Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 10th March 2003, 05:12 AM   #11
diyAudio Member
 
ashok's Avatar
 
Join Date: Jun 2002
Location: 3RS
Default A DIY saviour !

Hi Vic,
You should be truly one of the few DIY saviours. For anyone who likes to build things , it is like doing things blind if there is no test gear for measurement. Often these are not affordable for many DIY'ers.
Your programs and GUI look very nice and I hope everyone who uses it will give you feedback to tell you how good or bad it is so that you can hone it to make it the best software on the Net.

The Net generally has almost all the info one wants. But it helps a lot ( and saves time ) if explanatory text is also available from the same site. So it would be a good idea if users ( and others) send you some write up on using the different signals and equipment etc which you can edit and put up along with your downloads. That would make it a complete site !
I think the users owe this to you - the software is after all FREE.
This a fantastic opportunity for all the young audio DIY'ers to get in even deeper without having deep pockets.
Great work. And thanks to Kay again.
Ashok.
To Vic and all at Dazyweb
__________________
AM
  Reply With Quote
Old 10th March 2003, 08:35 AM   #12
Bricolo is offline Bricolo  France
diyAudio Member
 
Bricolo's Avatar
 
Join Date: Nov 2002
Location: Grenoble, FR
Default Re: Thanks!

Quote:
Originally posted by vrbalthezr
I checked my site stats and there was this incredible 360 hits (normal is 40) today which I traced back to this one message (thanks for the plug).
Thanks, everyone for checking out my site. Any input on bugs or feature additions will most certainly be welcome! The SG-2102 just got warble tones as of last week (for you speaker designers) mainly due to a request from my boss who needed them for test tones..

Vic
DazyWeb Laboratories
read my post, just before yours

that's for feedback
  Reply With Quote
Old 11th March 2003, 03:27 PM   #13
diyAudio Member
 
Pjotr's Avatar
 
Join Date: Nov 2002
Location: Netherlands
Hi,

This is to show what can be done with 24bit soundcard. It is a M-Audio Audiphile 2496.

I did investigate Vic’s freeware but I could not manage it to get it working with the M-Audio sound drivers. It seems to work only with the Windows mixers and as such I can only access the Advance sound chip on my MSI main board. Here is some work to do Vic.

So for the test I used AudioTester V2.0, shareware from http://www.sumuller.de/audiotester/ This software is limited to 20 min. sessions unless you register for $18,- But it has more capabilities and does the job better IMHO. Excuse me, no offence Vic.

Attached a screenshot of the response of the soundcard. Exempt for the sine sweep response al measurements were done with averaging set at 100 times. Sampling was 48KHz en I used a Rife-Vince window for the FFT’s. It is made with a 1m loop through cable made of double screened CATV coax. The most even response is with a dirac pulse. But this has the disadvantage of its very low energy of such a pulse. It is probably not suitable to do acoustic LS measurements because of the environmental noise. Best way to this is with a synchronised MLS measurement. But I have found no freeware or low-cost shareware capable of doing such MLS measurements. So for doing LS measurements you are left with white noise or a sweep. Or buy more expensive shareware from http://www.purebits.com/ for instance (around $300).

At the noise floor there are some traces of 50Hz hum and its 150Hz harmonic entering through the loop through cable. This is something to take care of when doing measurements. Digital induced noise from the computer is virtually absent despite the card is powered from the computer PSU, a slight spur at 8 KHz and 18 KHz.
Attached Images
File Type: gif resp.m_audio.gif (14.5 KB, 868 views)
  Reply With Quote
Old 11th March 2003, 03:29 PM   #14
diyAudio Member
 
Pjotr's Avatar
 
Join Date: Nov 2002
Location: Netherlands
Here is a picture of the THD. THD was much lower at –10dB output then at 0dB output (card is not clipping at 0dB). I have traced down this to the buffers of the codec chip used. AudioTester takes into account up to the 9th harmonic but does not use harmonics higher than ½ Fs. You can set the harmonics to take into account by hand if you wish. The stated 0.0003% in the picture is for the –10dB signal. This is quite low and lower than can be reached with a DIY build distortion analyser IMHO. Note that I have offset both measurements for clarity, look also at the accompanying scale at the left.
Attached Images
File Type: gif thd dist m_audio.gif (11.9 KB, 858 views)
  Reply With Quote
Old 11th March 2003, 03:32 PM   #15
diyAudio Member
 
