Record High Frequency Achieved

eldavojohn writes "Researchers at UCLA Henry Samueli School of Engineering and Applied Science managed to push our control of frequencies to another level when they hit a submillimeter 324 gigahertz frequency. As any signal geek out there might tell you, this is a non-trivial task. 'With traditional 90-nanometer CMOS circuit approaches, it is virtually impossible to generate usable submillimeter signals with a frequency higher than about 190 GHz. That's because conventional oscillator circuits are nonlinear systems in which increases in frequency are accompanied by a corresponding loss in gain or efficiency and an increase in noise, making them unsuitable for practical applications.' The article also talks about the surprising applications this new technology may evolve into."

Hardly the highest frequency!

[oskay]

Precision phase coherent control of lasers has become possible in the last ten years- Laser beams at frequencies exceeding 1 PHz (10^15 Hz) have been precisely controlled, phase locked, and tuned to have frequencies that are *exact multiples* of our best microwave frequency standards (e.g, cesium). It works the other way too-- our most precise microwave-frequency signals come from divided-down optical frequency references now! See also: 2005 nobel in physics.

Re:How they did it

[ToxikFetus]

The researchers first generated a voltage-controlled CMOS oscillator, or CMOS VCO, operating at a fundamental frequency of 81GHz with phase-shifted outputs at 0, 90, 180 and 270 degrees, respectively. By linearly superimposing these four (or quadruple) rectified phase-shifted outputs in real time, they ultimately generated a waveform with a resultant oscillation frequency that is four times the fundamental frequency, or 324 GHz.

This sounds a lot like a phased-lock loop [wikipedia.org]. And yes, from the article, it appears as though this does have pretty good scalability. TFA said 600 GHz is achievable. 324 GHz a nice because fog is transparent at that frequency.

Sorry, been done before and topped...

[AetherBurner]

Check out http://www.arrl.org/qst/worldabove/dxrecords.html [arrl.org] for the Amateur Radio DX records. This was achieved long ago and at higher frequencies. Highest RF frequency used for a confirmed two-way communication was 403 GHz between WA1ZMS/4 (FM07ji) and W4WWQ/4 (FM07ji) on 21-Dec-2004 over a distance of 1.42 kilometers. Achieving a frequency is one thing but being able to use it is another.

T-rays

[kebes]

This technology is another step along to road to widespread technology exploiting Terahertz radiation [wikipedia.org], which is the region of the EM-spectrum between IR and microwaves. Near the end of the article, they mention the possibility of creating imaging systems that can, for example, see through clothes. These applications of so-called T-rays have in fact already been demonstrated. For example, the image in this article [thznetwork.org] shows a man concealing a knife, which is easily visible in the T-ray image. (See also some other pictures here [thznetwork.org].) T-rays reflect strongly off of metals but can penetrate to varying extents through things like clothing and tissue. The military and security applications are obvious. However it would also bring up new kinds of medical imaging, and has been investigated for quality control, too (for example, scanning the inside of foods in assembly lines, etc.). In the previous link I put, there is an example of scanning through a Hershey bar, where you can see the positions of the nuts.

Suffice it to say this is an area of active research that may have many, many applications.

Re:Hardly the highest frequency!

[insignificant1]

Yes, people have achieved higher frequencies, and controlled them very precisely, as you point out; however, such oscillators aren't CMOS oscillators. That's the news, that they've built a CMOS oscillator at such a frequency, not that they have achieved the highest frequency ever to be controlled (which would be a joke). Not exactly what the /. headline implies, though.

Re:Hardly the highest frequency!

[jcorno]

That's an optical frequency. Well, UV, but still, totally different from what they're talking about. Your example has to do with electronic states of matter. They're talking about circuitry.

Re:How they did it

[zippthorne]

Mod parent down. You can indeed do this. They superimposed rectified quarter-phase signals. In fact, it is a pretty common effect that has been known about since at least the invention of the http://en.wikipedia.org/wiki/Rectifier>rectifier.

