Vinyl LPs - Wavering Pitch

Discussion forum for Amadeus users

Moderator: Martin Hairer

Post Reply
cegreen
Posts: 21
Joined: Fri Apr 04, 2008 3:00 pm
Location: Pennsylvania

Vinyl LPs - Wavering Pitch

Post by cegreen »

Hello Again,

I have a late 50's classical LP that has noticeable (but not terrible) pitch wavering on sustained notes.

I'm currently guessing that this is probably the result of an off-center hole.

The turntable's tracking force and anti-skate have been carefully adjusted, the LP is flat (no obvious warping), and the pitch wavering is reasonably consistent throughout, not just at the beginning or end of a side.

Is there anything I can try to minimize the wavering pitch?

Thanks in advance.

-Chris

rfwilmut
Posts: 255
Joined: Fri Nov 17, 2006 1:19 pm

Post by rfwilmut »

I'm afraid the best solution is resolutely old-school - I doubt it would be practicable to do it electronically.

Put the record on a turntable and as it turns note how the grooves swing towards the centre and out again. Note where the record is positioned when the grooves are furthest away from the centre. Stop the turntable, turn the record by hand and get as accurate assessment as possible of the position where this happens. Mark the label next to the centre hole in line with this position.

Now you need to enlarge the hole very very carefully. Use a reamer, or a Yale-type doorkey can work. Very gently enlarge the hole on the side where the mark is. Do this in very small increments. Test it by placing the record on the turntable and pushing it so that the mark you made is against the spindle. check how the grooves are swinging (easiest done with a pickup on it).

Keep doing this until you get minimum swing. Be careful not to overdo it as it will be pretty well impossible to correct.

You should be aware that the error won't necessarily be in the same position on both sides, so you should mark both and then consider carefully whether you think you can do this to get both sides correct before starting.

You won't get a perfect result but you may be able to reduce the problem moderately well.

Sonic Purity
Posts: 82
Joined: Sat Nov 10, 2007 11:58 pm
Location: Pasadena, California, U.S.A.

Post by Sonic Purity »

rfwilmut wrote:I'm afraid the best solution is resolutely old-school - I doubt it would be practicable to do it electronically.
I agree, and your method is what i use.
Now you need to enlarge the hole very very carefully. Use a reamer, or a Yale-type doorkey can work.

Keep doing this until you get minimum swing. Be careful not to overdo it as it will be pretty well impossible to correct.
I use a standard round file. It only takes one or two light strokes to take off enough center hole material in many cases.

While it is true that making the hole too large makes it vastly easier to get vastly worse results, the upside is that, placed carefully on a non-changer turntable, there is no chance that the spindle will block the ability to hit exact center. For large-hole 17 cm records, i often omit the spindle adapter entirely when digitizing, so i have total freedom to exactly position the record.

Some here may remember (or have) the Nakamichi Dragon-CT (Centering Turntable). This turntable had a whole separate tonearm hooked up to a servo system that would measure the eccentricity before playback, with the record rotating. The record was then precisely braked, and a “pusher” applied to push a sliding top platter with the spindle across the surface of the main platter beneath (very slightly, of course). The platter was slowly rotated 90° and the process repeated. This is in effect the same method applied electromechanically, with great precision.

***
I have thought about this in the past, and wonder whether it might actually be possible to correct this form of repeatable wow electronically. Seems to me that this form of wow is basically taking the original audio and frequency modulating it at a low amplitude and very low frequency (i calculate 0.55 Hz for 33.3 RPM). If the wow could be detected, then i would think that it could be demodulated along the lines of an FM radio demodulator. The span of frequencies to demodulate would be small (basically fixed for each speed, though we have to account for speed inconsistencies in the analog capture), fixed for a given recording side digital file, decidedly well below the practical low frequency limit of phonograph playback, and the amplitude range limited. Since i believe Martin is highly skilled at math, perhaps this problem may intrigue him.

Anyone ever thought about/looked into this?
))Sonic((

cegreen
Posts: 21
Joined: Fri Apr 04, 2008 3:00 pm
Location: Pennsylvania

Post by cegreen »

Good Morning (it is here, anyway),

Thanks for your thoughts and suggestions. I suspected that the solution was probably to modify the hole and the position of the record, as you suggest.

However, since my father, also a record collector, was doing this from time to time in the 1950's, I was kind of hoping that we would have progressed a bit since then. I guess not.

I have an old Denon DP-61F turntable that was recently tuned-up by the factory techs, who were amazed and pleased to be able to see and work on it. It has a servo-controlled tonearm that is designed to maintain the tracking and anti-skate forces within tolerance in real time.

This seems to work quite well with wavy or otherwise warped discs (I have heard solutions for this problem that involve thick plate glass and a warm oven, but I have never dared to try this with valuable discs.)

When my turntable first came out (early-mid 1980's?), I wondered whether the next step in servo control would be to monitor any in-and-out motion of the tonearm and adjust the rotational speed of the platter accordingly in real time. That never happened, probably because of lack of demand.

