Yes, i believe the shown design is such a lowpower embedded design, which Eva also mentioned regarding applications of filterless class D. Just wondered how the idea would/could translate into something usable as a stand alone amp -especially with a usable degree of class A -like operation
Of course it makes sense to invest into the most important recource of an audio chain: the current. I noticed that so often. For example a TC2000/TP2050 Tripath combo sounds with an 10 ampere expensive industrial SMPS much better than a cheap notebook SMPS. Less noise and more relax/authority in the sound.
I think it depends on the speaker´s character. I would imho suggest or test myself UCD with difficult to drive speakers. As I said before: a musical fidelity A1 was 20 years ago a bestseller with over 100.000 pieces sold. Nobody can explain the phenomenon why an amp sounds for a big crowd so good. As Kevin Voecks explained for speakers and this also is valid for amps: from bad measurements you can conclude to a bad sound but the reverse "from good measurements to a good sound" does not strictly work.
Sound travels at approx 345 meters per second in air. When you are listening 2 meters away from the speakers what you hear is not the original signal, it's the original signal with 6 milliseconds of group delay across the audio band (plus other components). Filters (and amplifiers) behave almost in the same way across their pass band. Speakers are much worse because their group delay is not constant.
Is that distortion? Nerd thinking.
Signal does not propagate infinitely fast. Everything adds delay/phase shift.
distortion and hearing
Yes, every deviation from the original* signal is distortion.
*original though is a construct as every link in the audio chain is a construct necessary to transfer (transform) the signal from a mechanical groove or surface inscriptions to a eletrical domain (then some times A/D to D/A) before the electrical chain ends in electroacoustic transformation mechanically connected to the inner ear. Every piece off copper, air, metal or semiconductor is a source of distortion though also a link, hence an important link. Funny though that the ear seems like an active perceptive device which attention and interpretative abilities can detect even unmeasurabel deviations orders of magnitudes lower than other sources of distortion (The acoustic environment is a distortional nightmare though often less intrusive than certain eletrical distortion artifacts...)
food for thoughts
Yes, every deviation from the original* signal is distortion.
*original though is a construct as every link in the audio chain is a construct necessary to transfer (transform) the signal from a mechanical groove or surface inscriptions to a eletrical domain (then some times A/D to D/A) before the electrical chain ends in electroacoustic transformation mechanically connected to the inner ear. Every piece off copper, air, metal or semiconductor is a source of distortion though also a link, hence an important link. Funny though that the ear seems like an active perceptive device which attention and interpretative abilities can detect even unmeasurabel deviations orders of magnitudes lower than other sources of distortion (The acoustic environment is a distortional nightmare though often less intrusive than certain eletrical distortion artifacts...)
food for thoughts
Thedealer,
Yes the only true meaningful way to know is testing in my own context of implementation. According to speakers, the "trouble" is that I´m constantly developing new speakers, so the amp should as the UCD claims to be, able to perform consistant under different loads. I usually get 3 to 4 designs through pr year, though all feature high (relatively) sensitivity, low electrical and acoustical compression and energy build, they do vary in impedence and are very sensitive to upstream changes. An amp to ad not subtract as little as possible would be a very nice working tool
regards,
O.K. you mean his work in the past should be relatively regarded in comparison to the rest.
But nowadays I cannot imagine that class A/B amps in the majority are better than in the past.
The market has it´s own rules and idealism has nothing to do here. On the other hand I see 2 main
facts which should mark the future of consumer electronics.
- the nowadays consumer behaviour which supports more audiovideo or homecinema gear.
- ecological aspects (energy crisis) which should open more and more the way for class D.
The newer generation (no matter which class) of amps will (must) sound better than in the past.
In times of www you cannot make big profits with crap. The world shrunked to a village where messages go around very quick there ...
I keep seeing it brought up, but IMHO energy availablity has absolutely *nothing* to do with Class D. There is no evidence to even seriously consider the possibility. Class D will thrive in the general marketplace simply due to economics. Cheaper to make and cheaper to buy. "It's all about the money." Not fossil fuels and renewable energy sources.
If it was a factor, then low-power output would also overtake high-power output, and that certainly isn't happening.
If it was a factor, then low-power output would also overtake high-power output, and that certainly isn't happening.
Yes, i believe the shown design is such a lowpower embedded design, which Eva also mentioned regarding applications of filterless class D. Just wondered how the idea would/could translate into something usable as a stand alone amp -especially with a usable degree of class A -like operation
seems unlikely. Take it as a challenge! nothing is impossible ;-)
I looked up that part. They guarantee to meet EMI spec with up to 100mm of speaker lead. (4")
Their use of spread-spectrum clock certainly helps lower the EMI at any one point, I wonder much could it be spread?With current tech, you could maybe make a nice set of powered speakers. Put the amp in the cabinet and tune it to the speakers.
