To get rid of CD scratches, is toothpaste enough or do you have to use sandpaper?

[pionphil]

This person says using car wax wasn't enough and he had to use sandpaper and buff it up with rubbing compound. 

 

 

 

However this person recommends toothpaste: 

 

 

So, is toothpaste enough or would sandpaper and rubbing compound be needed? 

[AudioDoctor]

I bet they are still made but there used to be little hand powered devices that would polish up CDs and I bet they're a lot safer than going at it with sand paper.

[sandyk]

1 hour ago, AudioDoctor said:

I bet they are still made but there used to be little hand powered devices that would polish up CDs and I bet they're a lot safer than going at it with sand paper.

I used to have one. 

You also used to be able to get this done at many Video stores for a small charge. They used special machines to clean up their customer abused rental discs.

[MarkusBarkus]

44 minutes ago, sandyk said:

They used special machines to clean up their customer abused rental discs.

...hey, those allegations were never proven!
 

Admittedly, I have used Armor-all, but not since sometime in the mid-90s, when my disc storage was something less than "Audiophile Style."

[bluesman]

Are the scratches causing any problem other than irritation every time you look at them?  If not, I'd leave them alone.  CDs have 2 things working for them that vinyl does not have:

 

1. There's an error correction methodology on CDs called CIRC (Cross-interleaved Reed–Solomon coding) that enables accurate reconstruction of missing data bursts.
2. The laser that reads the disc focuses below the surface, not on it.  So unless the damage is severe enough to interfere with the laser beam's ability to read the dots and generate a stream of 0s and 1s, there should be no effect on SQ. Scratches don't generally cause an audible problem unless they're so bad that they cause skipping or prevent playing at all.

 

I played with several methods for cleaning up old CDs I got from others over the years, just to see if it made a difference.  It does not, unless a scratch is big enough or an area is opaque enough to impede reading.  I tried everything from toothpaste to flour of pumice to jeweler's rouge to auto body finishing sandpaper (moving progressively from 150 to 1000 grit).  But the method I found to work best is actually the way we used to deal with oil leaks on British motorcycles back in the '60s.  It leaves a flat, smooth surface - but it does remove some of the material.  So removing scratches that penetrate close to the data layer means thinning the disc enough to make it useless.

 

You need a suction cup big enough to hold in your hand (but not so big that it extends over the center hole when affixed to the disc), a sheet of thick glass, a little light oil, and auto valve grinding compound in three grades (coarse, medium, and fine).  A glass shelf, table top, or other solid sheet is easy to clean, extremely smooth, and flatter than any other surface you're likely to find easily.  Clean the glass and the CD well - if you leave even a tiny bit of debris, it will add its own scratches. 

 

Smear a little of the coarse grinding paste on the glass, add a drop or two of oil to it, and spread the slurry around a bit with a finger. Stick the suction cup to the top of the CD to use as a handle and gently rub the disc against the abrasive ooze on the glass. You'll soon notice that it's smoothing the bottom of the disc.  As soon as you detect some improvement by eye or by feel, wipe the glass and the disc clean, rinse, and repeat with the medium grinding compound.  Then go to the fine. 

 

Don't overdo it!  If a scratch or area of surface opacity is truly causing skipping or failure to play, only take it down far enough to fix the problem - it does not have to be gone.  As alternatives to the oil / grinding compound mix, you can use adhesive-backed sanding pads like this one or self-adhesive discs made for use on rubber sanding discs.

[pionphil]

1. Quote

   There's an error correction methodology on CDs called CIRC (Cross-interleaved Reed–Solomon coding) that enables accurate reconstruction of missing data bursts.

   Does this only work when playing the CD live? Not when ripping the CD? 

[fas42]

Man, these are scary techniques!! 🤒

 

I haven't been game to go beyond high shine car polishes - if the first polish doesn't "cut it", then I redo, over and over again, while using strong pressure in the polishing action. This literally heats up the plastic - in one case I overdid it, and the plastic surface deformed, slightly - no longer perfectly flat. ... But I never went beyond "diamond glaze" cutting agents - too much then needs to be done to eliminate the fine scratches of the coarse cutting materials.

[kumakuma]

 

 

My solution for "fixing" severely scratched CDs

[bluesman]

16 minutes ago, fas42 said:

too much then needs to be done to eliminate the fine scratches of the coarse cutting materials

That's why you use the 3 step process from coarse to fine.  LIght pressure on a glass surface will do the job very well if you're careful and gentle.  It's actually very easy.

