Sound Quality Vs. Measurements - Page 441 - diyAudio
Go Back   Home > Forums > Member Areas > The Lounge

The Lounge A place to talk about almost anything but politics and religion.

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 17th April 2012, 11:20 PM   #4401
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
Difference in MODEL performance. What do you think this tells me with respect to my wife's critical hearing? I have not succeeded in making one behave like the other.
Attached Images
File Type: png 120.png (84.0 KB, 116 views)
File Type: png 951.png (72.2 KB, 115 views)
  Reply With Quote
Old 18th April 2012, 01:14 AM   #4402
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
The next thing I noticed is how much distortion in the model the output filter causes. But, when one adds some semblance of a load, without it you get instability and other such bad things. This leads me to a couple of conclusions:
The amp and the speaker must be designed together as a system. Output filtering and compensation taken as one.
Moving to active crossovers to eliminate the inductors in the crossover may do more for the result than this week's fancy VAS.
What I don't know: How bad is the distortion in an active crossover?

This is one of those things when you look at it you have to wonder how it works at all. My hat's off to those who have actually made it pretty darn good.
  Reply With Quote
Old 18th April 2012, 01:39 AM   #4403
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
One more couple of conclusions: use output stage with fast short local feedback to avoid troubles caused by added to already complex semi-stable system some more resonant thingies. Add plain resistor between them to decerease Q a little bit.
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 18th April 2012, 04:27 AM   #4404
benb is offline benb  United States
diyAudio Member
 
Join Date: Apr 2010
That makes me think, the output stage is (depending on design, but is very often) essentially a buffer, a current gain stage. Why not take the global feedback point from the INPUT of the output stage, instead of the actual output stage? This would surely isolate the feedback path from any variations in speaker impedance.

On the other hand, there's got to be something "wrong" with the idea, because I've never seen it done.
  Reply With Quote
Old 18th April 2012, 05:45 AM   #4405
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Quote:
Originally Posted by benb View Post
That makes me think, the output stage is (depending on design, but is very often) essentially a buffer, a current gain stage. Why not take the global feedback point from the INPUT of the output stage, instead of the actual output stage? This would surely isolate the feedback path from any variations in speaker impedance.

On the other hand, there's got to be something "wrong" with the idea, because I've never seen it done.
Nothing is wrong with this idea. And it sounds much better than with single global feedback loop.
Well, may be "wrong" is the result of measurements that designers and manufacturers want to show in order to get proper sales figures. But boutique designers don't need that, so they use it quite often without any odd noise. Except "No global feedback", or "low global feedback" phrase. That means "high" feedback may exist, but outside of "global" loop.
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 18th April 2012, 05:52 AM   #4406
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 109
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Quote:
Originally Posted by tvrgeek View Post
Moving to active crossovers to eliminate the inductors in the crossover may do more for the result than this week's fancy VAS.
What I don't know: How bad is the distortion in an active crossover?
Make the active crossover a digital one (and FIR to bypass any possibility of limit cycles and hence noise modulation, or use double precision IIR) and the distortion's just that of the DAC. Pick your DAC poison carefully
__________________
Seek not the favour of the multitude...rather the testimony of few. And number not voices, but weigh them. - Kant
The capacity for impartial observation is commonly called 'cynicism' by those who lack it.
  Reply With Quote
Old 18th April 2012, 06:07 AM   #4407
Previously: Kuei Yang Wang
 
Join Date: Nov 2002
Hi,

Quote:
Originally Posted by tvrgeek View Post
LL? I understand toroid, R core, IE core, and double C. I thought one of the quite experienced designers here was advocating R core.
There are many forms of mains transformers. The eastern block was big on double-C-Core and LL.

These are in my limited experience (I only tested a few samples of each type) the best choice for audio mains transformers, if correctly wound. Best in fact are 400Hz double-C-cores (made for Ships, Planes, military etc.) which need to be de-rated for 50Hz service, with a tiny airgap.

These are then wound using the knowledge of how audio transformers are wound to generate transformers with a defined, narrow passband (including some passives build into the transformer) and very large common mode rejection ratio all the way to very high frequencies.

It does cost more, but if you buy 100's or 1,000's of pieces it is normally a small price increase and no problem, though the transformer makers will think you crazy applying so much effort for mains transformers, they all are happy to take your money.

Torroids are by far the worst solution to audio mains transformers, though if you know and understand what their problems are you may be able to compensate for them using external means and at least mitigate some of the issues. Again, I see this extremely rarely implemented in any audio gear and given the preponderance of badly implemented torrid torroids in audio gear it is little surprise that the mains filter industry is booming.

In fact, the whole mains powering implemented without thorough understanding of what goes on (and hence creating problems) accounts probably for 70% or more of the sonic differences reported for interconnects and mains cables as well as other mains related items and surprisingly it remains "unmeasurable" as the meter readers often are as clueless as the high end designers of what is going and in absence of measured differences and the presence of claimed sonic differences you get the usual fight...

To a real pro neither fact is surprising and they are trivial to reconcile, so we usually look at the hissy fits and general spats that break out as soon as someone mentions sonic differences caused by cables and wire.

Ciao T
  Reply With Quote
Old 18th April 2012, 06:36 AM   #4408
Previously: Kuei Yang Wang
 
Join Date: Nov 2002
Hi,

Quote:
Originally Posted by tvrgeek View Post
Difference in MODEL performance. What do you think this tells me with respect to my wife's critical hearing? I have not succeeded in making one behave like the other.
Without seeing the Schematics I cannot comment how to change one into the other. What does interest me though, which one is it that your wife prefers? Am I correct that it is the second one?

