Measurement Consistency - diyAudio
Go Back   Home > Forums > Design & Build > Equipment & Tools

Equipment & Tools From test equipment to hand tools

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th December 2011, 05:51 AM   #1
diyAudio Member
 
Join Date: Nov 2009
Location: Vienna
Default Measurement Consistency

Dear All,

I cannot afford a Audio Precision or a DScope, but I have the following equipment, probably more than most of DIYers:

Picoscope 4262 16bit scope with FTT functions
Picoscope 4424
Leader 1702 (Distortion Meter)
Tektronic 720
Panasonic VP-7201A (Signal generator)

I like to experiment with power amps, DACs, and pre-amp.

I have no problem in qualitative measurements; also harmonics distribution is very straight forward.
The problem comes when I try to make quantitative measurements.

Let' do an example: measuring harmonic distortion with the Picoscope FFT.
Depending on the number of beans that are used in the spectral analysis, or the algorithm chosen, the values are different, it can easily vary of 15 %.

Another problem is the digital signal generators. I can use only a PC, as my equipment does not have any digital generator. Depending on the SW the performances are different. Even within the same software, I can see different shapes of artifacts changing sample frequency or bit depth, and not always the highest is the best.

I often find myself in a situation where I have too many variables, and I am not sure if what I see comes for the DUT or from my test equipment.

Is there a standard for audio measurements? what is used in professional equipment ? How do you do your measurements ?

Best Regards,

Davide
  Reply With Quote
Old 30th December 2011, 06:08 AM   #2
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
PC mother board chipsets, associated electronics can be poor compared to the full potential - I've measured <90 dB on PC loopback with RMAA for a "HD" chipset, 24 bits

but for distortion test waveforms and amp measurements within the Nyquist limit this is still way better than any cheap digital scope

a better soundcard or USB ADC/DAC may be had for a few $100

a cheap universal disc player can also have "not bad" analog output, you can burn the test signals into CD or DVD-A format
  Reply With Quote
Old 30th December 2011, 08:48 AM   #3
diyAudio Member
 
thaumaturge's Avatar
 
Join Date: Aug 2011
Location: Utah
As a previous metrology engineer I OUGHT to be able to anwer this.... but maybe not.

For starters run the tests multiple times in a row using exactly the same setup and equipment. Then shart changing one thing at a time until you find where the inconsistencies are coming from. I'd try an external sound card, such as the Beringer USB (~$40 US). I wouldn't expect great consistency out of the $4-$6 Deal extreme USB sound modules. If you suspect your UUT (unit under test) what do you get when you run just straight through without it?

Those are the best places to start that I can think of. Next would be your PC power supply.

Doc
__________________
Ne timeas a facie mulierum ea ignorare
  Reply With Quote
Old 30th December 2011, 07:57 PM   #4
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
Of course, window shape will have big impact on the FFT. It is a bit of an art I am afraid. I used to use Hamming, but have started to use square. I use TrueRTA for my spectrum analyzer. Is it absolute? No, but it is consistent. It seems very close to the SA in SoundEasy.

I sprung for an e-mu 1616 external mic preamp A2D. 192K 24 bit. I can reliably measure to -135 dB. Only trick is the e-mu being a semi-pro music unit has 600 Ohm inputs so I had to build a very clean little buffer for my probes. I had a Creative audesy but it failed, or I blew it up. I did not have clamps on my probes. It was not bad, but the external mic preamp and extra cables were a problem. I just built a new interface box eliminating most of the connectors. Die cast box, battery powered. Made a big difference.

Watch out for ground issues, that can make things wander.

Second on the power supply. That is what I am trying to beat down now.

Cable dress, lengths, clean connectors, clean switch contacts, on and on.

I have a pile of digital test files. Some I made, some I bought. As I try to measure the difference between a source and result, it does not have to be as perfect as one may think. I astually use the tone generator in TrueRTA as much as anything. It is cleaner than my HP 209A, but then there are times when a big old dial is just right!

