Hi everyone.
Whilst suffering from the curse of insomnia one night I got to thinking how to improve the quality of LP captures onto a PC.
Being as a stylus can not be in two places at once when wriggling through those grooves (hence - I believe - the lower channel seperation compared to a CD for example), would there be any advantage to halving the speed that the record rotates to 16.65 RPM, and recording at 32-bit floating point? Then afterwards digitally speeding up the recording to the correct value before saving to 24/96 or one's preferred bit depth?
This may result in a sound quality increase as the stylus is 'flipping left and right' at the usual rate, but at a slower speed, which may - or maynot! - result in slightly more detail retrieval? Due to the slower rotational rate I believe the flywheel effect is lessened, thereby making tracking a little easier for the stylus? Hope this is making some kind of sense lol!
Regardless of the theory, I'd like to try it out. Is the best way of getting my Rega to rotate at this speed by using something like a Variac?
Cheers,
- John
Whilst suffering from the curse of insomnia one night I got to thinking how to improve the quality of LP captures onto a PC.
Being as a stylus can not be in two places at once when wriggling through those grooves (hence - I believe - the lower channel seperation compared to a CD for example), would there be any advantage to halving the speed that the record rotates to 16.65 RPM, and recording at 32-bit floating point? Then afterwards digitally speeding up the recording to the correct value before saving to 24/96 or one's preferred bit depth?
This may result in a sound quality increase as the stylus is 'flipping left and right' at the usual rate, but at a slower speed, which may - or maynot! - result in slightly more detail retrieval? Due to the slower rotational rate I believe the flywheel effect is lessened, thereby making tracking a little easier for the stylus? Hope this is making some kind of sense lol!
Regardless of the theory, I'd like to try it out. Is the best way of getting my Rega to rotate at this speed by using something like a Variac?
Cheers,
- John
I can't help you with Rega ( not familiar with ) - but - that's old xperience that both tape and LP have better sound and measurements when speed is doubled ........ (not halved )
Speed doubling is when for recording - more tape going past the heads and (so the theory goes) better recording quality.
Not sure about the RIAA playback needing to be adjusted... can anyone shed some light on this?
Not sure about the RIAA playback needing to be adjusted... can anyone shed some light on this?
Playing back an Lp at half speed will certainly result in more accurate playback at the high end, BUT a loss in the bass end: all frequencies are halved, so if the cartridge has a response down to say 30Hz, then this means it can actually only recover frequencies down to 60Hz as recorded on the LP, once the pitch is corrected later.
The RIAA response has to be corrected too; all turnover frequencies have to be halved, i.e., time constants doubled.
Half speed mastering was common once for the advantages it gives the cutting head - the inertia of the moving parts is half as much of a problem. Same advantage with playback mechanics.
Since you are correcting the RIAA curve, you can also compensate for the unwanted bass rolloff introduced by the cartridge low end limit.
Another possible problem: turntable rumble will double in frequency, bringing it up to a more audible part of the spectrum. On a good turntable rumble will be low enough not to matter.
What drive system does the Rega use? Belt? You'd need to get a motor pulley exactly half the diameter.
I've sucessfully modified a Marantz direct drive to 16.66RPM to play my CBS extra long play discs, but then theres no pitch correction afterwards.
The RIAA response has to be corrected too; all turnover frequencies have to be halved, i.e., time constants doubled.
Half speed mastering was common once for the advantages it gives the cutting head - the inertia of the moving parts is half as much of a problem. Same advantage with playback mechanics.
Since you are correcting the RIAA curve, you can also compensate for the unwanted bass rolloff introduced by the cartridge low end limit.
Another possible problem: turntable rumble will double in frequency, bringing it up to a more audible part of the spectrum. On a good turntable rumble will be low enough not to matter.
What drive system does the Rega use? Belt? You'd need to get a motor pulley exactly half the diameter.
I've sucessfully modified a Marantz direct drive to 16.66RPM to play my CBS extra long play discs, but then theres no pitch correction afterwards.
Thanks for your reply Steerpike 🙂
I hadn't realised there was so much involved, and - sadly - much of it is beyond the skills I have to implement (esp. all the RIAA correcting needed). Still maybe one day when older and wiser 😉
Cheers,
- John
I hadn't realised there was so much involved, and - sadly - much of it is beyond the skills I have to implement (esp. all the RIAA correcting needed). Still maybe one day when older and wiser 😉
Cheers,
- John
I don't do any LP to PC copying myself, but I have read on the WWW that there ARE various plug-ins for software like 'Audacity' that allow you do do RIAA correction afterwards in the digital domain, and that these corrections are customisable for odd playback speeds.
