Vinyl quality

[Martin Bush]

Hello all

I've been listening to lots of vinyl recordings recently as well as reading about the mastering and cutting process.

Something I've read (in fact I read it here some time ago too) is that the closer you get to the end of a disc, the worse the sound is and for that reason the chaps doing the cutting try to avoid the last 15% of the available space. These comments were made in reference to new remasters of classic albums.

So this then got me wondering about 7" singles. Due to their size are they effectively lower quality cuts than they would have been if they had been track one on an LP?

I know LPs spin at 33rpm and 7" singles usually at 45, which may have some effect (I have LPs that play at 45 and singles that run at 33... But I guess those are the exceptions).

I have to say to some extent I don't care as I'm buying and enjoying lots of 7" singles at the moment. But I am curious all the same.

Martin

[Ted Kendall]

Deterioration of quality towards the end of the side is primarily caused by the lower linear groove speed as the centre is approached. As the bumps get closer together and the groove curvature increases, it is more difficult for a tip of finite size to follow them. Elliptical tips are less prone to end-of-side distortion than spherical for this reason.

Of course, the linear speed of a 45rpm groove is higher than that of a 33rpm for the same recorded diameter.

How far you go towards the centre is just one of the factors which have to be juggled when constructing a cut - using less land means that the grooves have to be more tightly packed, risking pre- and post-echo, or the modulation level has to be reduced. If the music at the end of the side is quiet, it is easier to cut further into the centre without excess distortion. A lot of ths art is having to be re-discovered by the new generation of cutting engineers, and sometimes they don't get the optimum balance.

[GrimJosef]

All other things being equal, it's possible to record at higher quality if the disc is spinning faster. Some music has been re-issued at 45rpm on 12" discs (obviously it may then need two discs instead of one) for this reason.

As I understand it the main reason for poorer quality as the stylus gets closer to the centre of the disc is the geometry of the tonearm relative to the turntable. In a perfect world the stylus would move inwards along a straight line which would pass through the turntable's central spindle. In practice the stylus has to swing on a circular arc at the end of the tonearm and for one reason and another the compromises that result are worse in the centre of the record than they are at the edge. A partial solution is to make the tonearm longer, so the arc is closer to a straight line. 12" tonearms as opposed to the regular 9" ones are fairly common. A complete solution is to use what's called a 'parallel tracking' assembly which isn't strictly a tonearm at all but is a motor-driven carriage e.g. http://www.thecartridgeman.com/conductor.htm. They fix the original problem but are tricky (=expensive) to engineer.

Cheers,

GJ

[paulsherwin]

As Ted says, it's basically down to the rotational velocity. Mastering engineers used to try to avoid modulation peaks in the final minutes of a pressing, though this often conflicted with the artistic requirements of the artists and producers. Some people used to claim that the Now/K-Tel compilations sounded better than normal LPs as the modulation level was reduced to cram more music in, so there was less compliance distortion.

45rpm will give better sound than 33rpm, all other things being equal.

[vidjoman]

Of course Pathe got it right in France back in the days of 78’s starting from the centre of the disc. Most music tends to start quietly and build up to a loud finale. So the longer linear track length near the outside of the disc could more easily cope with the increase in sound level.

[Ted Kendall]

Pickup geometry isn't of huge significance in end-of-side distortion, since the arm is (or should be) aligned for zero error at the minimum recorded diameter. That said, I use a Revox parallel track arm for 78 transfers...centre start has been advocated for centuries, but has foundered on the rocks of inertia.

[dave walsh]

The recording convention was broken somewhat from time to and the Rolling Stones Aftermath LP [1966] certainly did this. The session engineers seemed to be very co-operative and the final album contained more content than usual. The twelve tracks are all of a conventional length except the final one/side one Martin "Go'in Home". This is nearly 12 minutes long. It has been described as a little muddy but in the context of the time and the type of music it works very well. I thought it was acceptable then and is now but theoretically, you couldn't/shouldn't do it! As Ted says, there are a lot of technical factors to skillfully get right, depending on expectation, plus the personal less measurable aspects.

Aftermath was treated as something special at the time for squeezing more on an LP at 33 rpm but [thinking about it] Dylan's Highway 61 [1965] had Desolation Row and a lot of other content without any adverse comment. When he put out the [first?] double album in 1966 [Blonde on Blonde] "Sad Eyed Lady" was as long as Go'in Home but had a whole side to itself

Dave

[bluepilot]

Originally Posted by GrimJosef

As I understand it the main reason for poorer quality as the stylus gets closer to the centre of the disc is the geometry of the tonearm relative to the turntable. In a perfect world the stylus would move inwards along a straight line which would pass through the turntable's central spindle. In practice the stylus has to swing on a circular arc at the end of the tonearm and for one reason and another the compromises that result are worse in the centre of the record than they are at the edge.

