Re: [ARSCLIST] Dynamic pitch correction programs (was: wire recorders)

[Eric Jacobs <ericj@xxxxxxxxxxxxxxxxxxx>]

As the adage goes, an ounce of prevention is worth a pound of cure.  This
is clearly such a case.

I use a Simon Yorke S7 turntable, which has a removable spindle.  I've
made my own clear plastic jig with concentric circles and a hole in the
middle for centering both the record on the platter as well as the
record weight.  And this set-up will handle up to 20" discs.  No need
to worry about eccentric wow.  No need to enlarge the spindle hole.
It only takes a few minutes.

Although perhaps many things can be corrected after the fact in the
digital domain, doing it right in the analog domain can sometimes be
much much quicker and easier, and less risk of introducing digital
artifacts.

Eric Jacobs
The Audio Archive


-----Original Message-----
From: Association for Recorded Sound Discussion List
[mailto:ARSCLIST@xxxxxxx]On Behalf Of Mike Richter
Sent: Friday, July 23, 2004 11:59 AM
To: ARSCLIST@xxxxxxxxxxxx
Subject: Re: [ARSCLIST] Dynamic pitch correction programs (was: wire
recorders)


At 04:47 PM 7/23/2004 +0000, Don Cox wrote:
>On 23/07/04, Mike Richter wrote:


> > Yes and no. Eccentric wow is periodic but also variable in amplitude.
> > The amplitude change depends on the pitch of the grooves, so may be
> > predictable or may be a function of the music content.
>
>You are thinking of the variable pitch cutting used for LPs? I don't
>think that was used for 78s.

I am simply saying that the rate of change of the magnitude of pitch
variation depends on the pitch of the recording. True, that was intended to
be well-behaved on 78s and was designed to vary only in microgroove, but
the statement holds.

>But would that affect the off-centre wow? The size of the bumps in a
>road is not affected by the slope of a hill.

It very definitely does. Eccentric wow is many times more audible near the
center of a disc than at the edge.

Eccentric wow arises because the off-center spindle makes the radius change
significantly during a single revolution. The percentage change of radius
for a given eccentricity is greater the shorter the radius. Thus, the
effective pitch change (or time shift) is greater at short radii than at
long.

> > The frequency
> > of the time variation is exactly one cycle per rotation.
>
>And the rotation time could be fixed by playing all discs at a known
>rpm, set by a stroboscope. However, most restorers seem to prefer to try
>to match the playing rpm to the recording rpm, as closely as possible.
>So unless you know this exactly, you would need to adjust.

If there is no information on how the disc was played (e.g., one begins
with a tape made by an unknowable process), then the period is determined
just as phase is estimated: by listening to the end of the recording where
the amplitude is largest. If there is information on the speed of playback,
that can materially affect the accuracy of correction.

As someone suggested, picking up hum is a good way to get a reference - but
that won't help discs pre-1925 and is far from easy on any 78s. Something
about frequency response. :-)


Mike
--
mrichter@xxxxxxx
http://www.mrichter.com/