Modulation efficiency

[danielbriggs]

Agreed,
However I'll give it some more thought.
Off the top of my head, 940/980nm may be suitable? Cheap enough for relatively high powers.
Also, if the "heater" IR pump was placed orthogonally to the 1064->532nm beam, colour output would be unaffected?

[flecom]

my question would be simply, would PCAOM's work with most of our lasers? most of the drivers have standard analog lines as selectable wavelengths, which 473, 532, 635, 642, and 650 are not standard Ar/Kr lines... wouldn't this increase losses? also most PCAOMs I have seen have tiny apertures, further increasing your losses especially if you have a non-642 red

also isn't polarization an issue with PCAOMs? considering a lot of people are combining lasers with PBS cubes this would also be an issue?

[DZ]

Couple tests of my own: Top image, top trace is green, bottom trace is red. Bottom image, top trace is blue, bottom trace is modulation signal. What a mess, we pay good money for this! At least it looks decent, for the most part. Lasers used: Green Lasever, Red is a maxymodule with an LOC diode, Blue is LaserWave.

[ElektroFreak]

Lately there has been a solicitor by the name jiepu-crystal in the BST section. It just happens that the company this person works for produces a new kind of bonded crystal lasing medium which reduces thermal lensing effects and improves efficiency. I think this technology could be the key to some major improvement in modulation response..

http://www.jiepu.com/en/2008/1017/article_2.html

[ElektroFreak]

473nm blue is typically noisy even when running CW, but the CW fluctuations occur at high frequency so that when the beam is scanned it appears like TV static.

It's not surprising to see blue systems respond the worst to modulation. They are naturally more unstable.

@flecom: regarding PCAOMs, I'm not qualified to answer any detailed technical questions about them since I have extremely limited experience with them.

@danielbriggs: It is quite possible that pumping a different side of the doubler with some other diode may impair the production of green, yes.

[mixedgas]

A 4 channel pcaom driver contains 4 programmed oscillators. Last time I ordered one, they were 8$ a piece, A RF amp that should be 50-125$ in quantity, 4 modulators at 6$ a piece and about 30$ of other parts plus a 10$ or so RF combiner. It needs a milled AL case to act as the heat sink for the RF amp. The RF amp module has a precision milled flat as a heat sink, so you need to match the flat to a milled space on the heatsink. The Milled AL case also keeps you from radiating spurs and harmonics in the aviation, local government, and business bands, not exactly places where you want to leak RF. Add in the cost of running a run of circuit boards and a switching psu that cranks out 24V at a amp for the RF heater. I mean RF driver...... yes, they get warm.

Yes, I have built PCAOM drivers and have the RF tools to do so. Its NOT difficult at 4 channels. But 8 channels, well, buy it from NEOS. I have the parts laying around for a 8, and 8 preprogrammed crystals.

If we look at a 8 channel neos driver, it has 647 and 633 red, and 532 green already programmed. the channels are 3-4 nm wide. To cover the wide range of reds, you want PLLs,not crystals, to generate the reds. Thanks to cell phones, PLL chips are now cheap.

Most greens are not that bad. Its the blues right now that need AOs.

What you have to look at is the r to g and r to b and g to B delays in their leading edges.

Feed all three lasers your testing the same signal, not a color corrected signal out of the Pango color wizard. Prepare to be shocked.

I'd also like to see before and after shots of the BENNER/DZ color delay board.

If we use a ADAT, we can play the exact same show to the lasers dozens of times. Use tracks 1,2,3 to play back RGB. Use tracks 4,5,6 to record the light output. use track 7 to store a sync signal for the scope and some sort of time code or pulse from pangolin so we know where we are on the tape. And YES, the adat can record on 3 channels while playing back on 4.
Digitize with a Echo Layla or Echo Gina, using the fiber port on the back of the ADAT. James Lehman has a Layla, and DRLAVA has a Gina. Others here have them as well.

while I'd rather use slow shear wave like a pcaom, there are OTHER ways of doing color if N is 3 or less. This is probably very lossy compared to a PCAOM, but it CAN be done.

http://www.google.com/patents?id=N3A...age&q=&f=false

Note the issue date.

There are some tricks that can be done by running short pulses through the pump diode and pulse position modulating them. PPM is not hard. But nobody who has done this publishes it, they all just want to license it.

Right now, concentrate on the blue...

Didnt I participate in this same discussion about 1 year ago, Deja Vu...

Nice scope photos folks, keep em coming.

