Stereophile Test CD 3 Signals for Testing Amplifiers & CD Players

Signals for Testing Amplifiers & CD Players

[21] Home Theater Reference & Demagnetizing Tone (DDD) 0:23
1kHz at -20dBFS in both channels

The first 10 seconds of this track consist of a steady tone with a frequency of 1kHz. During the subsequent 15 seconds, it decays in level to silence. Its level of -20dBFS—20dB below the maximum peak level of a CD or laserdisc player—corresponds in a THX-specified system to an acoustic level of 85dB spl. To determine the maximum output level of your CD player, feed its output into an AC voltmeter—Fluke, Beckmann, and Radio Shack all sell suitable handheld models—and multiply the reading, which will be somewhere in the region of 200mV (0.2V) by 10.

To compare preamplifiers, power amplifiers, or CD players/processors, measure the level of this tone with an AC voltmeter at the loudspeaker terminals, and adjust the volume control so that the meter reads 1V (1000mV). As long as each component being compared produces a measured level that's within 10mV of 1V (990mV-1010mV), volume differences will not confuse the auditioning.

The decay part of this track will also eradicate residual magnetization in the ferrous components in your system. Just play the track from beginning to end. Repeat if you feel it necessary.

[22] 1kHz Squarewave at -12dBFS (DDD) 0:15
[23] Multitone signal, octave spacing (DDD) 0:15
[24] Multitone signal, 1kHz spacing (DDD) 0:15
[25] Multitone signal, 500Hz spacing (DDD) 0:15
[26] Multitone signal, 500Hz spacing with gaps (DDD) 0:15
[27] 1kHz sinewave (dithered) at -90.31dBFS (DDD) 0:20

In theory, because the CD system is limited to a maximum bandwidth of half the sampling rate—22kHz—it can't actually reproduce squarewaves. A squarewave can be shown by Fourier analysis to comprise a series of odd-order harmonics regularly dropping in amplitude with increasing frequency. For perfect reproduction of a 1kHz squarewave, therefore, we would need to be able to reproduce the 1kHz, 3kHz, 5kHz, 7kHz, etc., components, all the way to infinitely high frequency. However, as the CD system will not reproduce the harmonics above the 21st, at 21kHz, the 1kHz squarewave in track 22 will not have a true square shape, but instead will look like fig.5. It looks as though there is overshoot and ringing before and after each transition from high to low and vice versa; in fact, what you see is what is termed Gibbs Phenomenon—the effect of omitting the high harmonics that would otherwise "square up" the waveform.

Fig.5 Waveform of 1kHz squarewave restricted to 22kHz bandwidth.

The interesting test tones here are tracks 23-26, which feature a large number of pure tones with spectral gaps in between them. The spectra of these signals are shown in figs.6-9; those with spectrum analyzers will found them most useful to look at the performance of data-reduction schemes or perceptual coders, such as ATRAC used in MiniDisc and PASC used in DCC. For example, I used them to examine the performance of the DTS Zeta algorithm, outlined in a recent "Industry Update" article (Stereophile, March 1995, p.33). Track 23 features an series of tones spaced an octave apart, ranging from 20Hz to 20kHz. Because the frequencies of the test tones will generally not be aligned with distortion harmonics—the fundamental tone in the final three tracks is 350Hz—signals such as these can also be useful tools for examining the performance of amplifiers. Spurious tones such as distortion harmonics, intermodulation tones, and noise will poke their heads up between the "picket" fence formed by the fundamental tones.

Fig.6 Track 23, spectrum of equal-amplitude, multitone test signal, with octave spacing and a combined amplitude of -10dBFS (linear frequency scale, 20dB/vertical div.).

Fig.7 Track 24, spectrum of equal-amplitude, multitone test signal, with 1kHz spacing and a combined amplitude of -10dBFS (linear frequency scale, 20dB/vertical div.).

Fig.8 Track 24, spectrum of equal-amplitude, multitone test signal, with 500Hz spacing and a combined amplitude of -10dBFS (linear frequency scale, 20dB/vertical div.).

Fig.9 Track 26, spectrum of equal-amplitude, multitone test signal, with 500Hz spacing but gaps between (linear frequency scale, 20dB/vertical div.).

The final track enables those who can fine-tune the DACs in their CD players or D/A processors to optimize low-level linearity.—John Atkinson

COMMENTS
mav52's picture

eveytime I try to order this test disc it brings me to the review.  Not the page to move to check out.

BacHolz's picture

As above, attempts to order this "Stereophile Test CD 3" brings me to the review page, not checkout.

John Atkinson's picture

Quote:
attempts to order this "Stereophile Test CD 3" brings me to the review page, not checkout.

I just went to the shop page for Test CD 3 - https://ssl.blueearth.net/primedia/product.php?productid=4&cat=2&bestseller=Y - and when you click on "Add to Cart" it works properly. However, when I went to the Test CD page - https://ssl.blueearth.net/primedia/home.php?cat=2 - and clocked on "Buy Now" yes it opened this page.

Clinking on "Buy Now" for Test CD 2 brought up a page for both Test D 2 and Test CD 3.

We will fix this problem.

John Atkinson

Editor, Stereophile

hellmwc's picture

Great test CD,... The recordings are exceptional in every respect.

Regarding the CD booklet; Is this available as a PDF file for download?

Thank you,

William Hellman

John Atkinson's picture
hellmwc wrote:
Regarding the CD booklet; Is this available as a PDF file for download?
Unfortunately not. But the booklet text was identical to what is published in this Web article.

John Atkinson
Editor, Stereophile

Satya's picture

I am unable to use any of the links in the article or in the comments. I would like to purchase this CD.

Cheers

X