other lasers by changing the refractive index of a crystal in a nonlinear way.

[laserist]

Neos demoed a 2 axis AOM at the IPS meeting in 1994...

[kecked]

But it was like microradians Deflection. Damn good for adaptive optics however.

[buffo]

But it was like microradians Deflection.

Right. If this new technology can manage even 10 degrees, it will be a game changer.

Alas, the fact that they *aren't* bragging about a wide deflection angle makes me suspect that the max deflection is quite small.

Adam

[laserist]

But it was like microradians Deflection. Damn good for adaptive optics however.

That wasn't my memory of the demo. According to a Gooch and Housego pdf here:

https://www.optoscience.com/maker/go...oductionAO.pdf

Acousto-Optic Beam Deflectors (AOBD) are used to control the position of a laser beam as well
as modulation. By careful broadband design of the transducer and by varying the frequency of the
drive signal, the angle that the laser beam is deflected will change. The AOBD typically deflects the
laser beam over a fraction of a degree to a couple of degrees with a resolution of a few hundred spots
to an upper limit of about two thousand spots. Typical diffraction efficiencies are 40-70 percent.

[kecked]

Fraction degree to couple degree. No idea what a spot is. Nice paper.

couple issues. The self focusing is wavelength dependent per the equation and the wavelength is drive frequency depndant. Thus you would need to deflect each beam rgb separately and then rejoin them. Unlikely to work well. BUT it might work in a pcaom like configuration if multiple modes can be setup in the crystal. Plus special shaping optics ect…

I’ll stick to scanners for now. Still for some long distance throw it could work. Like say the side of a dam or mountain.

now that’s a quick 5min read. Your mileage will vary.

I do however wonder if the step size is small enough if you could use something of a lens to amplify the deviation.

[laserist]

High speed analog electronics isn't remotely my thing. I just remembered seeing a demo back in 94. I was honestly more taken with the PCAOMs they were also showing off. I'd left laser shows behind a decade before having realized that Laserium was in a death spiral. Amazing how long it took to auger in. I was at the conference as part of the introduction of the Zeiss APQ refractor telescope line.

[kecked]

High speed analog electronics isn't remotely my thing. I just remembered seeing a demo back in 94. I was honestly more taken with the PCAOMs they were also showing off. I'd left laser shows behind a decade before having realized that Laserium was in a death spiral. Amazing how long it took to auger in. I was at the conference as part of the introduction of the Zeiss APQ refractor telescope line.

I love gas through pcaom. Nothing is better. No divergence issues, no edge color. It’s perfect…except the dot but ok.