Vinyl Record Total Harmonic Distortion (THD)

[Craig Maier]

Using the Diamond Cut software (spectrum analyzer), I measured the %THD (Total Harmonic Distortion) of my record transfer system using a 300 Hz reference tone on my high quality test record (Hi Fi News*) and from the Make Waves Generator. The numbers are as follows:

Preamplifier + Soundcard:

0.0092% (Measured with the Make Waves Generator as the input to the preamplifier via an L pad attenuator)

Test Record + Turntable + Phono Cartridge + Preamplifier + Soundcard:

1.3 %

It is interesting to note that the dominent source of distortion is not the electronics, but the media (and maybe the phono cartridge) itself.

*Note: This record can be purchased from Tracer

keywords: Vinyl LP Distortion, Record Distortion, Vinyl LP % THD, Record Harmonic Distortion

[DougMac]

Very Interesting

What is the generally accepted audible threshold of THD? In other words, if you played the 300hz tone from the tone generator followed by the signal from the LP, would the average listener be able to hear a difference?

Also, I would think this would be an optimum result. Would it be reasonable to assume that a more complex signal would lead to more THD coming from the media?

Also would you mind sharing the TT/Cartridge combination used in your test?

Doug

[Craig Maier]

The test was run using a VPI HW-19 turntable, serial # 1945. The Phono Cartridge was a Stanton 500 with the LP stylus (new condition). The Preamp was a home brew job that I designed myself about 25 years ago.

My incremental* threshold of perception for detecting harmonic distortion using a 1000 Hz sine wave is around 0.28 %. You can easily measure your own sensitivity to this parameter. Create a sine wave file using "Make Waves". Use a signal in the 300 to 1000 Hz range. Then, bring up the following two items in the software:

1. Spectrum Analyzer set for % THD mode (see your users manual for the proper way to set that up for measuring Harmonic Distortion, which is critical).

2. VVA in 12AT7 mode, "Full Range" mode with the Operating Point set roughly in the middle and the Drive control set to around 20 %.

While previewing the VVA, adjust its mix control to vary the amount of distortion introduced as indicated on the % THD readout in the Spectrum Analyzer while the VVA is operating in Preview mode. Have someone use the bypass button to create a blind test. When the system is bypassed, the distortion will be in the mud and when it is not bypassed, the distortion will be variable depending on where you set the Mix control of the VVA. That way you can determine your own sensitivity to this parameter.

As for whether more complex signals produce differing amounts of distortion, please refer to the Diamond Cut users manual in the definitions section under TIM or Transient Intermodulation Distortion for a brief description of that phenomenon.

*Note 1 - I use the word incremental, because this distortion is the value added on top of what my audio reproduction system introduces into the signal. I have a very high quality system, so it's distortion levels are very low. {JBL 4410 Studio Monitors, Crest FA 901 Power Amplifier**, Home Brew Preamplifier, bi-amped for the sub-woofer***}

**Note 2: This Power Amplifier is EIA rated at 280 W/Channel into 8 Ohms with less than 0.05 % THD at that output power level. As with any power amplifier, it would introduce a much lower level of distortion at normal listening levels in a sound lab (provided that normal levels are not defined as "the threshold of pain!").

***Note 3: The performance of the sub-woofer and its power amp do not come into play with this type of test since Harmonic Distortion are multiples of the fundamental and do not contain sub-harmonics and 300 Hz to 1 KHz are above the 100 Hz crossover frequency.

[DougMac]

Thanks

I'll have to try the test.

I have no problem with those who prefer vinyl to CD. I do take exception to those who try to justify their position by claiming vinyl is a technologically superior system. THD is only one factor, but if I choose to add coloration to the reproduced sound, I want it to be on my terms.

Doug

[Craig Maier]

Quoting DougMac:

" . . . but if I choose to add coloration to the reproduced sound, I want it to be on my terms."

---------------------------------------------

That is why we invented the Virtual Valve Amplifier around 10 years ago now.



I think that shows that Rick and I feel the same way about it!

[DJBohn]

Craig,
For those of us still using conventional outboard phono preamps (with internal RIAA), would we first need to apply a reverse RIAA to the test signal fed to the phono input for checking phono preamp/sound card THD? (I've got to get one of those records for the complete test...)

DougMac,
Years ago the audio magazines claimed through blind listening tests that the average listener couldn't hear Total Harmonic Distortion below about 3%, but that was durring the LP era. Most consumer equipment at that time would have to be very good to just have a combined THD of 3% or less. Combined record & playback of reel-to-reels & cassette decks usually kicked up the total THD in the audio chain quite a bit.

[Craig Maier]

DJ - Good question.

No - you want to have the RIAA EQ in the signal chain, else the % THD indicated will be incorrect (it will read too high compared to reality). So, do not reverse the RIAA for the test if you are using an RIAA preamp.

For those folks using flat preamps, you must apply the software RIAA curve BEFORE applying the signal to the spectrum analyzer so that the system % THD measurements represents reality.