But lets try light feedback first. In the old days pango outputted a point clock and a bunch of other useful signals, but that ended with the QM2000 as nobody was using them...

Steve

[ElektroFreak]

But lets try light feedback first. In the old days pango outputted a point clock and a bunch of other useful signals, but that ended with the QM2000 as nobody was using them...

Steve

I definitely agree. Optical feedback is a straightforward and fairly simple solution, making it the ideal choice.

[mixedgas]

my question would be simply, would PCAOM's work with most of our lasers? most of the drivers have standard analog lines as selectable wavelengths, which 473, 532, 635, 642, and 650 are not standard Ar/Kr lines... wouldn't this increase losses? also most PCAOMs I have seen have tiny apertures, further increasing your losses especially if you have a non-642 red

also isn't polarization an issue with PCAOMs? considering a lot of people are combining lasers with PBS cubes this would also be an issue?

633 is close enough to 635 to work. 532 is a standard pcaom line, and pretty sure 473 is in there.
You reprogram the driver, not the crystal

As for polarization, you cant have everything. The way NEOS solved this for a 20-30 watt system was two crystals..

A normal AO will modulate the second polarization at reduced efficiency.

The answer is to use AO for blue, its the most expensive in dollars per milliwatt, hardest to modulate in a linear fashion, and the most sensitive to modulation. Blue is inherently polarized. IF you PWM modulate the AO as I have done in the past, most of the AM "noise" is averaged out of the blue beam and your image will look better, as well as having improved linearity that is inherent in fast PWM. That way you can use a "TTL" surplus AO.

Your only options are galvo modulation, EO or AO/PCAOM. Mems is not there yet.

The blue doubling crystal has to be held to .01 C or better, and .001 c is preferred. Better to NOT modulate the blue pump diode at all.
Folks, keep in mind that LBO/BBO/KTP are ceramics and insulators and do NOT conduct heat well.

One long shot would be to put a high electric field across the blue doubler and disrupt the doubling process, but I doubt anyone is going to try that. I've seen that done as a fine control loop in one precision laser.

Steve

[flecom]

633 is close enough to 635 to work. 532 is a standard pcaom line, and pretty sure 473 is in there.
You reprogram the driver, not the crystal

As for polarization, you cant have everything. The way NEOS solved this for a 20-30 watt system was two crystals..

A normal AO will modulate the second polarization at reduced efficiency.

The answer is to use AO for blue, its the most expensive in dollars per milliwatt, hardest to modulate in a linear fashion, and the most sensitive to modulation. Blue is inherently polarized. IF you PWM modulate the AO as I have done in the past, most of the AM "noise" is averaged out of the blue beam and your image will look better, as well as having improved linearity that is inherent in fast PWM. That way you can use a "TTL" surplus AO.

Your only options are galvo modulation, EO or AO/PCAOM. Mems is not there yet.

The blue doubling crystal has to be held to .01 C or better, and .001 c is preferred. Better to NOT modulate the blue pump diode at all.
Folks, keep in mind that LBO/BBO/KTP are ceramics and insulators and do NOT conduct heat well.

Steve

ya I know the driver - not the crystal - is the issue but I have never played with a driver so I have no idea how difficult it is to retune them... I have a spectrum analyzer that goes to 22ghz so if anything I can retune it myself...

I have an Isomet AOM, just need an 85mhz driver :\

[mixedgas]

ya I know the driver - not the crystal - is the issue but I have never played with a driver so I have no idea how difficult it is to retune them... I have a spectrum analyzer that goes to 22ghz so if anything I can retune it myself...

I have an Isomet AOM, just need an 85mhz driver :\

Nope, you need a 80-90 mhz oscillator TTL clock in a can (digikey) a few minicircuits MAR or ERA MMICs, a few resistors for RF pads, and a minicircuits LRFMS-1, and the Wenzel RF amp at this link. http://www.techlib.com/files/class_c.pdf

normally AO drivers use these or similar for the final amp, and I have a pile of them,

http://www.diplex.info/cgi-bin/ampli...xref=Freescale



5-7 mA into the LRFMS-1 gates the RF full on. Neos usually uses two of them in series with a MRA between them, one for on/off and one for slight adjustments to the drive.

Isomets are anywhere from 250 mW to two watts for the crystal, but for the most part, 250 mW of drive is overkill for newer crystals.

Or just send me a working 22 ghz spec-An and I can send you a driver in trade. I'll even pay the shipping both ways
Steve