Question about Acousto-optic devices.

[krfkeith]

So I've been reading up on the subject on acousto-optic devices, especially with respect to laser projectors. They seem like a very interesting technology, save for one major shortcoming: the low maximum deflection angle. As such, you run into an analogous problem as with electrostatic vs magnetic CRTs. Electrostatics are much faster, but they deflect the beam very little and so the tube has to be looooooooong. With AOMs, you'd end up with needing a screen 20ft away to get even a visible display area which just isn't practical. The question I have is whether you could use a lens, like in a regular CRT or LCD projector to "enlarge" the image. I drew a schematic of what I meant here http://i.imgur.com/6Xz0ng7.png

Now obviously the dimensions are exaggerated, and that's in only one dimension, but would the fundamental concept work?

[mixedgas]

Its not that easy. The laser beam diameter gets expanded by the ordinary lens, just as much as the angle gets expanded.

You need upstream optics to pre condition the laser beam, but this this is tricky, because the beam can then overfill or underfill the AO cell's active area.If you go to a larger cell, the cost of the AO increases exponentially.

As you are looking for two deflection directions, you need two sets of cylindrical preconditioning optics, one for each axis, then a final projection lens. When I see diagrams of these systems, they usually have 7 to 9 lens elements. This is far different then the angle expansion lenses used for laser shows.

Look for books by Adrian Korpel, he started out with the team that did laser video for Zenith TV.
Steve

[-bart-]

AFaik accousto optic deflectors are monochromatic, so producing an rbg co-lineair beam will be a challenge.

[krfkeith]

AFaik accousto optic deflectors are monochromatic, so producing an rbg co-lineair beam will be a challenge.

Oh I know, my plan would be to use a beam of UV light to illuminate a phosphor screen.

[hakzaw1]

PM sent-- welcome to PL from Brazoria..

[swamidog]

PCAOM's are not monochromatic.

AFaik accousto optic deflectors are monochromatic, so producing an rbg co-lineair beam will be a challenge.

[mixedgas]

PCAOM's are not monochromatic.

Swami, there is a slight difference between using a AO for modulation vs deflection. For deflection you keep the amplitude of the RF drive constant and change the frequency. This results in a change in deflection angle. For modulation you change the amplitude of the RF drive and keep its frequency constant, resulting in a constant deflection angle. For each wavelength in the PCAOM, the frequency is adjusted so they have the same modulation angle for each wavelength. The PCAOM driver is actually 4,6,or 8 drivers in the same unit, with their outputs summed.

Because of Raman-Nanth and Bragg laws being a function of wavelength, the deflection angles are wavelength dependent.

Look at the end of the PCAOM crystal, most of them have a correction prism as part of the crystal end. This aids in the undeflected wavelengths escaping the crystal at a large angle for separation.

You can actually use a AO as a grating if you slightly overdrive and mis-adjust it.

In fact, a cool way to do spectral analysis of at band of radio frequencies is to make a AO with a broadband piezo as the driver, mix the radio spectrum with a much higher RF driver frequency, and place a CCD camera down range. Each radio channel corresponds to some light exiting the AO cell at a different angle. So you can watch a whole spectrum at once without much hardware. The only delay is the acoustic transit time across the crystal, so this is far faster then doing a FFT or a swept RF receiver.


Steve

[swamidog]

thanks. i did not know that! this gas stuff is still pretty magical to me.

karl and i were missing with my pcaom and that argon i bought at selem. we were actually kind of confused by the diffraction grating effects we were seeing until we realized the expanded beam from the collimator was too large for the crystal.

Swami, there is a slight difference between using a AO for modulation vs deflection. For deflection you keep the amplitude of the RF drive constant and change the frequency. This results in a change in deflection angle. For modulation you change the amplitude of the RF drive and keep its frequency constant, resulting in a constant deflection angle. For each wavelength in the PCAOM, the frequency is adjusted so they have the same modulation angle for each wavelength. The PCAOM driver is actually 4,6,or 8 drivers in the same unit, with their outputs summed.

Because of Raman-Nanth and Bragg laws being a function of wavelength, the deflection angles are wavelength dependent.

Look at the end of the PCAOM crystal, most of them have a correction prism as part of the crystal end. This aids in the undeflected wavelengths escaping the crystal at a large angle for separation.

You can actually use a AO as a grating if you slightly overdrive and mis-adjust it.

Steve

[mixedgas]

AFaik accousto optic deflectors are monochromatic, so producing an rbg co-lineair beam will be a challenge.

Oh, it can be done with a DSP and Mathemagic :-)

The efficiency of this method is poor (less then 30%) and it produces beat frequencies which cause crosstalk between colors. Which is why single crystal RGB never caught on until the PCAOM solved the problem.

However one cannot argue with Uncle Theodore. Yes, that historic Uncle Theodore:

https://www.google.com/patents/US408...ed=0CFUQ6AEwBQ

I spent a lot of time researching AO color before NEOS decided to come out with the highly discounted 4 channel PCAOM at 1680$, vs 3200 for the 8 channel back in the 90s. I inherited a small sum from Grandfather and bought the 4 channel PCAOM. B

Steve

[mixedgas]

One other thing to keep in mind is AO has a severe XY resolution limit.

Steve