Vinyl Asylum

Welcome Licorice Pizza (LP) lovers! Setup guides and Vinyl FAQ.

Return to Vinyl Asylum

Message Sort: Post Order or Asylum Reverse Threaded

Measuring Intermodulation Distortion to adjust VTA

124.168.106.100
Posted on March 18, 2013 at 06:17:51
ianm0
Audiophile

Posts: 306
Location: AU
Joined: June 1, 2006
 Analogue Productions' Ultimate Analogue Test LP, Side 1, Track 9 is for setting VTA by measuring IMD (Internmodulation Distortion) using a IMD tester. Not too many people have this tester.

Is there any way measuring the distortion using an oscilloscope?

 

Hide full thread outline!
    ...
RE: Measuring Intermodulation Distortion to adjust VTA, posted on March 18, 2013 at 11:03:17
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 As Daverz stated, you need a spectrum analyzer. Furthermore, you need a test record with IM test tones. I have measured IM distortion using a software based spectrum analyzer by first recording the IM test tone to 24/96 high-resolution digital and then importing the files to my computer for spectrum analysis. However, I didn't use the measurements for setting VTA. I basically just wanted to see if I could measure IM distortion. The cartridge bodies were level with the vinyl playing surface. Here are some of my measurements.


 

RE: How to digitize and record?, posted on March 18, 2013 at 16:53:10
ianm0
Audiophile

Posts: 306
Location: AU
Joined: June 1, 2006
 Pardon my ignorance. How do you "record" the analog output from your preamp/amp to a digital file? Need an ADC?

 

RE: How to digitize and record?, posted on March 18, 2013 at 18:50:37
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 I use an Alesis Masterlink hard disc digital recorder, which is capable of making 24/96 digital recordings. Then I transport the files to my computer for analysis.

Unfortunately, the Alesis Masterlink is no longer in production, so you would have to find some other method for digitally recording vinyl from your preamp tape output. Many people use their computers with an appropriate ADC interface component.

One thing I forgot to mention about my IMD measurements is that I got much lower values of IMD when playing lateral grooves as opposed to vertically modulated grooves. Consequently, vertically modulated grooves might be the best choice for adjusting VTA. Unfortunately, all my test records are old and were cut with 15-degree VTA, so I don't think they would be appropriate for modern cartridges with 20 to 23-degree VTA.

Interestingly, the Shure V15VxMR had significantly higher levels of both harmonic and intermodulation distortion than the Audio Technica OC9ML/II.

The following sets of graphs are from vertically modulated grooves:






These next sets of graphs are from laterally modulated grooves:




 

Thresholds of perception, posted on March 19, 2013 at 00:53:23
KlausR.
Audiophile

Posts: 2154
Joined: November 17, 2004
 John,

"Interestingly, the Shure V15VxMR had significantly higher levels of both harmonic and intermodulation distortion than the Audio Technica OC9ML/II."


I just finished my piece about VTA and SRA, and found some info on perception thresholds:

* Klippel, “Speaker auralization – subjective evaluation of nonlinear distortion”, AES preprint 5310 (2001)
* Vanderkooy, „Another view of distortion perception“, Convention e-brief, 133rd convention of the Audio Engineering Society 2012


According to Vanderkooy threshold for harmonic distortion with music material is 5-10%. For IMD Klippel mentions thresholds of 1-3% (sine tones, two-tone signals), 1-5% (music material that resembles two-tone signals, such as flute), 10% (complex music material), >30% (music containing percussion and synthesized sounds with high transients and dense spectra).

Klippel states that "Fundamental tones and distortion components falling in one critical band width cause a roughness of the sound which is quite
audible." I'm using the Shure and have never heard something to that effect.

Klaus

 

RE: Thresholds of perception, posted on March 19, 2013 at 11:17:15
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 The findings by Kippel and Vanderkooy seem to defy logic. Normally, one would think that distortion harmonically related to the music would have a higher threshold for detection than IM distortion, which is not harmonically related to the music. However, perhaps this thinking is wrong. Certainly, my measurements show that phono cartridges produce significantly higher levels of IM distortion than harmonic distortion, yet they still seem to sound good.

However, I've always preferred the sound of low-output moving coils to all the Shure cartridges I've owned over the years.

Best regards,
John Elison

 

RE: Thresholds of perception, posted on March 19, 2013 at 23:31:02
KlausR.
Audiophile

Posts: 2154
Joined: November 17, 2004
 "The findings by Kippel and Vanderkooy seem to defy logic. Normally, one would think that distortion harmonically related to the music would have a higher threshold for detection than IM distortion, which is not harmonically related to the music."

