In article , Dave Plowman (News)
wrote:
In article , Sumatriptan
wrote:
Good morning experts ;-)

I've been going through my HFN test record. Most of it makes sense but
I don't understand my pink noise results. They show an approximate 10
dB / decade tilt, which is what I would expect from pink noise. For
example, here's a frequency analysis of a 'perfect' pink noise sample
generated by Audacity.

http://www.nu-ware.com/Misc/Screensh...nk%20Noise.png

Here is the result of the HFN pink noise test using my M55E cartridge.

http://www.nu-ware.com/Misc/Screensh...nk%20Noise.png

However, when I look at test results around the web I see essentially
flat curves from pink noise tests. Like this:

http://www.tnt-audio.com/sorgenti/shure_m97xe_e.html

So, what am I missing?

I'd say it's likely impossible to record perfect pink noise to vinyl.

FWIW I checked the 'pink' and 'white' noise bands on a variety of test LPs.
They were all wildly different in ways that didn't simply mimic the
response of the cartridge I was using.

The best way I found to do measurements of frequency response was to use a
stepped or swept tone test band from a test disc.

http://www.audiomisc.co.uk/HFN/LP4/NewLampsForOld.html

shows some examples using the Analog Productions test LP.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

In article ,
Jim Lesurf wrote:
FWIW I checked the 'pink' and 'white' noise bands on a variety of test
LPs. They were all wildly different in ways that didn't simply mimic the
response of the cartridge I was using.

I have a digital equaliser which generates what it claims to be pink noise
- and uses that to auto-eq a speaker system. It's very different from the
pink noise Audacity generates.

My feeling is pink noise may not mean the same to all. ;-)

--
*Modulation in all things *

Dave Plowman London SW
To e-mail, change noise into sound.

Dave Plowman (Ignorant Nutcase) wrote:


I have a digital equaliser which generates what it claims to be pink noise
- and uses that to auto-eq a speaker system. It's very different from the
pink noise Audacity generates.

My feeling is pink noise may not mean the same to all. ;-)


** Pink Noise is a general term for any random noise that has the characteristic of equal energy per octave - as opposed to "White Noise" which has equal energy in equal amounts of bandwidth.

However, a given pink noise source will have band limits - a lower and upper frequency beyond which the noise is no longer "pink" or of constant level. When represented on a scope screen, pink noises with different band limits look very different.

Also, pink noise comes in small doses - such as of 1 octave or 1/3 octave too - in which case the centre frequency is quoted.


.... Phil

On 11/03/2015 15:47, Dave Plowman (News) wrote:


I have a digital equaliser which generates what it claims to be pink noise
- and uses that to auto-eq a speaker system. It's very different from the
pink noise Audacity generates.



According to Wiki, 1/f pink noise power density falls off at 10
dB/decade of frequency.The Audacity pink noise plot at:

http://www.nu-ware.com/Misc/Screensh...nk%20Noise.png

Looks close to that.

On 11/03/2015 13:59, Jim Lesurf wrote:

FWIW I checked the 'pink' and 'white' noise bands on a variety of test LPs.
They were all wildly different in ways that didn't simply mimic the
response of the cartridge I was using.

The best way I found to do measurements of frequency response was to use a
stepped or swept tone test band from a test disc.

http://www.audiomisc.co.uk/HFN/LP4/NewLampsForOld.html

shows some examples using the Analog Productions test LP.



The HFN test record I got doesn't have stepped tones, just a rather fast
10 Hz to 20 kHz non-linear sweep. This results in a 'tilted' 10
dB/decade spectral response curve like the pink noise test.

BTW the variety of FR results in your article was interesting and put my
own results into some context.