[78-L] Removal of hiss on 78 transfers

[Steven C. Barr]

----- Original Message ----- 
From: "Jeff Lichtman" <jeff at swazoo.com>
>>In a posting I submitted several months ago, (and which was shot
>>down at the time by Dr. Biel), I stated that I have believed for
>>many years that it is technically feasible, (if not already
>>possible), to capture the music completely intact from the grooves
>>of a 78 without any noise whatsoever.
> This is already being worked on at Lawrence Berkeley National Laboratory:
> http://irene.lbl.gov/
> This sound restoration system was developed for use by the Library of 
> Congress.
>
> This optical system does not produce noiseless playback. On a worn
> record it is not such a simple thing to determine the position of the
> cutting stylus. Groove damage can make it hard to tell where the
> groove is at all using this technique, and this optical system has to
> interpolate in order to "play through" bad spots.
>
Also note that the "surface noise" heard on 78's is caused by the fact that
the shellac-based compound of which they are/were made contains various
"fillers!" The particles of these can be of sizes which, when included in 
the
walls of the grooves, produce their own UNdesired "apparent" sound
signals. The path of a lateral-record groove actually reflects the recorded
waveform; at 78 rpm, the outside turn of the groove spiral is about 9.9*pi
inches long (31.1 inches) which must be traversed in 1/78 of a minute
(1.3 seconds) or c. 23.9 ips. A 50Hz tone has a duration of .02 seconds
so its "wave" in the groove is about .48 inches long at the outside or about
.22 inches long on the inside turn! A 5KHz tone has waveforms .01 times
as long...or .005" outside and .002" inside! These numbers approximate
the sizes of grains of "filler"...meaning there is NO easy way to establish
whether a waveform ,003" long is "surface noise" or is actually part of
the recorded signal...?!

Steven C. Barr