Behringer DCX2496 digital X-over - Page 238 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Line Level

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

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 16th January 2010, 08:29 PM   #2371
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Quote:
Originally Posted by panomaniac View Post
Thanks for the measurement Michael.
.
You are welcome


Quote:
Originally Posted by panomaniac View Post
.
I wish I could get Steps to work for me. I always get glitches, but never at the same place. This results in nasty distortion spikes. No other software does that on my setup. Might have to try a different computer.
.
Have you checked
- buffer size ?
- sampling rate being the same for STEPS and your soundcard ?

Michael

Last edited by mige0; 16th January 2010 at 08:37 PM.
  Reply With Quote
Old 17th January 2010, 12:43 PM   #2372
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Back again – found a small mistake in one measurement and hence deleted former posting

########



Here is another interesting measurements of a ~ 1:6 small signal transformer made by WSW, I had at hand.

This is a rare piece of ancient Austrian PRO audio gear, manufactured by Vienna's "Wiener Schwachstrom Werke" - later on, taken over by Siemens Kangfilm to my knowledge.
http://www.tube-classics.de/TC/Klang...ilmhistory.htm

This transformer is roughly double the size of the previous one :
22mm x 28mm x 34mm (versus 20mm x 24mm x 24mm)


Anyway – lets have a look – it performs quite different :

Fed by 1Vrms:

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.


Fed by 0.1Vrms

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.


It is interesting that this one, compared to the former one, seems to perform way better in the lower frequency department.
Fed by 100mV, distortion is down to ~ 24bit territory, up form ~200Hz – not that bad !


Though this small signal transformer clearly too isn't suitable for 1Vrms line levels, it already performs quite well at 100mV level – or even slightly higher.
Frequency response seems to be excellent for both transformers IMO.

Wonder how different core materials behave in the lower and upper FR range, so I am looking forward to the measurements of better known (and available) device.



Michael
  Reply With Quote
Old 17th January 2010, 01:27 PM   #2373
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Cool work. I'm a little unclear on the graphs, which one is what level. But I do see the same sorts of things I've been measuring. Smaller transformers have more 3rd harmonic in the low end. And of course that rises with signal level.

That is an advantage that opamps seem to have - less distortion in the low end than most transformers. From 1K up is another story.

I'll keep working with Steps- maybe buffer settings will do it.
__________________
Take the Speaker Voltage Test!
  Reply With Quote
Old 17th January 2010, 01:36 PM   #2374
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Here's a couple of graphs real quick to show the differences in distortion between the stock DCX and the modified with transformer only output.
Attached Images
File Type: png 5K-8K-dcx.png (11.1 KB, 124 views)
File Type: png 5K-8K-transfo.png (11.2 KB, 121 views)
__________________
Take the Speaker Voltage Test!
  Reply With Quote
Old 17th January 2010, 02:48 PM   #2375
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
looks like the dcx+TX is about 25dB less distortion than a standard dcx.
__________________
regards Andrew T.
  Reply With Quote
Old 17th January 2010, 06:12 PM   #2376
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Yes Andrew, pretty much. At least at the top end. But the stock output has less distortion than the transformers in the low end of the spectrum. The transformers I tested have a 3rd harmonic that rises at lower frequencies. You will see that in some of the graphs below.

I'm curious about this 3rd order harmonic, because I measured a number of transformers last year and these showed very little distortion rise at the bottom end. And I don't know why the difference. I'll have to figure it out.

These plots show how the DCX stock output has a distortion that rises with frequency and the transformer has a distortion that drops with frequency. To my ear, this must be why I feel that the transformers remove a lot of digital glare or brittleness that I don't like. The swept FR is so close between these two there is no point to post it.

I'll try to figure out why I'm now seeing H3 distortion at the bottom end with the transformers that I did not measure last year.
Attached Images
File Type: png multitone-DCX.png (11.6 KB, 107 views)
File Type: png multitone-transfo.png (12.0 KB, 74 views)
File Type: png IMD-sweep.png (6.6 KB, 80 views)
File Type: png Intermod.png (15.3 KB, 77 views)
__________________
Take the Speaker Voltage Test!
  Reply With Quote
Old 17th January 2010, 06:40 PM   #2377
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Quote:
Originally Posted by panomaniac View Post
Cool work. I'm a little unclear on the graphs, which one is what level..
Thanks.
Concerning plot level, simply refer to the schematic above each plot. You see "the chain of signal flow" from left to right (no connections shown - but markers instead). Each plot is correctly calibrated.

Basically the soundcard output was (digitally) attenuated to get the 1V rms at first hand – which in all measurements is kept constant.

To perform measurements at attenuated levels (–20dB) I simply inserted a differential voltage divider that keeps its source impedance the same 100 Ohm as the soundcard's output.
If you cascade that differential voltage divider you get attenuation of –40dB (well roughly).

The setup limits are always shown first by a simple loop back measurement (no DUT inserted) to confirm validity.

Measurement is performed by processing the transformers output through a line receiver / line driver pair – and then feeding the mic-in of the soundcard.
For measurement of –20dB (100mVrms at transformer-IN) the mic gain was turned up by ~20dB - for measurement of –40dB (10mVrms at transformer-IN) the mic gain was turned up by ~40dB
This somehow complex procedure provided the cleanest and most consistent results over the wide range of levels.
At any change of mic-gain, STEPS was re-calibrated of course.



Quote:
Originally Posted by panomaniac View Post
I'll keep working with Steps- maybe buffer settings will do it.

One other thought – in case you have a setting for "internal" or "external" clock – check "internal"

Michael

Last edited by mige0; 17th January 2010 at 06:59 PM.
  Reply With Quote
Old 17th January 2010, 06:42 PM   #2378
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Pano, these dual tone and multi-tone measurements are very revealing ! - thanks for showing.
I have to test that for my soundcard too!

Michael
  Reply With Quote
Old 17th January 2010, 06:57 PM   #2379
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Quick question:
- What does the third plot from left show us ?
- which level actually is "0dB full scale" in your measurements - is it full out of DCX (no digital or analog attenuation)??

Michael

Last edited by mige0; 17th January 2010 at 07:07 PM.
  Reply With Quote
Old 17th January 2010, 07:19 PM   #2380
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Hey Michael.
That 3rd chart is the IMD sweep as done by RightMark. I don't know much about the measurement, the RMAA doc does tell me much. The far right chart is 60Hz and 7KHz IMD.

All the measurements in ARTA were done with the digital signal at -6dBFS. That was to prevent overdriving anything. I did my best to adjust the soundcard inputs to -6dBFS as well. So a sort of "digital unity gain." Obviously the stock DCX outputs had to be (analog) attenuated much more than the simple transformer outputs to reach that level. Similar to what you are doing.

So the output of the DCX was never full scale, but -6dB. Driven with a digital input only.
Not really sure why ARTA places the noise floor below -120dB. That should not be possible working in 16bits, as I was. RightMark tells me the noise floor is about -94dB.
__________________
Take the Speaker Voltage Test!
  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



New To Site? Need Help?

All times are GMT. The time now is 02:47 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