I do not agree you cannot hear the difference, i have both formats, the difference is staggering.
back to Class-D
You cant tell people what they can and cant hear.
I cant tell the difference between class A, AB and D or the two cd formats.
In one way that is an advantage, it means i can buy the chepaest one and be happy with it.
Hi,
yes, no produces thd,because each component has the same delay, referred to a point of arrival.
This is because the speed of propagation does not depend on the characteristics of the sound, but rather the characteristics of those who propagate. eg. its speed in air is 340mt/Sec But at 20°. If we change the temperature or humidity, change the speed.
Taking advantage of this behavior in a speaker, is that you can amplify or attenuate certain precise frequencies. eg. in a two-way speaker, if we move the tweeter in axis (forward / reverse), relative to axis of the bass, can attenuate or amplify the middle frequencies. great way to keep the high cut of the tweeter filter (it works clean and does not break).
(I think that reason of "amplify or att. is simple) no?
Regards
Roberto
Last edited:
OT clean up.I enjoyed reading this thread, especially as I learnt a bit more about class D along the way. As I see it, advances need commercial imperatives and the only major commercial imperative with amplifiers is with class D. Class D is already good and will continue to improve. It's a bit like the golden age of turntables; the research money is in line with the demand and TTs reached their zenith in the eighties. Demand dropped off and the advances came to a big slowdown. It's not to say some companies don't continue to work on TTs, Class A etc but you need to go where the money is and low priced, highly efficient and small Class D amps have the consumer demand spotlight on them.
Eva, one day it would be good to know what amps you have designed so I can compare with my other class D amps
Eva, one day it would be good to know what amps you have designed so I can compare with my other class D amps
Eva, one day it would be good to know what amps you have designed so I can compare with my other class D amps
+1
Juhleren,
When I follow your link of #213 I see some Spanish webiste? with class D boards based on Philips TDA8920 and TDA8950.
Years ago I built a TDA8920 amp with a very good linear supply. This amp was compared with some Hypex UCD amplifiers (UCD180 and 400). At that time the Philips amp was found to sound better than the Hypex amps, more coherent with more live and a better sense of rythm. It was before UCD offered an other input IC and improved power supply. I wonder how they would compare now.
When I follow your link of #213 I see some Spanish webiste? with class D boards based on Philips TDA8920 and TDA8950.
Years ago I built a TDA8920 amp with a very good linear supply. This amp was compared with some Hypex UCD amplifiers (UCD180 and 400). At that time the Philips amp was found to sound better than the Hypex amps, more coherent with more live and a better sense of rythm. It was before UCD offered an other input IC and improved power supply. I wonder how they would compare now.
For what it's worth, one of the reasons I've become interested in Class D is indeed because of their energy efficiency relative to other approaches.
Please correct me if I am wrong, but...
Replacing the sinewave with the sequence of discrete pulses (with fixed frequency PWM)
we are getting something that contains the 1st harmonic, or what we are to amplify
and some high order ones, which we are to filter out.
And the amplitude of this 1st harmonic will depend on carrier to signal frequency ratio or quantization ratio.
Due to that D-class has a peculiar distortion mode that does not exist in linear topologies.
The higher the ratio the lower the "ripples" of such approximation.
It's yet another reason to have the carrier frequency as high as doable.
That doesn't actually relate to my post, but it does speak to the title of this thread. Unless fidelity is only incidental to efficiency.
Efficiency of Class D
Contrary to the general conception og Class D as the most energy efficient amp design, I think some marketing has been too uncritically adopted.
When measured at max rated output, yes they are very efficient, but at normal use delivering 1W or less, they suffer from some idle consumption making them running hotter than class B amps (or low bias class A/B).
Only when I play extreemely loud my class B amp gets hotter than my ICE amp which is always lukewarm to touch though it claims to be 93% efficient, which it in practical use of course isn´t at all
Contrary to the general conception og Class D as the most energy efficient amp design, I think some marketing has been too uncritically adopted.
When measured at max rated output, yes they are very efficient, but at normal use delivering 1W or less, they suffer from some idle consumption making them running hotter than class B amps (or low bias class A/B).
Only when I play extreemely loud my class B amp gets hotter than my ICE amp which is always lukewarm to touch though it claims to be 93% efficient, which it in practical use of course isn´t at all
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.