DVD Optics question

Back in the day, they still used pits, but it was PWM (continuous time, not clocked or otherwise encoded).

AFAIK, CD media works on interference, so you could get a 0-100% intensity modulation with variable depth pits -- that's percent out of whatever the full scale range is, which is going to depend on reflectivity, dot size and so on, and may be a small fraction of the detector's average intensity. So you'd want to design the signal with DC restore and AGC. It will, of course, be very sensitive to fingerprints and scratches, as LaserDisc was.

Tim

-- Deep Friar: a very philosophical monk. Website:

On a sunny day (Sat, 6 Apr 2013 22:06:33 -0700 (PDT)) it happened 5k3105 wrote in :

Does not work very well, it is about reflectivity, and that varies a lot. There are 'dropouts' too, you need error correction and that requires digital. It would vary between media too, how would you know during recording? Forget it.

The old Philips video disks used digital FM recording, even those flopped because there was no real error correction and lots of dropouts in the video, all sort of otehr artefacts too.

So the only way is digital with a lot of error correction.

Hi Tim,

Even if it was possible I don't have the knowledge or equipment. I was hoping optical would be a good 'non-degradable read' medium for analog recording since it's plentiful and cheap.

Hi Jan,

Is there any media a standard DVD RW laser can burn reliably? Is there a 'stamping' process that could more reliably image a disk?

Second, I imagine there is no way to burn a horizontal, planar wave- shape into the disk?

Sorry, I just can't seem to find much information about this subject - maybe because it's obviously impractical for people who do know about it?

Thanks all.

On Sun, 7 Apr 2013 18:48:55 -0700 (PDT), 5k3105 wrote:

Although I didn't work on DVD-RW, I did work on the first writable CDs. (Before they existed and shortly after.)

They used (and I assume the concept is still used today with DVD-RW) some polymer layers that are dyed. I guess the best way to imagine this is that there, let's say, a "spongy" polymer layer that absorbs red light and "shellac" polymer layer above it that absorbs green light but doesn't absorb red. The disk starts out with these two layers nice and flat, when erased. To write a "1" (or "0" depending on how you want to interpret the physical result) all you do is hit the surface with a red laser at some precise point. The light is absorbed by the spot in the red dyed polymer "spongy" layer and warms up there. As it does so, the "shellac" layer above it (thinner) warms more slowly and only by thermal conduction out of the red-absorbing layer. The spongy layer puffs up when it gets hot like that and the warm, shellac layer gives way because it was softened. When the red laser beam is removed, the shellac layer dissipates its heat rapidly and laterally into the surrounding shellac polymer layer and cools more quickly (and hardens) than the underlying red- absorbing and thicker spongy layer below it. So as the spongy layer cools as well over time, the puffed out bump is "held" in place by the shellac layer above it. To erase the bit, you use a green laser to warm up only the shellac layer. The underlying spongy layer then restores itself to its previous "non bump" volume, pulling the warm shellac layer back down and restoring the "bit." Reading is simply being able to detect a bump or a non-bump.

The above description is how the writable and erasable first CDs worked and may still be how the DVDs operate. The colors chosen will be different with DVD-RW, of course, to allow for density differences. And there may be additional difficulties related to achieving that density that have required additional solutions I'm not aware of, now. But that gives a general idea of the approach, I think.

So far as I know, although one could attempt to design an optical and mechanical system to achieve it, no one has implemented variable height bumps. My guess is that there is already enough variation in the processes that such differentiation isn't likely to be very fruitful. And in cases where the polymers could be redesigned better and new hardware developed to work with that for analog use, there is no market for the extra expenses.

So it's binary.

Besides, the focusing system depends on much of this as well. It uses a quad diode detecter oriented in 45 degree rotation, with diagonal sum and then a differencing amp, along with crafted astigmatism in the optical column to achieve rapid focus as the disks rotate underneath. I suspect that modifying the usual design and polymers would impact this, as well.

It's digital. It's binary. If you want analog, you'll need to convert back and forth, I suspect.

Jon

Thanks Jon,

That seems to explain that the media itself is only good for binary data.

I appreciate the info.

Current DVD-RW drives use high intensity red lasers with power outputs up in the 300mW range or even higher for recent Dual Layer models. With this kind of intensity of light you can relatively easily burn or melt black colored thermoplastic. Black ABS or PVC (chloride warning here!) should be workable at reasonably slow speeds of disk rotation.

At the same time the last (closest to the disk) focusing lens in the player is floating on a voice coil that lets it move left-right (perpendicular to the radius of the disk) and up-down i.e. closer to the disk or further away.

You could make a system that uses that latter degree of movement to shift the focal point and vary the size of the hot spot thus varying the width of the groove that the laser burns in the material. Perhaps the horizontal movement could also be utilized. The voice coil could even be moved by analog means it you prefer (although PWM would probably be easier)

I'm not sure how would you read all that back but you can record (by optical means only) a groove of varying depth and width. I imagine if you have a transparent base to your disk, you can cut the groves all the way through (so you forgo the depth and only go by width) and then shine another laser through the groove and analyze intensity of the light received on the other side of the groove by a photodiode. The wider the groove the more light you receive. I imagine this could be converted back to sound by analog means as well.

So, yes, it is binary as designed but you can make it analog - there seems to be just enough hardware in it to do that. I've no idea what kind of sound quality we are talking here - most likely something pretty crude.

I think the above procedure you describe would ruin the surface tracking method that involves the use of astigmatism and the circle of least confusion used to track warps in the disk. Once you start messing with the focus like that, you have lost the ability to track warping and once you have lost that, you really cannot reliably read back variations you think you wrote before (and probably didn't go the way you wanted them to because of the warps.)

Jon