M Woofer Tester II is more accurate at measuring low Ohms than my FLuke.
  Reply With Quote
Old 3rd January 2012, 02:38 PM   #5
diyAudio Member
 
Join Date: Nov 2009
Location: Vienna
I'll try the generator of TrueRTA. I'll try to post some pictures of the artifacts that I see in digital signals (assuming the DAC is kind of transparent). I think a good starting point is the characterize the source.

Happy new Year,

Davide
  Reply With Quote
Old 3rd January 2012, 10:07 PM   #6
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
Absolutely.
  Reply With Quote
Old 9th January 2012, 06:05 AM   #7
diyAudio Member
 
Join Date: Nov 2009
Location: Vienna
So Today I had time to make some measurements for the characterization of digital test signal sources. The hardware available as source was:

1) Desktop PC with digital out.
2) Desktop PC + Hiface
3) Laptop Panasonic CF-U1 + Hiface
4) Macbook pro + Hiface

Software for PC TrueRTA and WG140 generators
Software for Mac Sonfbird and Itunes, with test tones generated with WG140

As a DAC I used a Buffalo II + Trident + Legato (All with shunt PSU)

As scope I used the Picoscope 4262.

Conclusions:
A) The desktop PC was hopeless, for quantity of noise, both with his digital output and with the Hiface.
B) The laptop are much quiter when on batteries. The first two pictures show the spectrum with the same configuration with and without PSU.
C) WG140 has less noise than TrueRTA. picture 3 and 4 are an example.
D) The shape of the noise can change when changing the sampling frequency. Not sure if this comes from the DAC or from the source itself.
Pictures 5, 6 7 show the same setup, with the SF goinf from 44.1,96 to 192.
E) I could not see any difference netween Itunes and songbird.
F) The number of samples in the scope change the measurement of the amplitude of the harmonics and the noise floor. I settled on 16k as a compromise between speed and accuracy.

Just to clarify, all this I do not think that can be estrapolated to audible differences in reproduction. The point is that if I see a component of my signal, I want to know if it comes from the DUT or from somewhere else.
After today I think that most of the inconsistencies I was seeing were due to the power supply of the source. I would be curious to see if replacing the normal switching psu with a quiter one would change the story.

More to come,

Davide
Attached Images
File Type: png MAC-HF-IT192-100-BAL-PW.png (30.9 KB, 128 views)
File Type: png MAC-HF-IT192-1000-BAL-BAT.png (29.3 KB, 124 views)
File Type: png LT-wg96-HF-BAT.png (26.9 KB, 125 views)
File Type: png LT-HF-RTA96-1000-BAL-BATT.png (30.0 KB, 124 views)
File Type: png LT-wg44-HF-BAT.png (28.0 KB, 123 views)
File Type: png LT-wg192-HF-BAT.png (28.4 KB, 24 views)
  Reply With Quote
Old 12th January 2012, 09:12 AM   #8
diyAudio Member
 
Join Date: Nov 2009
Location: Vienna
No comments?

D.


Sent from my N-04C using Tapatalk
  Reply With Quote
Old 12th January 2012, 10:37 AM   #9
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
You would expect noise floor to change with sample rate (see "binning"). Apodization (windowing) will also have an effect. For a desktop computer, you might do better with a good quality PCI interface- I've used an M-Audio 192 in two different cheap HP computers and had remarkably low noise.
__________________
You might be screaming "No, no, no" and all they hear is "Who wants cake?" Let me tell you something: They all do. They all want cake.- Wilford Brimley
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Peerless HDS Exclusive - consistency check Multi-Way 6 17th July 2008 11:44 AM
Tweeter unit consistency? clm811 Multi-Way 3 19th March 2007 12:22 AM
Consistency of enclosure types? Kev06 Multi-Way 2 6th October 2006 08:20 PM
Need some measurement help mazeroth Multi-Way 5 24th April 2006 07:22 PM
how do you do a measurement? aarono Multi-Way 0 12th April 2005 01:50 PM


New To Site? Need Help?

All times are GMT. The time now is 12:48 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2