Vinyl & needle magnified 1000x

Sometimes, during a listening session, I just stare at my gear and appreciate the technology behind music playback. It's fascinating.

 

The left and right channels are inverted in the diagram, are they not? As I understand this, the waves of the left channel should be on the right side of the diagram.

 

I think these longitudinal images are very deceptive, because the effective focal length and depth of field are very large.

Here's an example to demonstrate the effect I'm talking about:



If you looked at the scene from above, you might well find that there are five or ten yards between each runner. You could probably lie them all down flat and have acres to spare.

One reason I think this is that I made some transparent moulds of some records and examined them under high magnification. The grooves are actually much smoother than I was expecting when viewed directly from above. I checked, and this is not a quiet patch:



I think the angle and perspective of the shots in the original post dramatically exaggerate the lumpy impression of the record surface by massively foreshortening it. A view from above would give a better impression of the real ratio between longitudinal and transverse displacement.

One tell-tale in the first picture is that even though it's a close-up of a stylus, you can actually see the curvature of the groove. That shows that the picture contains a very compressed t axis.

 

not really, the pictures from the first post contains a lot of high frequency information, the picture that you show I could bet is mostly vocal range and bass, also notice the "separation groove" and the blank space between songs at the left.

 

That

That's stunning. Thanks. Despite the ancient technology at play, to me it seems even more mysterious and unfathomable how you turn that in to incredibly resolving music than digital signals.

 

5 tons of pressure???

 

From the Handbook For Sound Engineers [4th ed., 2008]:
Channel Orientation
The groove shall be recorded for reproduction with the
right-hand loudspeaker(s), as viewed from the audience,
actuated by movement of the groove wall, which is farther
away from the center of the record.

 

It drives me mad when people say that. You can't measure pressure in tons!

 

From the same source I quoted above:
The vertical tracking force applied to the stylus is
divided between the two walls. Each wall is experiencing
force that is equal to the total vertical force times
the cosine of 45° or 0.707, Fig. 27-26. For instance, if
the vertical tracking force (VTF) is 1 g, each groove
wall will experience a force of 0.7 g.

A very small area of contact exists between the
stylus tip and the groove so the pressure against the
groove wall can rise up to many thousands of pounds
per square inch. For instance, if each wall receives 0.7 g
of force applied through the contact area equal to two
ten millionths of an inch (0.2 × 10–6), the pressure is
7726 lb/inch. With such high pressures and force of
friction between the stylus and the vinyl, the outer skin
layer of the record material melts as the tip slides over
the plastic and then refreezes almost as fast as it melted.
Since the melting temperature of the vinyl is about
480°F, the same temperature exists in the contact area.

 

Maybe not an accurate measurement but the concept is there.

I used to use the line as a stereo salesman, "A stylus has a tough job. It has to trace vibrations as tiny as the wavelength of light while experiencing forces as high as 100 Gs." Being a stereo salesman, we all know I didn't care if that was true. But by now I've done the research.

I've posted on this before but by taking in to account the size of a line contact diamond, you can show that it is capable of tracing vibrations that, at a minimum, get to within an order of magnitude of the wavelength of visible light. It might actually be capable of tracing wiggles in the groove that get down to the wavelength of light. That's still to be determined in my book.

As for the forces involved, 100 Gs is conservative. When the diamond stylus is moving sideways to trace wiggles in the groove, the acceleration involved might be as high as 250 Gs. Again, we get into the same quandary as above. Just as you don't measure pressure in tons, you don't measure force in Gs. You measure it in Newtons. But the acceleration of the stylus can easily exceed 100 Gs.

 

If you're tracking at, for example, 1.75g then the force exerted is 1.75g. If you take that pretty small force and concentrate it over a very small area then you get very high pressure. But it is still only 1.75g of force. Putting a needle on a record doesn't conjure massive forces out of nowhere.

You are confusing force with acceleration.

 

I gather that you're expressing some sort of disagreement with the quote that I posted, but the exact nature of that disagreement is not clear to me. If you are in fact expressing disagreement, what exactly are you disagreeing with?

 

No, I am not.

As a stereo salesman, yes, back then I did. I am not now. I should have made this clearer in my post.

As we all know from Newton, force equals acceleration times mass. F=MA. Acceleration is a component of force. The needle is being accelerated by a lot. It's experiencing accelerations easily as high as 100Gs and maybe a lot higher. Fortunately for our records, the mass is very low. The force of the needle on the record groove is relatively low and manageable. Otherwise the stylus would tear the groove to shreds with one play.

 

Oops, my mistake. I don't disagree with what you quoted; somehow I mixed it up with an earlier quote. My apologies.

 

I thought G was equal to 1 times the force of gravity? (So when you say 100's of Gs you mean very strong forces.) But I see that we agree F=mA so all is well and our physics teachers can rest easy .

 

Got it. Thanks.

 

This is utterly unbelievable to me. The frequency of visible light is 10 orders of magnitude higher than the upper range of human hearing. Please show us. Or post a link to where it was posted before.

P.S. to the OP, awesome pics, thanks for posting!

 

Sure. Easy.

The one difficulty is that the forum won't let me use superscripts. I need to use scientific notation for this explanation. In the following argument, 10-9 is actually 10 superscript -9, or ten to the minus ninth power. Similarly, 10-6 is ten to the minus sixth power.