Long story short: a full-wave rectified sine wave will have 2x the frequency of the original. Even if the original is a PURE SINE WAVE. The output however is no longer a pure sine wave. You can get a pure sine wave if you have the right filters, but you're going to lose quite a bit of gain.

No amount of filtering can extract a "higher harmonic" from a pure sine wave. Perhaps you could filter out any harmonic frequency you desired from a square wave, or sawtooth wave, but it's going to have terrible gain, and I don't think that's what they did: a square wave superimposed with itself pi out of phase and rectified is a constant voltage.

Re:Hardly the highest frequency!

[oskay]

The work with the CMOS circuits is clearly an important achievement.

However, both the Slashdot title ("Record High Frequency Achieved") and summary ("...managed to push our control of frequencies to another level ...") do seem imply that frequency control has not been possible at frequencies that high before. So, it's important to point out that while it's a record, it's only a record within context. (Records within context are fun; you can do anything with them. For example, I hold the bicycle land speed record for all persons with my SSN.)

In any case, it's *not* totally different. Both are examples of frequency control, which is it's own discipline that spans precision timing and applications in all frequency ranges, from RF (on chips and in free space) to optical (on chips, in fibers, and in free space) and beyond.

Re:First Guess: +1, Patriotistic

[Anonymous Coward]

Only at /. does "paranoid conspiracy theory" get +1 interesting...

Sounds like extension of the push-push oscillator

[MetaDFF]

What they did sounds like an extension of the technique used in push-push oscillators to "double" the oscillation frequency.

The basic principle behind a push-push oscillator is that two out-of-phase signals of fundamental frequency f_o are combined such that the fundamental signal and the odd harmonics cancel, while the second harmonic at 2*f_o add constructively. In the case of a push-push oscillator, you only need two signals 180 degrees out of phase. This could be generated with a differential VCO.

Using a push-push oscillator is a well known technique for increasing the frequency of oscillation of a VCO beyond the fMAX of a transistors at a given process node.

The only disadvantage with push-push oscillators is that you end up losing a lot of power as the second harmonics's power will always be much smaller than the power in the fundamental frequency of the VCO.

Re:How they did it

[leighklotz]

This sounds a lot like a phased-lock loop
It doesn't sound like a PLL to me; a PLL has VCO in it, and this is a VCO, but the VCO is just the oscillator part.
I.e., where's the phase comparator?

It sounds more like a quadrature oscillator with 4 outputs. Oscillators have an inherent need for a 180 degree phase shift, and a quadrature oscillator gives you two outputs 90 degrees out of phase. This one gives you 4 outputs 90 degrees out of phase, which seems a bit of a trick.

It may be some variant on the Bubba Oscillator [google.com], which uses 4 stages to reach the 180 degree inversion, but of course the output of each of those is 45 degrees.

Re:How they did it

[tlhIngan]

There's actually no indication of feedback at all here (which is the whole point of a PLL). In general, actually, feedback slows systems down. They do mention that they are using a VCO -- also used in PLLs -- but I get the impression that the purpose here is to, say, generate frequency modulated radio signals; such a modulator would be an open-loop system.

Perhaps the technique is standard frequency mixing [wikipedia.org], a standard technique used in practically every radio receiver these days. It's basically a three terminal device - you feed in two signals, and a third one appears. If the mixer is your standard physics lab ideal mixer, you get the sum and difference frequencies at the output. (In reality, you get the sum, difference, and a bit of bleed through of the original signals). It's used by radio receivers to downcovert the original signal to a 10.7MHz IF (which is how things like "radar detector detectors" work - by detecting the VCO output which would be the expected frequency plus or minus 10.7MHz, and how some radar detectors use non-standard IFs to prevent this). So they'd have three mixers, which can be completely passive devices, first two combine two to get the doubled frequency, then the last one to get the quadrupled one.

All it really needs is a non-linear device to make mixing happen. If you've every been near a transmitter and heard the radio go nuts, it's because the local transmitter is causing the input amplifier to go non-linear and mix its signal with your desired one, also known as intermodulation distortion.