RIAA Response question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, that should be a user selected option. I did both versions for each sampling frequency.

Hi Scott, I've just found this old comment of yours. Great idea (to be able to switch on or off, Allen Wright's 50kHz boost) ... but do you have to be careful how you arrange the wires from the switch to the RIAA network? If you simply twist together the wires - is that OK?


Thanks,
Andy
 

PRR

Member
Joined 2003
Paid Member
Has anybody ever tested the RIAA accuracy of common record lathes? I would not be shocked to see fairly gross errors.

I would be shocked if a stock Neumann or Scully had anywhere near 1dB error 50Hz-15KHz.

Most high-class cutting facilities are further tweaked so the total response (not just the lathe) is as the engineer wishes.

There are "simple precise" ways to calibrate a cut independent of the pickup's foibles. One elegant way is to cut a series of high tones, and then shine a light at an angle. Typically this gives a "wedge" reflection, and the straightness of the reflection edges IS the cutter's response at the groove.
 
Hi Scott, I've just found this old comment of yours. Great idea (to be able to switch on or off, Allen Wright's 50kHz boost) ... but do you have to be careful how you arrange the wires from the switch to the RIAA network? If you simply twist together the wires - is that OK?

My comment was for doing it digitally with an FIR or IIR filter. Any shorting of the wires should be OK, BTW the effect of this extra time constant is very small in any case.
 
I don't think FIR would be a good choice. On RIAA decode you are adding a lot of low end. I would expect the pre ringing to be audible.

You can create a minimum phase FIR RIAA that has virtually no pre-ringing. In any case the pre-ringing is not at a low frequency. BTW Audacity has only a linear phase FIR which has substantial deviation from ideal and no one seems to notice.
 
Last edited:
You can create a minimum phase FIR RIAA that has virtually no pre-ringing. In any case the pre-ringing is not at a low frequency. BTW Audacity has only a linear phase FIR which has substantial deviation from ideal and no one seems to notice.


All the FIR filters i have run into are linear phase. I thought linear phase was inherent to FIR as minimum phase was inherent to IIR.
 

PRR

Member
Joined 2003
Paid Member
Has anybody ever tested the RIAA accuracy of common record lathes? I would not be shocked to see fairly gross errors.

Here is what Neumann claimed in 1973:
15Hz...16000Hz +/-0.5dB

For 1973, half-dB accuracy was pretty good stuff.

I have seen the electronics of later cutting amps. They are very precise equalizers with parts selected to 1% precision (0.08dB).
 

Attachments

  • Neumann-SX74-1973.gif
    Neumann-SX74-1973.gif
    14.8 KB · Views: 162
The SX-74 is very linear. The secondary self resonance is at about 16k. It will be very linear up to there.

PRR asked me about the RIAA circuits. I had asked him not to post the schematic because of the religious nature of some concerning vinyl. I am a cutter by profession and some mystery to the unwashed is helpful. This seems like calm technical bunch. So here goes.

The link is to the later SE66 IC based RIAA encode module. There is an earlier discreet version. This is from the earlier VG66 amplifier rack.The later SAL74 amplifier rack is vastly more common. It was introduced in 1974.

Neumann SE66 - Pro Audio Design Forum
 
There are "simple precise" ways to calibrate a cut independent of the pickup's foibles. One elegant way is to cut a series of high tones, and then shine a light at an angle. Typically this gives a "wedge" reflection, and the straightness of the reflection edges IS the cutter's response at the groove.

That is called the Buchmann and Meyer light pattern method. Most of the important papers on disc recording pre date the AES and were published by the Acoustical Society of America in the U.S. The book “Disc Recording and Playback” Edited by H.E. Roys holds all the secrets. The AES compendiums should be subtitled “ weird, wacky and esoteric papers on disc recording and playback. The Roys book has the good stuff.
 
I know this is OT but what are some practical reasons you would want to use an FIR minimum phase filter instead of a plain old IIR?

Not particularly but an accurate IIR for RIAA requires a little trickery itself at low sampling rates especially if you are trying to bother with the 50k time constant. In this world people tend to obsess over mathematical perfection. It has all appeared here over the years in several threads.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.