Listening Test... A Practical Demonstration That Amplifiers Have Their Own Sound. - Page 12 - diyAudio
Go Back   Home > Forums > General Interest > Everything Else

Everything Else Anything related to audio / video / electronics etc) BUT remember- we have many new forums where your thread may now fit! .... Parts, Equipment & Tools, Construction Tips, Software 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 26th February 2014, 10:26 AM   #111
diyAudio Member
 
Max Headroom's Avatar
 
Join Date: Jan 2013
Location: West Australia, straight over the road from the beach, natural ambient sounds only.
Hi Pavel, are you testing two seperated monobloc amplifiers, or a stereo power amp.

Dan.
  Reply With Quote
Old 26th February 2014, 10:31 AM   #112
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
It is a stereo power amplifier, but it has dual mono concept and two channels are totally separated with separated power supplies. Channel separation is excellent, as you may see in the attached measured plot.
Attached Images
File Type: png channel_separation PA4.PNG (21.8 KB, 116 views)
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote
Old 26th February 2014, 11:39 AM   #113
diyAudio Member
 
Max Headroom's Avatar
 
Join Date: Jan 2013
Location: West Australia, straight over the road from the beach, natural ambient sounds only.
Quote:
Originally Posted by PMA View Post
It is a stereo power amplifier, but it has dual mono concept and two channels are totally separated with separated power supplies. Channel separation is excellent, as you may see in the attached measured plot.
Ok, so that is not quite the same as a typical stereo amp, hence your earth common connection.
Yes, you ought to be pretty happy with that degree of channel isolation.

Dan.
  Reply With Quote
Old 26th February 2014, 11:58 AM   #114
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
Power supply inter-channel modulation is thus highly suppressed.

To me, it is a 'typical' stereo amp . I have abandoned common power supply concept years ago, as it compromised both parameters and sound.
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote
Old 26th February 2014, 01:05 PM   #115
diyAudio Member
 
Max Headroom's Avatar
 
Join Date: Jan 2013
Location: West Australia, straight over the road from the beach, natural ambient sounds only.
Quote:
Originally Posted by PMA View Post
Power supply inter-channel modulation is thus highly suppressed.

To me, it is a 'typical' stereo amp . I have abandoned common power supply concept years ago, as it compromised both parameters and sound.
Agreed, and of course.
I have tried 4 transformers (1 per each rail) and infinite earth common (sizable copper sheet) and the result was quite spectacularly good in many respects, notably centrality and stability of mono content, 'índependance' of stereo content, and portrayal of depth information, including sounds from behind the listener.
Absolute polarity also became, errr absolute.

I find there is distinct (coherence) value in combing two channels correctly into one enclosure for unbalanced input connected systems.

Dan.
  Reply With Quote
Old 26th February 2014, 09:30 PM   #116
fas42 is online now fas42  Australia
diyAudio Member
 
fas42's Avatar
 
Join Date: Jun 2012
Location: NSW, Australia
Blog Entries: 11
Dan, interested in your experiments in creating a sizable local earth - this is something I've thought about, but haven't played with. What sizes of sheet are we talking about, and did you determine any relationships or optimum size for the material?

Thanks,
__________________
Frank . . . the truth is, I just like a bit of ASMR ...
  Reply With Quote
Old 27th February 2014, 06:57 AM   #117
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
Even with a 'sizable local earth' you need to consider and make a proper analysis where the ground currents flow. Otherwise a 'sizable local earth' would bring no advantage and would not exclude possible hum/buzz.
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote
Old 27th February 2014, 09:21 AM   #118
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
Quote:
Originally Posted by PMA View Post
Attached you will find a test schematics. 1A current was forced from left channel of the amplifier to its right channel via 6.8 resistor. FFT of the forced channel has been performed, as well as the FFT of the driving channel output.
You motivated me to do more tests !

I tried current injection but this would need at least a 8 ohm resistor, and a rather large voltage on the driving channel, to swamp its distortions. However this means I have to turn the volume knob... and then the noise on the channel under test goes up.

So, I have simply used a 0.5 ohm load resistor on one amplifier channel. The other channel is not used. The amplifier under test is driving a low voltage (up to 0.5V RMS) in this resistor, and I am simply measuring the output.

I have not used FFT, that would be too easy (and would hide too much stuff).

