Does anyone understand the physics of vinyl recording and playback?

[sputnik]

Ok, this is something I've wondered about since I was a kid but have been too lazy to ever research. Just some quick back of the envelope calculations indicate that the tangential velocity of an LP record passing under a stylus varies from around 100 feet per minute at the outer grooves to around 45 feet per minute at the inside grooves. This means that one second of music will occupy about 20 inches of groove length in the outermost grooves and only about 9 inches along the innermost grooves.

Does this mean that recording studios might want to place more complex music that may require higher modulation in the outer tracks and save the inside tracks for less complex music? I've never noticed any such correlation so my guess is that it does not make that much of a practical difference. But, as anal as the audiphile world is, it seems like someone would have made big deal about this at some point in the last hundred years.

It's also interesting to note that the tangential velocity of the outer groove of a 78 rpm record is about 160 feet per minute or 32 inches per second. So 78s should have the highest fidelity, right? Maybe there is some trade off in frictional forces or groove wear at some point. Also, is there something magical about 33.33 rpm or is it just a convenient standard to fit a half an hour of music on one side of a record? Does anyone know how and why these standards came about?

[BEC]

Here is one link.

http://www.recordtech.com/prodsounds.htm

Bob

[sputnik]

Excellent link. Thanks Bob.

[Al Klappenberger]

Sputnik,

It sounds to me like you have the details of 33 1/3 recording figured out pretty well. The reduction in speed at the inner grives is why they always sound bad at the inner groves. The only thing you didn't mention is the stereo aspect. Basically each channel of the stereo is one the two walls of the grove. I think the only trick they do to optimize the quality is to spread the groves farther apart during louder passages and comperess them closer during softer ones to increase play time. The dynamic ranges is also compressed to keep the signal above the surface noise. One outfit uses less of the inner grove area and spreads the music over a larger number of disks.

That's about all I know about the subject.

Al K.

[sputnik]

I wonder if there was ever any experimentation with stereophonic 78s.

ADDED: I thought this was an interesting factoid about 78s.

http://stereos.about.com/cs/glossaryandtools/g/78rpm_record.htm

[JJkizak]

If they used some of the new miracle plastics available nowadays they might be

able to press some real good records.

JJK

[DRBILL]

There was another thread on this same subject a while back where I asked why the recordings were not made from the center, out, since the last movement tends to be the loudest.

DR BILL

[WMcD]

There must be a diagram out there someplace. I'll try to pack a lot of things into one post, below. A history lesson, all on one page. Others should correct me, 'cause this is off the top of my head.

My recall is that the original Edison cylinders had vertical modulation. The acoustic cutting head and the play back needle both moved, essentially, up and down in time to the music. Makes sense if you recall Edison's work with a needle connected to the center of a diaphragm. (Horn loading on both ends, of course.)

The 78s changed this with the groove moving relatively left and right. There was some patent fight over some aspect this. The winner called itself "Victor" and the machine was called Victrola. The full story is out there someplace.

In the 30's the radio networks used very large transcription disks (18 inches?) which were shellac. I believe they were running at 33.33 rpm. Consistent with what was said above, the large outer diameter lead to high linear speeds and thus the relatively slow radial (rpm) speed was okay.

These were how half-hour "transcribed" radio shows were distributed (or delayed) when there was not a wire line being used for live broadcast. If you hear "Old Time Radio" it probably originated with one of these.

The radial speeds of 78 and 33.33 were established because of some gear ratio (obviously a fraction of whole numbers) used to gear down a synchonous motor. Apparently belts were not used at first. Maybe belts couldn't be used for the cutting process.

The 33.33 rpm had, I recall, several different systems contending with each other. It was eventally standardized. Probably the "micro groove" system. It used the horizontal modulation and the RIAA boost of highs at recording and cut at playback. This was mostly so that surface noise would be cut during playback and the highs maintained.

In microgroove, the grooves were placed more closely together than before and they were smaller. A special smaller stylus was needed.

The turntables of the day had to play both. Some pick-ups were actually two-in-one where there was a lever at the head of the arm to flip over to which ever one was needed. One for 78 and one for 33. There was a gear selector too, natually.

There was an issue of cross talk between adjacent grooves even in mono. It may have occured in the cutting process where stresses imposed in cutting one groove would affect the one next to it.

You can hear this sometimes where the lead in groove is quite and you hear the pre-echo of what comes in the next revolution.

It was also the situation that loud passages would cause great deviations or modulation. (A similar issue.) So there was a technique where during loud passages the pitch of the grooves would be increased to allow more space on either side. This required some look-ahead.

I think there was no stereo in 78s. One early 33 rpm stereo system used two pickups parallel to each other, spaced about 3 inches. There were two sets of grooves in the platter. It didn't go to far but must have gotten people hooked on the possiblity of two channels on one disk.

Eventually there was the single groove "stereo" which is what we have today (starting long ago, of course). Hard to descibe.

There are two sets of coils in the pick-up and they are a right angle to each other. Consider a V with a 90 degree angle. One "microphone" responses to northeast-southwest motion of the stylus. The other responds to northwest-southeast motion. In theory, motion along one axis does not affect the other. Actual isolation was (is) over 30 dB.

