Sound Quality Vs. Measurements - Page 794 - 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 2nd January 2013, 05:36 PM   #7931
jcx is online now jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
well certainly "Feedforward Error Correction in Power Amplifiers" JAES 1980 discussion of combining Class D and linear amps is beyond patent time limits - now "prior art"
  Reply With Quote
Old 2nd January 2013, 05:41 PM   #7932
diyAudio Member
 
Join Date: Mar 2004
Location: At the output stage
Send a message via Yahoo to mr_push_pull
Quote:
Originally Posted by jcx View Post
Feedforward Error Correction in Power Amplifiers
so, I'm not wrong in thinking that this is what Devialet actually does?
__________________
we all love a good ol' stereotype until it's against us
  Reply With Quote
Old 2nd January 2013, 05:57 PM   #7933
jcx is online now jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
can't say, don't know what they're up to but the feedforward paper has interesting analysis of how combining power amplifier outputs can be made to work, with switch mode amp + linear concept explicitly described
  Reply With Quote
Old 2nd January 2013, 06:10 PM   #7934
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by mr_push_pull View Post
I didn't mean it that way, I'm highly skeptical when "revolutionary" inventions are mentioned, especially in the context of audio. actually, the mere mention of the word "technology" (followed by an acronym) in a review makes me rise an eyebrow. I mean, damn it, this is what marketing guys in corporations do, they invent/change names hoping that everyone will buy into thinking there's really something going on there LOL

I'd be interested in learning more about this, I've searched for the mentioned patent, thanks.

(this still leaves the question: why is the HF noise of the "conventional" D-amps to be dreaded? I haven't found any credible explanation yet)
It's not necessarily to be dreaded. But the noise in the Devialet is high within the audio band.
  Reply With Quote
Old 2nd January 2013, 07:11 PM   #7935
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Quote:
Originally Posted by bcarso View Post
It's not necessarily to be dreaded. But the noise in the Devialet is high within the audio band.
This is a very slippery domain, Brad.

In analog technology, one is torn between two factors: lower input stage bias is good for the S/N ratio, but higher bias produces better slew rates.

So one has to juggle between these two much of the time.

Both can be done, but to be done right, they cost money to be made to work really well in both fields.
__________________
Per Aspera Ad Astra.
  Reply With Quote
Old 2nd January 2013, 07:30 PM   #7936
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by dvv View Post
This is a very slippery domain, Brad.

In analog technology, one is torn between two factors: lower input stage bias is good for the S/N ratio, but higher bias produces better slew rates.

So one has to juggle between these two much of the time.

Both can be done, but to be done right, they cost money to be made to work really well in both fields.
Optimal bias with bipolars appropriate to the impedances of source and feedback divider is a place to start (the ratio of e sub n to i sub n is ideally the same as the source resistance). But then (as amply discussed already and in other threads) the transconductance can be too high for easy compensation of the loop, so we degenerate with emitter resistors, and add noise thusly. We can have the best of both worlds with strategically placed inductors, to roll off the HF gain but allow low resistance at audio frequencies. See Jensen and his predecessors/successors.

JFETs are almost free from input current noise, for typical impedances, and their e sub n drops with the 1/4 power of drain current. So to a point more current is better, and the available current for slewing is higher. But the transconductances are typically lower than bipolar stages, and the loop is usually easier to compensate against oscillation.

If we look at the Devialet measurements, as opposed to their non-referenced claim of 130dB signal-to-noise, there is something wrong IMO. And it's not due to a lack of money
  Reply With Quote
Old 2nd January 2013, 08:02 PM   #7937
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Brad, what I referred to was that damned, niggling little voice inside your head that says why settle for 40 V/uS, when you can do like 80 V/uS?

And you know you have to sacrifice something for doubling your slew rate, and still you can't say to yourself, oh heck, I don't NEED 80 V/uS ...

We're really much like gamblers, always trying to beat the odds.
__________________
Per Aspera Ad Astra.
  Reply With Quote
Old 3rd January 2013, 05:16 PM   #7938
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
I wonder if class D had been invented in 1937 and class AB last week would we be be even more impressed by AB than I am ? Nearly the efficiency of class D with the purity of class A ? So simple to implemented that it can be attempted by music lovers rather than engineers ? I suspect class D could have been made workable in 1937 with beam tetrodes ( 807 especailly , TT21 was a post war version for transmitters based on KT88 ) ? Earth to Mr Wave where are you , what do you know ? The LS 50 ( Gu , FU 50 ) ?
  Reply With Quote
Old 3rd January 2013, 08:07 PM   #7939
diyAudio Member
 