The test signal is the sum of :
- a low frequency (LF) sinewave around 52.73Hz
- a high frequency (HF) sinewave at 16 kHz

The test program plays the signal, records the soundcard output, then detects the HF amplitude, and plots its variations according to the LF amplitude.

I am using synchronous detection on the HF, so each HF period gives one amplitude sample. This completely removes the HF from the results data. So it is better if both frequencies are completely unrelated, as this spreads the detected samples along the LF period. Also the funny LF frequency value is due to the fact that an integer number of periods have to fit into 16384 samples, because I was lazy with the soundcard code.

It also averages lots of periods (one test takes 30 seconds) because the level of noise on this soundcard is really over the top.

Variations of amplifier output impedance translate directly into HF signal amplitude variations. The LF signal sets the amplifier output stage current.

Here is a first, simple graph.

First, the legends : "1000/200mA" indicates 1000mA of LF current, plus 200mA of HF current. Remember this for the next graphs.

Top Left Plot : this is the amplifier output, over one LF period.

Bottom Left Plot : HF amplitude, synchronized with the time axis of the plot above.

Since the load is resistive, HF amplitude is proportional to output Gm (and inversely proportional to output Z).

I have displayed amplitude in % variation versus average. So, it is equivalent in % Gm variations versus average Gm.

So, we see :

- Low absolute value of current : this is inside crossover region, both output transistors work, gm-doubling occurs. HF amplitude is higher.

- High absolute value of current : only one transistor works, so gm is smaller. HF amplitude is smaller. The fact that the PNP and NPN outputs have different characteristics is clearly visible as the "valleys" in the curve are different depending on output polarity. The top transistor seems to have 2x the strength of the bottom one.

Top Right Plot : Same data, different view. Instead of time, the X-axis is low frequency current value.

Here, the Gm curve is characteristic of a little overbiased Class-AB output stage, which.

Bottom Right Plot :

Same as above, but there are two Gm plots, from the start of the test, with cold transistors, and from the end of the test, with warm transistors. At the beginning, the output stage shows under-biasing, then it warms up and becomes over-biased.

Remarks :

Global feedback is active, and there is still enough of it remaining at 16kHz to make the Gm variations very small. Still, the real behaviour of the output stage (open-loop), is clearly visible.

The green and blue curves use different values for the HF current, this does not seem to matter a lot here. See later for more on that.
Attached Images
File Type: png 1000mA-1.png (364.6 KB, 57 views)
  Reply With Quote
Old 27th February 2014, 09:27 AM   #119
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
Do you have a schematics of the amplifier under test, even a blind one or simplified?
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote
Old 27th February 2014, 09:35 AM   #120
peufeu is offline peufeu  France
diyAudio Member
 
Join Date: Mar 2001
Location: Lyon, France
It could be interesting to investigate self-heating of the output transistors.

Here, I plotted a snapshot of Gm versus current, over time, so it is a kind of 3D view.

Each curve is an average over 0.7 seconds. I spaced them vertically for clarity.

The first curve is at the top. It shows under-biasing, classic V-shape.
The next curve, directly below, is much flatter.
Then as time passes, transistors heat, over-biasing occurs, inverted V-shape appears.
It stays stable like this until the end of the test.

So, we can see this amplifier has a thermal time constant around 2-3 seconds for the output stage. It is an integrated amp (big hybrid circuit, japanese 80's style).

During quiet music passages, its natural bias is a bit underbiased. When it gets louder, after a few seconds it will heat up and overbias. When it gets quiet again, I suspect several seconds of severe underbias could occur. I need to fix the software to find out.

BTW, this clearly shows that measuring output Gm at DC with a slow data acquisition system is completely useless... just like most tests using stationary signals, like THD, etc., which measure the performance on stationary, constant amplitude signals, not on real music signals, which vary a lot. But I guess everyone knew that
Attached Images
File Type: png 1000mA-1-heating.png (203.0 KB, 54 views)
  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
Ultimate listening test ... test your ears and audio chain PMA Everything Else 378 21st November 2013 10:20 AM
Blind listening test with Orions, cheap pro sound and even cheaper DIY Godzilla Full Range 16 19th November 2010 07:10 PM
Project : Practical Class-D PWM Subwoofer Amplifiers with optocoupled Gate Drivers Workhorse Class D 9 20th January 2004 11:21 AM


New To Site? Need Help?

All times are GMT. The time now is 02:17 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