Pjotr's Avatar
 
Join Date: Nov 2002
Location: Netherlands
And here a picture of an IM distortion measurement:
Attached Images
File Type: gif im dist m_audio.gif (10.9 KB, 839 views)
  Reply With Quote
Old 11th March 2003, 03:35 PM   #16
diyAudio Member
 
Pjotr's Avatar
 
Join Date: Nov 2002
Location: Netherlands
As an extra, a picture of the performance of the Advance sound chip on the main board. The loop through cable was a simple screened cable with jack plugs. Look at the considerable mains noise entering through the cable. Also there is considerable digital noise from the main board. But the frequency response is reasonable flat within 1 dB from 20 Hz to 20 KHz.
Attached Images
File Type: gif performance.gif (17.2 KB, 543 views)
  Reply With Quote
Old 11th March 2003, 05:22 PM   #17
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
Christer, dumb question here, if I may. Do the distortion artifacts you see change if you change the frequency resolution or make a small change in the excitation frequency? I wonder if this might be a bin effect.
__________________
"The way to deal with superstition is not to be polite to it, but to tackle it with all arms, and so rout it, cripple it, and make it forever infamous and ridiculous."- H. L. Mencken
  Reply With Quote
Old 11th March 2003, 09:49 PM   #18
diyAudio Member
 
Join Date: Sep 2002
Location: Sweden
Quote:
Originally posted by SY
Christer, dumb question here, if I may. Do the distortion artifacts you see change if you change the frequency resolution or make a small change in the excitation frequency? I wonder if this might be a bin effect.
Not dumb at all, but difficult. I have been spending a large part
of the evening trying to do some more empirics on this. I couldn't
quite repeat the behaviour I thought I had previously
observed. Maybe it was there under some conditions, or maybe
it wasn't quite what I remembered. Anyway, the spectrum
does show a lot of harmonics all the way from the fundamental
up to 20kHz, some of them of the same magnitude as the
2nd and 3rd overtones. What I thought I had seen, and maybe
did see, was some very-high-order overtones with much larger
magnitude than the 2nd and 3rd overtones, which would seem
not quite reasonable.

Whether or not this phenomenon actually did show up, I had
previously observed exactly this phenomenon when doing
FFT on Spice simulations. Since I couldn't repeat it with the
soundcard (after using the program for a few days, I have to
say there are quite a number of bugs and oddities, sometimes
making it cumbersome to use) I reverted to some experiments
with Spice. I have often used signals where the time period is
an even multiple of the simulation step length, which gives the
behaviour I mentioned. Hence, I tried to change the signal
frequency slightly, as you suggested, and that made an
enormeous difference. I made the experiments with a single
sine signal and a purely resistive load, in order to analyze
a single undistorted sine wave. That is, the only errors that
should show up in the spectrum are due to the simulation
step length and/or the FFT algorithm.

Using a 1kHz sinewave, simulating for 20 cycles with 1ns
steplength gives sharply defined peak at 1kHz with a very flat
noise floor just below -180dB, plus some low but sharply
defined peaks in the MHz area, the largest about 30dB above
the noise floor. When changing the frequency to 1001 Hz and
keeping the simulation parameters, there is no nice flat noise
floor, but a sloping one. However, the high-order peaks are
gone. Maybe the sensemorale is to use a signal period that is
not an even multiple of the simulation step length. I am not
sure if this is a phenomenon due to FFT, in which case it
can occur in measurements, or due to simulation step length,
in which case it can only occur in simulations. I attach two
spectras to illustrate this latter Spice experiment.
Attached Images
File Type: jpg fft_1001_20m_1n.jpg (30.7 KB, 463 views)
  Reply With Quote
Old 11th March 2003, 09:50 PM   #19
diyAudio Member
 
Join Date: Sep 2002
Location: Sweden
Hm, thought I could attach several images, but obviously not,
so here's the other one.
Attached Images
File Type: jpg fft_1k_20m_1n.jpg (33.1 KB, 445 views)
  Reply With Quote
Old 11th March 2003, 09:54 PM   #20
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
I think you have it nailed. This is a function of using a discrete FT.
__________________
"The way to deal with superstition is not to be polite to it, but to tackle it with all arms, and so rout it, cripple it, and make it forever infamous and ridiculous."- H. L. Mencken
  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
Visual Analyser & Nuova Elletronica LX.1690 KIT Oscilloscope + Spectrum Analyser fotios Solid State 25 9th March 2008 07:35 PM
Which Spectrum Analyser?? fmak Parts 25 1st April 2006 04:08 PM
HP 8590A Spectrum analyser A 8 Parts 3 4th April 2004 11:22 AM


New To Site? Need Help?

All times are GMT. The time now is 04:54 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