# Designing a Femptosecond clock.

1. May 7, 2013

### Edward Solomo

Hi, I had an idea for creating a clock that could time femptoseconds, although it's possible that this design would most likely fail and only create some crazy diffraction patterns, but anyway, here was my idea (I also didnt' want to waste forum space anywhere else but General Discussion with this idea haha).

--------------------------------------------------------------------------
Ok, so we start with a very flat torus and a very thin cross section, really just a disc with a hole in the middle, the inner radius would be 0.1591 meters, the outer radius will be 0.1592 meters. The inner circumference is exactly 1 meter, and the outer circumference is 1.0003 meters.

We divide the inner circumference into 10 nanometer segments, giving us a total of 100 million subdivisions, a division every 0.0000036 degrees; while also dividing the outer circumference into 100 million subdivisions.

Between each subdivision on the inner circumference, and the respective subdivision on the outer circumference, there would be a photon receptor, on the outer circumference, and a photonic CPU on the inner circumference, this CPU would be very simple, only processing how many times the photonic receptor was activated. It's very important that we are using technology that processes using photons. Each nano-CPU would be independent, except for being able to communicated to a "mother" processor how many times they've been activated, and being able to be told by the mother processor to delete their stored information.

Using the lowest energy X-rays, we can produce a continuous laser that is under 10 nanometers. We would then spin the disc, at 10 revolutions per second. At this rate we could accurately time nanoseconds, because the receptors would be individually activated by the X-ray laser 10 times per second.

Now if we speed this up to 10,000 revolutions per second, we can time picoseconds. The difficulty that confronts us this system is that it would not be feasible to increase the rate of revolution any more that it is currently going, as it's already spinning (on the edge) at 1/30,000 the speed of light.

One way around this problem would be to create a stack of 1,000 of these super thin discs, with each disc rotated 0.0000000036 degrees more than the one below it (1/1000 of 0.0000036). We would also need 1000 separate X ray lasers.

Now each receptor respective to the same numerical receptor on the disc below it, would be activating 1 femptosecond after the numerically respective receptor below. When the "event" ends, which would also have to be executed through photonic computing, the lasers will terminate instantly, and then you can compile the total time by taking a sum of all the activations of each receptor on each disk, the total sum being the number of femptoseconds.

Anyway, something tells me this idea wouldn't work at all. But it was a fun thought! Diffraction express!

2. May 8, 2013

### Staff: Mentor

Making precise clocks is all about managing uncertainties. With what precision can you put the photoreceptors? How finely can you focus the laser? What is the accuracy of the angular velocity of the disc? What is the noise in the photoreceptors? etc.

I think that you will find out that even one of those things will prevent you from reaching even microsecond accuracy with such a device.

3. May 8, 2013

### Edward Solomo

So basically even the tiniest bit of friction or sound wave vibrating the apparatus as a whole would ruin the idea.

I was thinking of using statistical elimination, so you could get a confidence interval, mean + standard deviation and other statistical methods to refine that measurement.

4. May 8, 2013

### AlephZero

You would have to run this in a vacuum, otherwise the spinning disk would act as a centrifugal air pump which would require a lot of power to drive it, and generate a lot of heat!

Aside from the rotor balancing requirements to spin this at the speeds required, the rotordynamcs of this could get quite "interesting", considering the rim is moving a lot faster than the speed of sound in a typical metal.

5. May 8, 2013

### Office_Shredder

Staff Emeritus

To do any kind of filtering on your femptosecond pulses you have to be able to measure how far apart you're receiving them, which requires a femptosecond clock to do...