At 06:44 AM 3/25/2014, Tom Fine wrote:
>1. for recordings from the "golden era", many of the microphones were
>incapable of receiving vibrations at 20kHz, much less higher, and properly
>transferring them into electrical voltages and currents. And, downstream
>from the microphones, input and output transformers acted essentiall as
>band-pass filters, unable to pass ultra-sonic frequencies. So what's up
>there at ultra-high frequencies on the LPs tends to be ringing, from
>components of the recording chain, from equalizing applied during
>recording, from RIAA encoding filters, from the cutterheads, from the mics
>themselves, etc. Also scrape-flutter from the tape machines of the day.
"Tape machines of the day" certainly had scrape flutter, but that doesn't
create a lot of energy above 20 kHz. Scrape flutter is FM modulation, with
modulating frequencies generally in the 2 to 3 kHz region. Looking at a 10
kHz tone that has scrape flutter, most of the energy is in the first set of
sidebands which are symmetrical above and below 10 kHz, i.e. from 7 to 8
kHz and 12 to 13 kHz. The energy in the 2nd set of sidebands (extending
out to 16 kHz) is quite low. Scrape flutter will only increase the
bandwidth of the original (undistorted) signal by about 3 kHz.
Harmonic distortion caused by high tape recording levels, however, will
create lots of energy above 20 kHz. Moderate overloads create a lot of 3rd
harmonic distortion, so if the bandwidth of the original signal was 15 kHz,
playback of the tape with a wideband reproducer will have energy up to 45
kHz. Greater overloads will also create 5th and 7th harmonic
distortion. I see this frequently when doing Plangent transfers, because
the bandwidth of the Plangent playback hardware is better than 200 kHz.
Looking at a spectrogram display, which I do routinely now that I use
Izotope RX, it's easy to see that big bursts of energy above 20 kHz
coincide with tape overload, and are likely caused by harmonic
distortion. However, I can also see that portions of the music which are
well below overload still often have energy up to 30 or 35 kHz. How much
of the energy above 20 kHz was actually emitted by the musical instruments
and captured by the microphones, and how much is distortion created by
equipment between the microphone and the tape recorder? An interesting
question. After the advent of multitrack recording, when master tapes
became a second generation recording, it seems quite possible that a lot of
the energy above 20 kHz comes from tape overload on the multitrack. On
first generation masters? I don't know.
-- John Chester
|