Tuesday, July 10, 2012

A word of warning

What I'm going to do today is leave a note so that if at any point anyone finds themselves in the situation I did one month ago, they might find this post and save themselves some panicked guesswork.

One of the CDs I shot lasers at
Let's rewind one month, then. I am shooting lasers at CDs. And I don't mean in the usual "use the laser to read the information encoded in the CD" sense. Nor the slightly less usual "use the laser to burn the information onto the CD" sense. I mean I grabbed a CD, stripped away part of the label and reflective layer, and aimed an He-Ne laser through the resulting transparent opening into a wall.

Spots of light caused by diffraction

Why would we do such a thing? The short answer is that CDs (and optic discs in general) work as diffraction gratings. The rough idea is that diffraction gratings allow you to divide a beam of light into a lot of beams of light, and since all those beams of light have the same source , they can interfere with each other in a way that is simple to calculate and visualise*. In particular, what you see when you shoot lasers through a CD is a number of bright spots on the wall behind. One of them is located exactly where it would be, if the CD was just an ordinary piece of transparent plastic. Just trace a straight line from the laser to the wall, and there you go. The other spots are located on either side, at places that can be found with something called the grating equation.

The grating equation relates the angle of the position where the bright spots are, the wavelength of the laser, and something called the "period" of the grating. In a CD, that amounts to the distance between two consecutive grooves, AKA the track pitch.

I already know the wavelength of the laser I'm shooting at the CD, and I can figure out the angle of the bright spots by measuring a couple of distances and using trigonometry. Which means that I can use all this data to measure the track pitch of a C. A distance, incidentally, that a bit of prior research on people who had done similar experiments had shown to be 1.6 micrometers  (a micrometre or micron is a unit of length equal to the millionth part of a metre. It has the symbol μm)

So, as I was saying, I am shooting lasers at the CD. I mark the location of the spots, measure all the relevant distances, do all the relevant math, and find a result: 1.49 μm. Well damn. That's not good.

But wait!, you're thinking. Aren't we talking about a difference of 0.1 microns? A tenth of a millionth of a metre? That's a minuscule difference!

It sure might seem that way, but the measurement I was doing was supposed to be much, much more precise. Specifically, with an error margin of 0.02 μm. A result of 1.49 ± 0.02 means that the biggest possible value for the track pitch, given all those measurements, is 1.51. 

So I recheck all my math, measure the distances again, etc, but nothing changes. and I start to get worried. All the other further things I was supposed to do with that CD would be completely pointless, with such a large error. I needed to know what was wrong! 

And so I turned to the internet, and explored the issue, and what do I come across? Well, that the ubiquitous figure of 1.6 μm that everyone keeps quoting is not, in fact, quite so. Certainly, there are CDs with that track pitch. Those are what we call 74 minute CDs. The much more common in modern times 80 minute/700 MB CD, why, that has a 1.5 μm track pitch.

The industry standards for CDs say that track pitch has to be 1.6 ± 0.1 μm, so of course, many assume that 1.6 is the most common. Certainly that was the case with every single previous experiment on the subject I'd checked previously. What happens is that nowadays, CD players are more reliable than they were when the CD was first created, so you can have a CD with grooves closer to each other and you won't have any trouble playing it. A tighter track pitch means more information can be stored in the same space, so naturally you want to take advantage of that and go to the lower end of the allowable range. Thus, 1.5 μm track pitch CDs.

You hear me, people of the future who might consider doing experiments with CD diffraction? The groove spacing or track pitch for a 700 MB / 80 minute CD is 1.5 μm! Don't let everyone else mislead you!

* In theory, there is no requirement that light be from the same source to interfere. In practice, however, what happens is that any observable effects of interference between different sources lasts for too short a time to be seen.

No comments:

Post a Comment