Quote:
Originally Posted by tvrgeek View Post
The next thing I noticed is how much distortion in the model the output filter causes.
Then something is not right. The output filter should not add distortion. If it does, it means you introduced an unfortunate interaction between it and the amplifier and load.

Quote:
Originally Posted by tvrgeek View Post
This leads me to a couple of conclusions:
The amp and the speaker must be designed together as a system.
It helps, but rarely feasible.

Quote:
Originally Posted by tvrgeek View Post
Output filtering and compensation taken as one.
So much should be obvious from at the stability requirements of feedback amp's.

To make this more accessible conceptually, IMNSHO the compensation should make the amplifier stable with resistive loads, the build-out network should resist RF ingress and decouple the Amplifier sufficiently from the load to ensure unconditional stability as system.

Quote:
Originally Posted by tvrgeek View Post
Moving to active crossovers to eliminate the inductors in the crossover may do more for the result than this week's fancy VAS.
Maybe, maybe not.

Quote:
Originally Posted by tvrgeek View Post
What I don't know: How bad is the distortion in an active crossover?
Invariably worse than that of a passive one. That much should be abvious, assuming quality components are used for the passive crossover (so not ferrite cored chokes and bipolar electrolytics).

Quote:
Originally Posted by benb View Post
That makes me think, the output stage is (depending on design, but is very often) essentially a buffer, a current gain stage. Why not take the global feedback point from the INPUT of the output stage, instead of the actual output stage? This would surely isolate the feedback path from any variations in speaker impedance.
Sure, but as in a well designed amplifier the output stage should be the majority contributor to non-linearity (if it is not, linearise the other stages first by using degeneration or local loops and make them more linear), why not drop feedback alltogether (hint - because our output is in class AB, that is why we cannot drop looped feedback - make a true class A Amp and run feedbackless is trivial).

Quote:
Originally Posted by benb View Post
On the other hand, there's got to be something "wrong" with the idea, because I've never seen it done.
It actually has been done. Krell comes to mind, they had a range of amplifiers with dynamic biasing that kept the output stage in Class A and no looped feedback from the output stage (but local loops from the VAS out).

The Cello Encore Amplifier (and AFAICT others from Cello) used no looped NFB but a variant of hawkesford error correction (which actually is fundamentally again NFB) at low frequencies and a complex set of networks to add some feedback at higher frequencies, keeping Class AB distortion contained.

In the DIY scene the LC Audio Zapsolute and Millenium are examples of Amplifiers that use only local degeneration and no looped feedback, commercially the Etalon Amplifiers from Hungary come to mind.

Former DIYA contributor "Jocko Homo" used to remove the global feedback loop to the input stage from his amp's and instead used a feedback loop around output stage and VAS (which was supposedly excellent sounding) and got promptly crucified by some lippy sh...t at the JAES when he tried to share some of those ideas...

There are other topologies that have been running without looped feedback too. For a really freaky one look up S.P.L.I.F. (I need to that Charles is a sly old Fox - if he was American he would named his idea J.O.I.N.T.).

Generally the price you pay for these topologies is greater measured THD, so Members of the holy Catholic Church of the Meter Readers and Double blinders (which are out in force at thishere Venue and show little respect for differing views) will promptly declare you a Heretic and set the Holy Inquisition on you should you be found to promote such circuitry...

As an aside, these low or no feedback around the output stage topologies usually benefit from "overbias", as this reduces the higher order components, so the distortion result is more akin to a good tube Amp (note, it does not make an SS Amp sound like a tube Amp)...

Ciao T
  Reply With Quote
Old 18th April 2012, 06:41 AM   #4409
Previously: Kuei Yang Wang
 
Join Date: Nov 2002
Hi,

Quote:
Originally Posted by abraxalito View Post
Make the active crossover a digital one (and FIR to bypass any possibility of limit cycles and hence noise modulation, or use double precision IIR) and the distortion's just that of the DAC. Pick your DAC poison carefully
In my limited Experience FIR filters are inimical to perceived "good sound". Correctly done IIR filters behave the same as classic passive/analog ones.

So the only benefit of digital crossovers is ease of adjustment including compensation for non-ideal drive units in speakers.

This is what argues strongly in favour of such units for DIY, if the digital crossover can be of sufficiently high quality, but less so for commercial designs where the resources exist to optimise passive system.

Ciao T
  Reply With Quote
Old 18th April 2012, 06:48 AM   #4410
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 109
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Quote:
Originally Posted by ThorstenL View Post
In my limited Experience FIR filters are inimical to perceived "good sound".
Perhaps the FIR filters you've experienced haven't been 'correctly done' ? Care to share more details?

Quote:
So the only benefit of digital crossovers is ease of adjustment including compensation for non-ideal drive units in speakers.
You use 'so' but I can't see how this (above) statement follows reasonably from your earlier ones.
__________________
Seek not the favour of the multitude...rather the testimony of few. And number not voices, but weigh them. - Kant
The capacity for impartial observation is commonly called 'cynicism' by those who lack it.
  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
Quality Control differences = variations in sound quality? KT Class D 3 4th June 2014 01:02 AM
Sound Card for Measurements Marik Solid State 2 2nd January 2012 09:59 PM
Sound Card Recommendations (For Audio Measurements) dchisholm Equipment & Tools 5 16th July 2011 10:40 AM
How to protect sound card during amp measurements? okapi Everything Else 13 2nd September 2008 04:06 PM
Sound cards - test and measurements jackinnj Everything Else 2 5th July 2003 04:02 PM


New To Site? Need Help?

All times are GMT. The time now is 07:25 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