I'm guessing that the pitch changes--wow--are caused by variations in the speed of the stylus traveling through the groove as a result of the record being off-center. Essentially like the result of playing an LP at 38rpm or 28rpm. Is that correct?

And as rfwilmut suggests, I was hoping that by now someone would have devised software that could sample the sound and make an appropriate correction.

For now, I think I'll leave well enough alone, keep my reamer in the drawer, and keep my fingers crossed that someone out there will find this an interesting problem to solve.

Regards,

-Chris

rfwilmut
Posts: 255
Joined: Fri Nov 17, 2006 1:19 pm

Post by rfwilmut »

cegreen wrote:I'm guessing that the pitch changes--wow--are caused by variations in the speed of the stylus traveling through the groove as a result of the record being off-center. Essentially like the result of playing an LP at 38rpm or 28rpm. Is that correct?
Yes: the linear speed of the groove under the stylus is, given a constant rotation speed, directly proportional to the radius from the stylus to the centre of the spindle. (So the linear speed almost halves across the whole side, hence the deterioration of quality towards the end of a record.)

When the record is off-centre, the radius is constantly changing, and with it the linear speed: hence the wow.

The electronic solution would be to key an instantaneous pitch changer to the rotational speed of the turntable: but you would have to exactly align its cycle with the offset in the centre hole or you would be making matters worse. Something which could instantaneously measure the variance in the pickup position would seem indicated, but it would have to be set to ignore the constant speed reduction as the record progresses. You can see why a reamer is a lot simpler.

As to warped records, placing one between two sheets of plain glass in a warm (not hot) oven is the recommended method, but you are right to be cautious - it's difficult to know how to set it up unless you have a number of disposable warped records to experiment on.

CDJonah_alt
Posts: 379
Joined: Thu Mar 15, 2007 3:57 pm

Vinyl LPs - Wavering Pitch

Post by CDJonah_alt »

Actully, I don't think one needs something as sophisticated as a "pitch
changer" because the time is changing. So one could fit a small section
of the digital waveform and then resample it. I did this once to
compensate for an old-fashioned _cheap_ tape recorder where the speed
was defined by the rotation of the takeup reel. The practical problems
of getting the correction algorithm in sync and staying in sync would be
a "bit of a" horror. If I were to try it, I would probably consider
doing it semi-manully. Record the thing, pick sections where the the
wow is clearly heard and then interpolate between the sections.

Chuck


rfwilmut wrote:
cegreen wrote:
I'm guessing that the pitch changes--wow--are caused by variations in the speed of the stylus traveling through the groove as a result of the record being off-center. Essentially like the result of playing an LP at 38rpm or 28rpm. Is that correct?
Yes: the linear speed of the groove under the stylus is, given a constant rotation speed, directly proportional to the radius from the stylus to the centre of the spindle. (So the linear speed almost halves across the whole side, hence the deterioration of quality towards the end of a record.)

When the record is off-centre, the radius is constantly changing, and with it the linear speed: hence the wow.

The electronic solution would be to key an instantaneous pitch changer to the rotational speed of the turntable: but you would have to exactly align its cycle with the offset in the centre hole or you would be making matters worse. Something which could instantaneously measure the variance in the pickup position would seem indicated, but it would have to be set to ignore the constant speed reduction as the record progresses. YOu can see why a reamer is a lot simpler.

As to warped records, placing one between two sheets of plain glass in a warm (not hot) oven is the recommended method, but you are right to be cautious - it's difficult to know how to set it up unless you have a number of disposable warped records to experiment on.




_______________________________________________
Amadeus forum mailing list
Unsubscribe / change settings at http://two.pairlist.net/mailman/listinfo/forum_list
--
Charles D Jonah
Building 200
Chemical Sciences and Engineering Division
9700 S. Cass Avenue
Argonne, IL 60439
630-252-3471 CDJonah@anl.gov

_______________________________________________
Amadeus forum mailing list
Unsubscribe / change settings at http://two.pairlist.net/mailman/listinfo/forum_list

cegreen
Posts: 21
Joined: Fri Apr 04, 2008 3:00 pm
Location: Pennsylvania

Post by cegreen »

If I were to try it, I would probably consider
doing it semi-manully. Record the thing, pick sections where the the
wow is clearly heard and then interpolate between the sections.
That approach would certainly work fine for me. I don't have a lot of records with serious wow problems, so it wouldn't be a problem to spend the necessary time.

-Chris

cegreen
Posts: 21
Joined: Fri Apr 04, 2008 3:00 pm
Location: Pennsylvania

Post by cegreen »

I came across a presentation about all of this which might be of interest.

It's in the form of a downloadable PowerPoint presentation from a company in the UK, and it outlines a detailed approach to the problems of identifying and correcting wow and flutter. I thought it was well done.

Here's the link (clicking on the link downloads the file--there is no online content at this address):

http://www.jts2004.org/english/papers/P ... ration.ppt

-Chris

Post Reply