Join Date: Mar 2004
Location: At the output stage
Send a message via Yahoo to mr_push_pull
I'm personally interested in class D as it has a lot of advantages. no, it's not the efficiency, f**k the planet. joking, but it has other "practical" advantages, like smaller size (and weight), smaller heatsinks (weight again), can accommodate a smaller enclosure (weight again), if used with SMPS weight and size can be reduced again. and if all that adds to a lower price (including lower P&P due to lower size and weight) and we find a way to make them really good, why not? things change, I think it's unhealthy to just say "this is how they made the good stuff back in the day, I'll stick to that". if I were to resort to the over abused car analogy, everyone kept building long throw engines for a long time in a false belief that they're superior torque-wise. no-one makes them anymore, they're a relic of the past. but go back in time 40 years and tell that to the people who made them.

what surprises me is that expensive (as in really expensive) implementations begin to emerge. the 40k Mark Levinson, the 20k TAD. they are huge, heavy and look more like class A behemoths. still, they keep getting not-so-enthusiastic reviews. my only interaction with class D is my current, UCD-based amp and a very short listen to a Jeff Rowland. if I were to comment on my amp's sound, I'd say it somehow correlates with what most reviews describe.

but why? what is inherently inferior with class D? what makes the Devialet better (if it really is better)?
__________________
we all love a good ol' stereotype until it's against us

Last edited by mr_push_pull; 3rd January 2013 at 08:28 PM.
  Reply With Quote
Old 4th January 2013, 06:56 AM   #7940
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Not sure about your car engine analogy? What is a "long throw" engine? One that has a stroke value bigger than its diameter value, as say 80x90 mm? That would be an undersquare engine, while a say FIAT 1.3 litre, 86x55.5 mm would be an oversquare engine, since its diameter is bigger than its stroke. If so, then you have it upside down, such oversqaure engines were all the rage 40 years ago, while today, most are even extremely oversquare, e.g. Honda's 2.4 litre engine has a bore of 88 mm and a stroke of 98 mm or some such.

I also wondered, so I talked to some friends active in the German auto industry (my old time friend works in Bosch, refining electronic fuel injection systema, for example) and was told that long stroke engines were easier to cool and to be made compliant with the Euro emission norms. With modern alloys, they can also be made to last very long, and with the introduction of compressors and turbo chargers to large series small and cheap cars, the issue of revs was off the table anyway. These days they get 100 bhp per litre specific power below the 5,000 rpm line, using turbos. By comparison, look at the just discontinued Renault Clio RS, it also had a 2 litre engine, it did produce 204 bhp (or 102 bhp per litre), but at 7,400 rpm, yet even so, its emission figures cannot compare with a standatd say VW 2 litre trubo blown engine, and its torque peaked at 201 Nm, whereas the VW engine has a torque of 350 Nm.

Last but not least, turbocharged low revving engines do much better using alternatve fuels, such as natural gas, alternative versions of diesel fuel, etc.

In short, the tables were turned by no whim of the manufacturers, but by ever more demanding emission and fuel efficiency regulation. An oversquare engie can also do it, but at a higher price point, so the industry simply took the easier way out, as it always will.

My own experience with digital amps is also very modest, 5 in all so far, none of them wildly expensive.

To dispense with the price issue - it varies from economy to economy, but locally, for the price of a digital amp, one can do an analog one for the same money or just a bit more, so their price starts to become interesting only after a certain power point, where they take the lead. That would be, off hand, say 150 WRMS or so. Not many people are into that kind of power, most find that say 50 WRMS is enough for them.

On the technical side, not one of the amps I did hear would truly satisfy even analog build criteria. Their power transformers were on the small side, as were their capacitors. I actually talked one of the owners into doubling his filter capacitor capabilities by simply adding two more of the same per channel (he had, as per the instruction manual, 2x10,000 uF per channel for a nominal 100 WRMS/8 Ohms output, so I made him add another pair of capacitors per channel, thus doubling his capacitance), and the result was audibly better. No revelation, but it did play better tunes.

My general feeling was that all of them were built by the book and measuring gear, not by audtioning.

And they were definitely all made to a price, the main sales pitch being "here's a lot of clean power for not a lot of money". The moment you started to add bits and pieces, such as reference voltages done as they should be, it suddenly stopped being cheap and started costing serious money. The price increased much faster than the audible benefits. They had great bass, solid midrange, but their treble always gave them away. You could never hear brass being struck by wood, it always came across as a sound you could not make out how it was made.

So, in my view, good, promising, but as yet, not convincing. I wouldn't be surprised if it turned out that no matter what the maths say, the output devices needed to be faster still, well into microsecond range, nanoseconds don't seem to cut it.
__________________
Per Aspera Ad Astra.
  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 12:02 AM
Sound Card for Measurements Marik Solid State 2 2nd January 2012 08:59 PM
Sound Card Recommendations (For Audio Measurements) dchisholm Equipment & Tools 5 16th July 2011 09:40 AM
How to protect sound card during amp measurements? okapi Everything Else 13 2nd September 2008 03:06 PM
Sound cards - test and measurements jackinnj Everything Else 2 5th July 2003 03:02 PM


New To Site? Need Help?

All times are GMT. The time now is 06:51 PM.


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