The cutting head had two solenoids cutting the groove in the same way.

Gosh, what smart cookies these guys were. One groove, two independent signals.

I like the 90 degree V shaped canyon analogy . . . and that was the trick. Again, in time, either one can move on the diagonal.

My recall is that the two channels were arranged with reverse polarity. I'm totally not sure of this, but it makes sense.

If they had the same polarity, during a loud passage (with most information in phase) the groove would shift between V and v and V and v, forcing the stylus up out of the groove.

When one channel has reversed polarity you get more like a shift iI to Ii to iI to Ii (where you have to imagine the left and right shift - I can't find anything better on this keyboard). [Edit: those letters looked good in the Courier which shows up on this end and comes out as a Helvetica when I look at it on the BBS. Let me now use Xx to xX to Xx to xX. It is simply to represent relative displacement.)

This makes sense because such lateral movement (high modulation) doesn't force the stylus up; but, rather, sideways. It is a still a challenge to pick a tracking force which keeps the stylus tracking, in the middle of its dynamic range, and does not chew up the vinyl.

Very interesting and brilliant technology. I'm happy with my CD's though.

Smile,

Gil

[sputnik]

Thanks Gil. Another question if you don't mind. How is the up & down and side to side motion translated into an analog signal? There does not seem to be any electrical component associated with the stylus. Is it simply a mechanical sensor? Does the cartridge and/or tonearm function as a tranducer as well as sort out the channels and phasing? I've used pressure tranducers - is it something like that? Thinking about this really gives some idea as to how smart and clever these old guys really were. I grew up just pushing buttons and waiting for the music to come out - it was magic. Thanks again.

[LarryC]

I'll take a whack at it until Gil gets back -- the cartridge is indeed a transducer, with an itty-bitty magnet and 2 itty-bitty coils (if stereo), which act as a generator when they move in relation to each other. Just like a big generator. The coils are at right angles to each other, and both are at 45 degrees of horizontal.

You can see at once that there are two kinds of carts -- a "moving magnet," where the magnet is moved by the stylus and cantilever and the coils are stationary, and the "moving coil" where there are actually two really fine coils on the moving cantilever. It seems like the MM can be designed to put out a much healthier voltage than the MC, perhaps because the coils can be bigger relatively speaking, and of course you can change the needle assembly with a moving magnet by pulling it out, unlike a MC which must be sent back to a factory since you can't pull it out with wires attached to the coils. There are others here who know a lot more, but I thought I'd at least answer your basic question.

Larry

[WMcD]

There is a little microphone in there, or two.

The groove is the music.

No diaphragm, the needle in contact with the groove moves the generator. The generator is a coil in magnetic field caused by a little magnet.

If you have a conductor in a magnetic field, and apply a force, it makes current. That is how the dynamic microphone works. Same thing. So we've got electricity which is an analog to the music. But it is small.

This small current is fed to an amplifier. Then we've got a more power which is fed to the speaker box.

Now we've got the speaker. Here the current is fed to a coil which is much bigger than the little brother in the phono pick up. It is sitting the field of very bigger magnet. It pushes on the diaphragm.

If nothing goes wrong in between, the speaker diaphram is doing about the same as what the needle is doing in the groove, but much more so.

I must say I don't know if I'm being too simple here.

Gil

[colterphoto1]

Gil sez:

"If nothing goes wrong in between, the speaker diaphram is doing about the same as what the needle is doing in the groove, but much more so.

I must say I don't know if I'm being too simple here."

No Gil, that's perfect! We enjoy the occasional simplicity. 'BUT MUCH MORE SO' no shtt sherlock!- I bout fell outa my chair! Thanks,

Michael

[LarryC]

I've always thought how amazing it is that an extraordinarily complex single waveform that contains all the fundamentals and overtones of a full musical group can be cut so accurately into a microgroove, picked up by nearly as precise movements of the stylus and signal generator to generate the same signal in the cartridge, amplified, and sent eventually to the speaker, which as Gil says does the reverse of the cart (and microphone) to send that same signal airborne with amazing accuracy into our rooms.

Larry

[sputnik]

Thanks Larry. That explains alot. I've used the terms without really knowing what was going on. The old vinyl technology is so elegant compared to just reading a stream of ones and zeros.

[pauln]

Consider that at the molecular level of the vinyl wall records are 40 bit.

[sunnysal]

to continue with the digital/analog comparison, I have always viewed the rotational speed of the LP as an analog (forgive the pun) for sampling rate in the CD realm. that is to say that 33 1/3 rpm would be like 44.1k, 45 rpm like 96k, and 78 like 192k...since you can lay down more detail for a given length when cutting the groove...however since the cartridge is forced to retrieve the info at that greater speed I suppose their might be trade offs...

It would be fun to think that the best of both worlds would be to record at the highest rpm and read at the lowest rpm, using digital tranformation to bring the pitch back up to normal, of course we would lose info in the AD/DA cycle...oh never mind...

my head hurts when someone makes me think...lol

tony