New toy!

[daedal]

@daedal: that 914nm line in Nd:YVO4 is responsible for 457nm DPSS blue. It actually gives good gain, which, combined with the superior stability and ease of use associated with Nd:YVO4, is why 457nm DPSS is superior in nearly every way to 473nm Nd:YAG-based blue.

You are right! I completely forgot about that. Doubling the 914nm line will result in 457nm. D'oh!

EDIT: Just noticed your links to the 491 and 543nm systems.. You may be right about the methods used to obtain those wavelengths, but it seems weird to me that SFM of 1064 and 914, one of which is a very strong line, gives only ~2mW of tops when SFM of 1342 and 914nm, both considerably weaker lines, gives up to 1.2W of 543nm. Those power levels seem skewed in the wrong direction to me..

Well... I could be totally wrong here, but wouldn't it be an issue to have a high gain + a low gain line competing? Hence why the 593.5nm and the 491nm lines are really low power (491nm maxes out at 2mW!). On the other hand, and as is the case with the 457nm lasers, as well as multi-vs-single line gas lasers, suppressing the high-gain line allows for a 'good' gain of the two other lines.

My thinking is that when the high gain 1064 and the low gain 1342 or 914 are both on at the same time, the coating is probably suppressing only a little bit of 1064. This would realistically not yield much gain in 914 or 1342. An even more possible scenario, is that no suppression of the 1064 line is done via coatings, resulting in poor 914/1342 output, and thus horrible summation powers. Whilst with 2 lower gain lines on full blast, and the 1064 line completely blocked, the gain would be almost equally divided with little need for AR/HR coating in either of the 1342 or the 914 lines.

I hope any of this makes sense?

--DDL

[danielbriggs]

Extract from an appropriate paper I dug up in the library today:

Based on the above analysis, because of the sub- stantial gain difference between two lasing wavelengths, to optimize high efficiency sum-frequency mixing, the loss values of each respective wavelength at the output coupler should be set to approximately balance the gain matching. Here, we mainly depend on increasing transmission loss of 1064nm in the cavity to optimize the gain matching by designing and coating the highly excellent optical thin film. To reduce the design difficulty, the left facet of Nd:YVO4 was coated with 1064/1342 nm high reflection (HR) coatings as were all reflective mirrors: The concave surface of the output mirror was coated with 1342nm HR, and 1064nm partly transmission. Practically, it is very difficult to achieve the exact transmission values of the output mirrors for wavelength 1064 nm because of experiment errors. After several experiments, eventually, using the output mirror with the optimum transmission value of 1% for 1064 nm, the maximum yellow laser at 593.5 nm output power was achieved. The final experiment result is given below.

Hope that helps,
Dan

[daedal]

Extract from an appropriate paper I dug up in the library today:



Hope that helps,
Dan

Oh, that's cool! My thoughts exactly

Thank you sir

Is this paper available online?

--DDL

[ElektroFreak]

Well... I could be totally wrong here, but wouldn't it be an issue to have a high gain + a low gain line competing? Hence why the 593.5nm and the 491nm lines are really low power (491nm maxes out at 2mW!). On the other hand, and as is the case with the 457nm lasers, as well as multi-vs-single line gas lasers, suppressing the high-gain line allows for a 'good' gain of the two other lines.

My thinking is that when the high gain 1064 and the low gain 1342 or 914 are both on at the same time, the coating is probably suppressing only a little bit of 1064. This would realistically not yield much gain in 914 or 1342. An even more possible scenario, is that no suppression of the 1064 line is done via coatings, resulting in poor 914/1342 output, and thus horrible summation powers. Whilst with 2 lower gain lines on full blast, and the 1064 line completely blocked, the gain would be almost equally divided with little need for AR/HR coating in either of the 1342 or the 914 lines.

I hope any of this makes sense?

