# Highest possible AC frequency possible today?

Tags:
1. Feb 16, 2015

### Jason White

What is the highest possible AC frequency being able to be generated. I don't care if this is usable in a circuit but as long as it can be physically made, what is the highest AC frequency possible? I've heard 70GHz was the max using an amplifier.

2. Feb 16, 2015

### Svein

Well, if not the absolute max, you run into size trouble. At 70GHz, λ≈2.8mm on a wire. That means, of course, that λ/2≈1.4mm. Feedback is going to be very tricky...

3. Feb 16, 2015

### Jason White

I'm a newbie with electrical engineering, i don't see why a wavelength would cause a problem with a wire. Can you go further into detail?

4. Feb 16, 2015

### Svein

When the signal has traveled λ/2, the phase has changed by π, which means that it has changed sign. Negative feedback has changed to positive & vice versa. Capacitive feedback from collector to base is usually negative, but after less than 2mm it turns positive - and you have an oscillator.

5. Feb 16, 2015

### Jason White

So 2.8mm is more than 2mm? i are you saying the frequency would be zoo fast it would act as a DC current since it isn't an "Oscillator"

6. Feb 16, 2015

### Mike_In_Plano

70GHz is low. The US was using 90GHz in gear in the early 90's

7. Feb 16, 2015

### Jason White

Ao i'm assuming they are in the tera hertz today?

8. Feb 16, 2015

### f95toli

It depends on what you mean by "AC". We can generate just about any frequency from DC up to gamma radiation.
There used to be a "gap" between microwaves (up at a few hundred GHz) and far infrared (above 1 THz) but that gap has been closed for a while, it is still a tricky frequency range to use in applications because the frequency is too high for conventional electrical transmission lines and too low for optical fibres. but it is possible and it is becoming more common (it is e.g. used in detectors, spectroscopy etc).

Btw, 70 GHz is not a very high frequency; you can buy "conventional" (albeit expensive) microwave equipment (generators, analyzers and so on) as well as cables etc. that work up to 110 GHz.

9. Feb 16, 2015

### Baluncore

The limit to generation is really the bottom of what is called the Terahertz gap. http://en.wikipedia.org/wiki/Terahertz_gap

If we make a rule that the AC must be carried on a wire or transmission line then we are probably going to be restricted to about 30 GHz.
Beyond 30GHz a waveguide would be needed to keep losses down.

Above 25GHz, significant power can be generated by a gyrotron; http://en.wikipedia.org/wiki/Gyrotron

10. Feb 16, 2015

### f95toli

That depends on the size, you can get by using "wires" as long the dimensions are much smaller than the wavelength; and since we can now easily make structures that are only a few hundred nm in size this means we can go to very high frequencies. I've designed (simple) circuits that operated at about 150 GHz (a device integrated into a log-periodic antenna). There are simple electronic circuits (flip-flops) that operate at several hundred GHz (I believe the record is something like 350 GHz).

Also, standard transmission lines (say coplanar lines or microstrips) work well up to about 100 GHz or so, and these are routinely used; but of course you need to be increasingly careful about the design (impedance control) and choice of materials to avoid excessive losses. .

PS i have seen at least one paper where the authors demonstrated transmission of 1 THz wave on what was essentially a twisted pair.

Edit:
The application notes for the PNA VNA's on the Keysight website has some nice pictures of what a some of this equipment looks like

http://www.keysight.com/en/pd-19092...k-analyzer?nid=-536902643.958548&cc=GB&lc=eng

Last edited: Feb 16, 2015