[sandyk]

59 minutes ago, bluesman said:

2.The laser that reads the disc focuses below the surface, not on it.  So unless the damage is severe enough to interfere with the laser beam's ability to read the dots and generate a stream of 0s and 1s, there should be no effect on SQ. Scratches don't generally cause an audible problem unless they're so bad that they cause skipping or prevent playing at all.

I disagree. Readability does matter.

Please explain why BluSpec etc. CDs with their improved polymer formulations sound better than RBCD with the exact same Data, and not only that, frequently rip at a much higher speed than the same on the RBCD version of BluSpec comparison sets.  Sony obviously agrees, or they wouldn't have provided these comparison sets.

[fas42]

12 minutes ago, bluesman said:

That's why you use the 3 step process from coarse to fine.  LIght pressure on a glass surface will do the job very well if you're careful and gentle.  It's actually very easy.

 

Yes, it's all about technique ... if I was very, very concerned about rescuing a particularly important CD I would most likely give your methods a go - in the end, one uses what one is familiar with, is comfortable with ...

 

I acquired quite a few ex-library CDs over the years - some of which had obviously been used as drink coasters 😉. My way of polishing always did enough to get them usable, mostly. These days, I would rip them - and burn them to CDR ... something I'm now having to do with one or two purchased new CDs; the silvering is letting go, in a small spot.

[fas42]

6 minutes ago, sandyk said:

I disagree. Readability does matter.

Please explain why BluSpec etc. CDs with their improved polymer formulations sound better than RBCD with the exact same Data, and not only that, frequently rip at a much higher speed than the same on the RBCD version of BluSpec comparison sets.  Sony obviously agrees, or they wouldn't have provided these comparison sets.

 

Where this comes into it is that the transports have electrical activity occurring the whole time the disk is being read - this activity will change depending upon the condition of the disk, and the format, and the nature of the pits; enough to impact sensitive analogue areas of the playback chain, if not enough has been done to make the latter robust from interference.

[sandyk]

5 minutes ago, fas42 said:

 

Where this comes into it is that the transports have electrical activity occurring the whole time the disk is being read - this activity will change depending upon the condition of the disk, and the format, and the nature of the pits; enough to impact sensitive analogue areas of the playback chain, if not enough has been done to make the latter robust from interference.

 You can still hear these differences on highly acclaimed players such as the MARANTZ SA-11S1 CD PLAYERS ,but the differences are not  as great, and that is with the RBCD version in a pristine condition.

 

[bluesman]

2 hours ago, sandyk said:

Please explain why BluSpec etc. CDs with their improved polymer formulations sound better than RBCD with the exact same Data, and not only that, frequently rip at a much higher speed than the same on the RBCD version of BluSpec comparison sets.

As I recall, BluSpec CDs are cut to Redbook spec using the same blue laser diodes used to make Blu-ray discs rather than the red lasers used for standard CDs. The original claim (for which I've never seen substantiation or any correlation with SQ), was that the shorter wavelength of blue light allows the beam to cut the pits with "far greater accuracy".  Sony claimed that this was the main reason for their allegedly superior SQ, and that their special Blu-Ray polycarbonate formulation added further improvement.  But there are many comparisons on the web that seem to show absolutely no difference in the same files on the two media, and SQ assessments seem about equally split between identical and slight unspecified or inconsistent advantages to BS.

 

I've seen claims that the specific polycarbonate used for Blu-Spec discs is somehow optimized for Blu-ray.  I've never been able to find any specific info on it and if / how it differs compositionally from the polycarbonate used to make standard CDs.  Sony claims only that it's 6 times smoother. Sony makes no claims about the optical proerties of the material, but I can understand how its optics might affect the read by allowing reduced or more consistent refraction and parallax.  Still and all, the surface finish on standard CDs seems not to be a barrier to accurate and complete data capture by the playback laser.

 

[sandyk]

53 minutes ago, bluesman said:

Sony claimed that this was the main reason for their allegedly superior SQ,

 Barry Diament also has a couple of the BluSpec comparison sets, and also finds that the BluSpec version sounds better, although for him the differences disappeared after ripping them. In my case I was still able to hear an improvement with the BluSpec version after being ripped, but it was less obvious.

Incidentally, with a friend's expensive Marantz SA11, you could hear the differences, but they were not quite as obvious as with the less expensive players.