Making a Rega
Hi
Good luck with your endeavours. OT but years ago, I have made my Rega Planar MK111 spin at 78rpm for my shellacs.
I did this by calculating the dia/circumference for a new pulley and still have it laying around.
At the time, the rubber band suspension of the motor was a real problem and I remember I also doubled-up on the rubber bands or added some extra ones.
bulgin
Hi
Good luck with your endeavours. OT but years ago, I have made my Rega Planar MK111 spin at 78rpm for my shellacs.
I did this by calculating the dia/circumference for a new pulley and still have it laying around.
At the time, the rubber band suspension of the motor was a real problem and I remember I also doubled-up on the rubber bands or added some extra ones.
bulgin
This here can probably do the RIAA dewarping in pristine quality
(same author as of the magnificent Click Repair):
Brian Davis' Equalizer
I've tried such things a couple of years ago. It was easy enough
with a DC-motor driven GyroDec.
I could not detect any improvement in treble replay quality,
but any tiny trace of 50Hz hum, which in normal operation is
totally inaudible, stands out like a sore thumb after it has been
translated to 100Hz!
That made me stop these experiments.
Anyone interested in a custom GyroDec power supply
that can do anything between 10 and 100 rpm ?-)
(same author as of the magnificent Click Repair):
Brian Davis' Equalizer
I've tried such things a couple of years ago. It was easy enough
with a DC-motor driven GyroDec.
I could not detect any improvement in treble replay quality,
but any tiny trace of 50Hz hum, which in normal operation is
totally inaudible, stands out like a sore thumb after it has been
translated to 100Hz!
That made me stop these experiments.
Anyone interested in a custom GyroDec power supply
that can do anything between 10 and 100 rpm ?-)
The newest version (14.2.0) of the free program sox
http://sox.sourceforge.net/
can apply RIAA equalization digitally.
But to make this whole scheme work with sox, you must
1) connect to turntable to a preamplifier with
flat frequency response, such as a microphone preamp
2) change the sampling rate field in the sound file header
so sox thinks that it was recorded at normal 33-1/3 RPM
Regarding 2), sox may have a command-line option to
override the sampling rate of the source, so 2) might
not not be necessary.
http://sox.sourceforge.net/
can apply RIAA equalization digitally.
But to make this whole scheme work with sox, you must
1) connect to turntable to a preamplifier with
flat frequency response, such as a microphone preamp
2) change the sampling rate field in the sound file header
so sox thinks that it was recorded at normal 33-1/3 RPM
Regarding 2), sox may have a command-line option to
override the sampling rate of the source, so 2) might
not not be necessary.
johnm said:
Steerpike said:
This is so funny. LOL at myself. I had this idea a few weeks ago, and I thought it was a great one. Oh well...... So much for my recent patent idea. I thought of some of the problems though, such as a 20HZ bass part would put the cartridge in to it's resonant frequency, and might start the arm jumping around. The hum never occurred to me, and far worse, the RIAA eq! Oh well. 🙂
Hey Steerpike, this idea started with me, when I was going to transfer some big reel 15 IPS tapes to digital n a machine that would only do 7.5 IPS. Do you think THAT might be a good idea? No RIAA, and hum and noise would be less to begin with. What are your thoughts on that one?
As Werner said, the tape recording is also frequency non-linear, using NAB or CCIR equaliation curves, that are different for different speeds.
Also, you have the same problem with the tape recorder having say a 30Hz low end frequency response limit, that when running at 'half speed', translates to a 60Hz roll-off, after speed correction.
Its quite a feasible solution though, if you have no access to a tape player that can do 15IPS.
Also, you have the same problem with the tape recorder having say a 30Hz low end frequency response limit, that when running at 'half speed', translates to a 60Hz roll-off, after speed correction.
Its quite a feasible solution though, if you have no access to a tape player that can do 15IPS.
Whoops! I should have remembered that about he Tape E.Q.
Looks like I had TWO bad ideas, instead of one. LOL- Status
- Not open for further replies.