--DDL

I think I see what you mean here.. you could easily be correct, but we're a bit over my head here in terms of the actual theory and math that governs these things. I don't have enough information at my disposal to say for sure, myself.

@danielbriggs: Awesome post! Thanks..

[Xytrell]

Better to just go with "magic", I think!

But now I want to know how magic, itself, works. Surely that must be more complicated than sum frequency generation.

[Stoney3K]

Google "Sum frequency mixing" and "Sum frequency generation" if you want more info. Now read some of the patents. Does your brain hurt yet? Because mine does... Better to just go with "magic", I think

Hum, I'm an electronics engineer, and my interpretation is, that this is just the optical equivalent of an AM radio. The two signals (beams) are multiplied in the time domain, which means frequency addition and subtraction. That's exactly the way every baseband-to-RF front-end works in electronics these days, since it would be a bit stupid to do signal processing on the GHz RF signal directly.

Frequency mixing can give you some neat stuff too, like FM synthesis (Yamaha DX7, classic sound!) or instruments like the Theremin.

[ElektroFreak]

But now I want to know how magic, itself, works. Surely that must be more complicated than sum frequency generation.

magicians never tell...

[buffo]

But now I want to know how magic, itself, works. Surely that must be more complicated than sum frequency generation.

Several people here have already posted the basics. Doc's waveform pictures are a good way to understand where the output frequency comes from. (Stoney3K's RF analogy is equally apt; if you've ever dabbled in radio you'll understand where he's coming from right away.)

But if you want to understand the precise non-linear process that extracts a beat frequency from two competing beams of IR light, you're going to have to spend a good bit of time studying up on quantum mechanics. This is where the "magic" comes in.

For example, just about everyone here understands that KTP, BiBo, or LiBo crystals are all capable of frequency doubling, right? After all, it's how we get 532 nm green, 473 nm blue, and so on... But if you asked someone to explain *exactly* what is going on inside that crystal on a sub-atomic level to change an IR beam of light into a visible beam of light, most people won't be able to answer you. And when I say "most people", I'm including college-educated professionals. Hell, I can't even really explain it, even to myself, and I've been studying it for some time now.

Non-linear optical processes are fiendishly complex, and are often completely counter-intuitive. It's a subject that positively lives in the world of quantum mechanics. I've spent a good amount of time trying to wrap my head around it (and bear in mind that I was trained as a nuc before I got old and fat, so I'm no stranger to the quantum world), but so far the true understanding of the process (on a quantum level at least) is beyond my grasp.

Then there's the math to consider. If you don't have at least a 600-level calculus class under your belt, you won't be able to follow the equations. (And since I understand calculus only in it's most basic form, I'm completely lost when I try to work through quantum math.) This is why it's easier to just say "magic". There are some things that I will never understand completely, and I'm OK with that.

The bitch is that if you really understand non-linear optics and can put the theories into practice, you can do some incredible things. One example is bending light around an object to make it disappear. Yeah, I know this sounds like "cloaking device" science fiction, but this has already been achieved with radio frequencies. And no, I'm not talking about stealth technology where you either limit the reflection of incident energy or control the direction at which it's reflected. This is actually diffracting the beam around an object so that it continues past it as if the object was never there. And it's been demonstrated in the lab with RF.

Figure out how to do that with visible light, and you've cracked open a whole new world of optics. Think hyper-powerful telescopes, sub-wavelength lithography, and even lasers with an order of magnitude better divergence than we have now. I have no doubt that one day we'll see some of these devices, but whether it's in my lifetime or not is anyone's guess. (Though we're getting damn close with the sub-wavelength lithography already, now that I think about it.)

Adam

[White-Light]

Just as an aside to the technical discusion, if anyone is looking for a yellow module, CNI make these up to 2.5 Watts in size although I'm unsure of the pricing:

http://www.cnilaser.com/yellow_laser589.htm

[Doc]

@ Buffo:

Regarding the RF light bending, have your read any accounts of the Philadelphia Experiment?