Defy logic? Not really, if you consider the phenomenon of auditory masking.

Boer et al., "Audibility of nonlinear distortion in loudspeakers", AES paper 4718:

"Masking is a common phenomenon in sensory perception. It describes the observation that a stimulus is clearly perceivable when presented in isolation, becomes imperceptible when presented together with another stimulus. The term target is used for the stimulus that becomes inaudible due masking, and the term masker is used for the other stimulus. In perception of nonlinear distortion the target can be seen as the generated harmonics, and the masker as the fundamental. The question is, in which amount the harmonics are masked by their fundamentals. For intermodulation distortion this is even more of influence. The sum and difference tones are close to one of the fundamentals and are therefore more likely to be masked. Obviously distortion components that are masked within the audio stimulus do not
need to be reduced. This is why auditory masking is of importance in this perception research."

A further issue probably to consider in this respect is whether or not the masker and the distortion product are within the same critical band.

In view of the above the threshold data make sense: the more complex the music, the more maskers are present in each critical band, the higher the threshold for IMD.

Klaus

 

RE: Thresholds of perception, posted on March 20, 2013 at 08:27:47
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 Tell me what "nonlinear" means. That has always confused me. Is there something called "linear" distortion? What is the difference?

Thanks,
John Elison

 

RE: Thresholds of perception, posted on March 20, 2013 at 09:44:06
KlausR.
Audiophile

Posts: 2154
Joined: November 17, 2004
 Linear distortion is distortion where the shape of the sine wave is not altered, only its amplitude (frequency response) or phase (group delay), e.g. by attenuation in electronic components, think of an equalizer. Linear distortion is signal level independent.

Nonlinear means that new frequencies are generated, harmonics or intermodulation products. Nonlinear distortion is signal level dependent.

Klaus

 

Thanks, Klaus!, posted on March 20, 2013 at 11:54:35
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 I guess I had never heard a good explanation, but now I understand the difference between linear and nonlinear distortion.

Thanks again,
John Elison

 

RE: How to digitize and record?, posted on March 18, 2013 at 19:31:38
ianm0
Audiophile

Posts: 306
Location: AU
Joined: June 1, 2006
 Thanks for the advice. I guess this IMD track is not for the average hobbyists. I was hoping that I could output the analog signal to an oscilloscope and get some meaningful interpretation by looking at the traces.

 

RE: How to digitize and record?, posted on March 18, 2013 at 21:16:25
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 The beauty of this digital age is that you don't really need any actual test equipment to make the IMD measurements. An oscilloscope can be useful for looking at waveforms, but with the IM waveform, you would see a picture like the waveform inserts that I included in one of the graphs.

I have used my oscilloscope to look at square waves and also to look at crosstalk waveforms for the purpose of analyzing changes in azimuth.







Crosstalk Measurements

 

RE: Measuring Intermodulation Distortion to adjust VTA, posted on March 18, 2013 at 12:10:00
Daverz
Audiophile

Posts: 2104
Location: So. California
Joined: September 24, 2002
 That's pretty cool. What software are you using for this?

It looks like the frequencies are 4 kHz and 200 Hz. So you can see the first harmonic of 4 kHz and then side bands at 4 +/- .2 kHz, 4 +/- .4 kHz, etc, and then another set at the 2nd harmonic of 4 kHz.

It would be interesting to see how the side bands change as the VTA is changed.

 

RE: Measuring Intermodulation Distortion to adjust VTA, posted on March 18, 2013 at 14:27:03
John Elison
Audiophile

Posts: 23900
Location: Central Kentucky
Joined: December 20, 2000
Contributor
  Since: January 29, 2004
 I was using Sound Forge software. I suppose there are other programs with spectrum analyzers. The thing I like about Sound Forge is that I can copy the test signal and upload it to my computer for analysis. I don't have to be playing the record while I make measurements. This saves my test records, which are hard to find these days.

Best regards,
John Elison

 

RE: Measuring Intermodulation Distortion to adjust VTA, posted on March 18, 2013 at 08:02:11
Daverz
Audiophile

Posts: 2104
Location: So. California
Joined: September 24, 2002
 Interesting, I hadn't heard of this before.

I think you'd need a signal analyzer (which could be done in software using the FFT*) to do it. I wonder if looking at just the first order IM distortion at 56 and 64 Hz would be enough.

One problem though: if you use that test record, would the setting only work for 200g records?

*I've written (a proprietary) signal analyzer in Python. But the basic software components are all open source.

 

Page processed in 0.026 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]