The wavelength of visible light is general given as 400 × 10−9 meters to 700 × 10−9 meters. 1 x 10−9 meters is a nanometer. I want to use micrometers, also called microns, which are 10−6 meters, instead of nanometers in my explanation. To do that, I have to move the decimal place over. That now makes the wavelength of light 0.4 x 10−6 meters to 0.7 x 10−6 meters. The wavelength of light is 0.4 to 0.7 microns. A micron is abbreviated as μm.

The size of a line contact/Shibata stylus varies with the manufacturer but I've seen cross-sections of the diamond generally spec'd out as anything between 6 x 75 microns down to 2.5 x 75 microns. link That smaller, width dimension, 2.5 to 6 microns, is what comes in contact with the walls of the record groove. It has to be smaller than the smallest wiggle in the record groove in order to trace it. If it is bigger, it will ride right over the wiggle. To get an image of that, imagine the point of an American football running over a wall covered with shingles. Where the edges of the shingles are smaller than the tip of the football, the football will ride over them. Where the breaks between the shingles are larger than the tip of the football, the football will trace them.

That means that a 20 kHz signal can't be any larger than 2.5 microns. We've already established that the wavelength of light is 0.4 to 0.7 microns. Here is where I get the vibrations in the record groove to within one order of magnitude of the wavelength of light. I only have to move the decimal point one number either left or right to compare the two.

Where it gets interesting is above 20 kHz. Back in the quadrophonic days, a Shibata stylus had to trace wiggles in the record groove as tiny as 100kHz, or 100,000 Hertz. I own a Lyra Delos cartridge, which is spec'd flat up to 50 kHz. I'm assume my cartridge's output is down only a few dB at 100,000 kHz. That's only one octave higher. If I can extend my cartridge's frequency response up to 100 kHz, even if it's down a bit in output, I get back that one order of magnitude difference between the size of the stylus and the wavelength of light. The wiggles in the groove that the stylus is now tracing are reduced to the range of zero point something microns, or right at the wavelength of light.

This last bit is speculation on my part. I have no problem demonstrating that the wiggles in the groove are within one order of magnitude of the wavelength of light. That's close enough for this old stereo salesman.

 

Thank you for the correction. It's important to have the right information. I had read in more than one post the opposite (incorrect) orientation.

 

I understand the photographic distortions from the near field to distant azimuth. This effect can be further enhanced by telephoto. The overhead view is more objective, however that groove contains no high frequency content that I can see. The typical groove of relatively high frequency and pitch (or modulation) does have sharp peaks with acute angular sweeps, (closer to that shown in the OP's photo examples) causing extreme acceleration of the stylus as it tracks them. The record groove is an amazing obstacle course the stylus must negotiate without hacking off bits of vinyl.

 

Knowing that the outer wall of the record groove is the right channel and the inner wall is the left channel allows you to set anti-skating by ear. I use the mnemonic device "left lower right raise" to remember what to do.

To set anti-skate by ear, you listen for mistracking on tough to track passages. If you hear mistracking in both channels equally, that generally means you have to raise your tracking force. What we want to hear is mistracking in one channel only. For that to be happen, the anti skating force often is set either too low or too high.

If the mistracking is in the right channel, that means the stylus has enough force to hold it in firm contact with the inner groove wall (the left channel) but there is insufficient force for the stylus to maintain good contact with the outer groove wall (the right channel). To get more force on the outer wall, you raise the anti-skating compensation. That pulls the arm outward, which should reduce mistracking in the right channel.

The opposite also is true. If the mistracking is in the left channel, that means the stylus is probably not in great contact with the inner groove wall. The anti-skating force is often set too high, pulling the stylus away from the inner groove wall. Lowering the anti-skate force will cause the arm to pull inward, which should improve tracking in the left channel.

Left lower right raise. Listen for mistracking and use that phrase to set your anti-skate force. If there is mistracking in the left channel, lower the anti-skate compensation. If the mistracking is in the right channel, raise the force. Left lower right raise.

 

Oops.

In my post above about wiggles in the record groove and that they are close to the wavelength of light, I was so conscious of getting the math right, I screwed up the writing.

The fifth paragraph in the post starts out, "That means that a 20 kHz signal can't be any larger than 2.5 microns." No, that's wrong. The opposite is true. It means that a 20 kHz signal can't any smaller than 2.5 microns." Please make the change in your mind. Mine is shot.

 

Regardless of the technical hoo-ha (as interesting as it may be), the pictures are awesome...

 

I won't debate you about the wavelength of light, however, I do know something about quadraphonic records. The CD-4 (for Compatible Discrete 4-Channel) record contained two carrier signals (one each for right and left) in the range of 30kHz-45kHz which contained the difference information to tell the demodulator which sounds went to the front channels and which went to the rear channels. Because the signal had to make it from the cartridge to the demodulator without being attenuated, the wiring and cables had to have an impedance of 100,000 Ohms, which is where I believe you're getting your 100,000 number from. The CD-4 system can be thought of as a sort of FM signal incorporated onto the vinyl. BTW, I believe the electron microscope images the OP posted are actually the grooves on a CD-4 record.

 

Thank you for the correction. You are right. I remembered the bandwidth used for CD4 reproduction incorrectly. My error.

I can get a stylus needing to read bumps in the groove close to the wavelength of light but I can't quite get there. I'm still at least an order of magnitude away.