The tracking shouldn't have very much to do with it. Any decent arm should give you the possibility of adjusting the tracking and you can set it up so that it is perfect at the centre of the disc if you want. It will be a bit off at other places but in practice you can adjust it so that errors are averaged out over the disc and the centre will be just as good as the outside.

The main problem is that in the centre of the disk the audio signals are squashed into a smaller space than at the outside. That makes it harder for the stylus to fit in the groove as the bends in it are sharper. The stylus will have difficulty in tracking high frequencies at high amplitude. Recording engineers don't try to avoid using the last part of the groove. They just try to keep the music quiet and avoid high frequencies.

Come to think of it, I wonder whether that's why it became common for singles to fade out rather that keep going at max volume until the end then suddenly stop.

[Nuvistor]

Originally Posted by vidjoman

Of course Pathe got it right in France back in the days of 78’s starting from the centre of the disc.

They may have done 78’s running from the centre but the circa 1910 discs ran at 90rpm starting at the centre and were hill and dale recording.

I had a customer bring in a standard record player and asked me to convert it to play those 90 rpm disks. Luckily he did not want the automatic working which was a bit of luck or I would have been stumped.

He was very pleased with it when completed, he also had a decent “HiFi” that he bought from me but was happy to listen to his Pathe records.

Just shows it’s the content that matters, not always the vehicle, he was a professional musician.

[bluepilot]

Originally Posted by Nuvistor

They may have done 78’s running from the centre but the circa 1910 discs ran at 90rpm starting at the centre and were hill and dale recording.

I think the later Pathés were 80 rpm. Earlier ones were "between 90rpm and 100rpm". Even with hill and dale recording, the hills and dales will be closer together in the inner tracks and more difficult for the "saphire" to follow.

[dave walsh]

If working outwards from the centre at whatever speed has the advantage of greater linear length and less compression by the end of the track, was it an example of the "better" configuration [inside out] being defeated by the inferior [outside in] simply because it was adopted more readily? Is it perhaps because "outside in" looked more intuitive? Did the multiplicity of factors involved meant there really wasn't much difference between the two configurations overall It's not hard to find technical arguments that suggest parallel tracking is probaly the best of all but that never took of in the same way-less easy to manipulate for the average user ?

Dave W

[Richard_FM]

The Guinness Book of Recorded Sound mentions an early long play system in the 1920s where the turntable speed varied depending on where the needle was. IIRC it would start at a fast speed & gradually slow down towards the centre.

It was supposed to also improve sound quality, but needed some fancy electrics to work.

Some Laserdiscs also did this, I can't remember if they were constant angular velocity or constant linier velocity.

[Ted Kendall]

The World Record Controller - it started slow at the outside and finished near 78 at the inside, the idea being constant groove speed rather than improved quality. The drive system was a knife edge idler akin to the Lenco arrangement, linked to the pickup and driving the record surface. Pemberton-Billing, who invented the system, was also a fully paid-up fruitcake...

[paulsherwin]

CDs have constant linear velocity of course.

[bluepilot]

Originally Posted by dave walsh

If working outwards from the centre at whatever speed has the advantage of greater linear length and less compression by the end of the track, was it an example of the "better" configuration [inside out] being defeated by the inferior [outside in] simply because it was adopted more readily? Is it perhaps because "outside in" looked more intuitive? Did the multiplicity of factors involved meant there really wasn't much difference between the two configurations overall

I think the main reason for starting in the middle was because swarf on the cutting lathe tends to move towards the middle. If you start in the middle the cutter moves away from the swarf. If you start at the outside you need something to clear the swarf away.

Originally Posted by dave walsh

It's not hard to find technical arguments that suggest parallel tracking is probaly the best of all but that never took of in the same way-less easy to manipulate for the average user ?

The tracking of a 9" arm is not bad. Just for fun I transferred some mono discs to my PC in stereo and wrote a program which cross-correlates the left and right channels. On a good record with the tracking well set up the offset between L and R is zero samples at 88200/s sampling rate from start to finish. (It doesn't work with stereo because the audio content is different).

I found that the errors recorded on the discs can be much worse. Some discs have a constant offset where presumably the cutting head wasn't parallel to the groove. Others have an offset which gets worse towards the end where presumably the cutting head wasn't aiming for the centre of the record. On a PC the tracking can be corrected later by deleting a few samples here and there.

[wd40addict]

The 16" 33RPM Vitaphone discs, used to provide sound for the first successful talkies, started from the centre and worked outwards. That way if you're using a steel needle it's at its sharpest playing the slower grooves, when it's getting blunt towards the outside the faster velocity will make up for it to some extent.

With a conventional 78 played with a steel needle you have the worst of all worlds.

If you transfer a set of 78s to a PC the difference in frequency response at the side changeovers can be quite noticeable.