panomaniac said:Hate to break up the little love fest going on here, but I want to ask a question.
Hi,
Personally I think it comes down to design. A Quad current-dumper is different from a Sony MOS-FET. Different design, different layout and different component choice. Class A with no switching distortion and Class B with, different design approach.
If all amplifiers did sound the same, aesthetics aside, then there would probably be only one amplifier available for the world to buy and that would be that.
Perception though is very complex. My partner and I often disagree on that a particular colour IS a particular colour and no doubt I think that my sytem sounds better to me than my partner thinks it sounds to her. You may think I am being contradictory (to my previous comments) here but I think that some people will agree.
Thanks
Gareth
panomaniac said:
I am quite confident that it is not necessarily one of those, I've made small changes to an amplifier that did not change any of these specs including phase (with my measuring equipment) and it change the sound of the amplifier. As Gareth also stated, even different components make a difference.
André
This is like the song that never ends.
There's the Consumer Reports crowd to whom any reasonably well designed 100W amp really does sound the same.
Then there's those who think removing the Bybee whatevers from John's Blowtorch suddenly makes it sound like gob sh*te.
Does anyone here really think a 10W SET amp won't be distinguished from a 10W class T AMP in a careful AB or ABX test?
I thought the interesting question was if two amplifiers of equal power rating that could not be distinguished by any conventional test could actually be proven to sound different. Or I think, more importantly, are things like $300 supply caps detectible on two otherwise identicle amplifiers.
BTW I like Trocken wine and I hate California Cabernet.
There's the Consumer Reports crowd to whom any reasonably well designed 100W amp really does sound the same.
Then there's those who think removing the Bybee whatevers from John's Blowtorch suddenly makes it sound like gob sh*te.
Does anyone here really think a 10W SET amp won't be distinguished from a 10W class T AMP in a careful AB or ABX test?
I thought the interesting question was if two amplifiers of equal power rating that could not be distinguished by any conventional test could actually be proven to sound different. Or I think, more importantly, are things like $300 supply caps detectible on two otherwise identicle amplifiers.
BTW I like Trocken wine and I hate California Cabernet.
Hi,
I agree that this argument could go on forever. Realistically speaking how could you go about designing a test that can differentiate whether one amp sounds better than another.
Whilst a lot of people seem to think that THD etc etc are the be-all and end-all of amplifier specifications this is not true. Whilst these figures can go a long way in determining the operability of a product they do not tell you if it performs.
Many valve amps of yesteryear have THD measurents above the 0.0000000000000 +/- 0.00000% range with a slew rate of 600kV/uS and quite frankly sound more involving (to my ears at least). Not all valve amps though.
Amps are different just like people are different and I think that maybe some people should be spending less money if they truly cannot tell the difference between, but again, more money does not mean better design/quality/sound.
Thanks
Gareth
I agree that this argument could go on forever. Realistically speaking how could you go about designing a test that can differentiate whether one amp sounds better than another.
Whilst a lot of people seem to think that THD etc etc are the be-all and end-all of amplifier specifications this is not true. Whilst these figures can go a long way in determining the operability of a product they do not tell you if it performs.
Many valve amps of yesteryear have THD measurents above the 0.0000000000000 +/- 0.00000% range with a slew rate of 600kV/uS and quite frankly sound more involving (to my ears at least). Not all valve amps though.
Amps are different just like people are different and I think that maybe some people should be spending less money if they truly cannot tell the difference between, but again, more money does not mean better design/quality/sound.
Thanks
Gareth
scott wurcer said:
Scott, it changed from similar measuring amps to all amps and back...
I've found that supply caps is very important (depending on the quality of the rest of course), even found differences between transformers from different manufacturers.
André
Hi André,
I agree that the power supply is important and that different reservoir caps here can make a difference to the sound, a 6700uF having less low-end than a 10000uF, which is no doubt due to the charge times and the amount of charge that can be held. Also bypass caps (for the reservoirs) of differing types, to my ears, sound different.
Today I have been working on my Sub and decided to swap some of the caps in the power supply section. The differences have been noticable in the detail and speed of the low-end, with the sub sounding less 'sluggish' and there is now more ability to define what is happening....more 'tuneful' I suppose.
Thanks
Gareth
scott wurcer said:
I don't know about $300 supply caps, but if you wanted to compare PSU components, the 10W SET would be the way to go since the PSU is part of the signal circuit(at least the last cap). I suppose the driver tubes could be included as well(different cap). Throw in a cascode input, and you've just put a stethoscope on your PSU. Hook 'em up to >93dB speakers and put your ear to the speaker. 😉
BTW... a 100uF/600V motor run cap can be had for ~$15 USD on epay. You don't need no stinking $300 caps...just a little real estate. 😀
pedroskova said:
SS amps like the QSC and it's clones also use the PS caps as part of the output circuit. i've seen a competitor (read "knockoff") of QSC that uses the same basic design, but a bunch of paralleled 20,000uf caps in the PS rather than the 1 or 2 that QSC uses.
the SET amp will sound much different than a 10W amp of other design primarily necause of damping factor. i think damping factor and the related output impedance vs frequency curve of an amp is worth investigating for this topic. if we find two amps that measure the same, they should sound the same, but they frequently don't,
so maybe we're missing a measurement.
one thing i have rarely seen investigated is an output impedance vs frequency curve. sure the two amps might have an identical damping factor at 200hz, but where is the breakpoint on the impedance curve.... 500hz, 1khz, 2khz? the two amps that should sound the same might have very different output impedances between 1 or 2 khz and 20khz, and that might be, however subtle, what makes them sound different
this thread seems to be burning up a lot of bandwidth
That's the crudest form of PWM possible, gives terrible distortion performance, and was rejected out of hand for being unworkable at the very beginning of the digital amplifier's development (or at least, in the first paper published on it by the AES, in 1984). PWM can be done much better - certainly, those huge clock frequencies you quoted aren't required - but it still doesn't change the fact that open-loop digital amplifiers are incredibly difficult to implement because of their PSU and filtering requirements.
If it were that easy it would have taken over long ago!
unclejed613 said:
The QSC I've opened the chasis on had dozens of parallel capacitors, in both the primary and secondary. (It was something in the PL 2 series I think, switching supply). The multiple parallel capacitors have the benefit of placing the equivalent series resistance in parallel.
unclejed613 said:
Given that a triode is about the least linear gain device next to an SCR, it should be easy to hear. This is particularly true in designs lacking negative feedback to temper the non-linearity.
unclejed613 said:
This is a very good point. If the output impedance changes with frequency or amplitude, it will be audible. As a quick back of the envelope calculation, a change in output impedance from 1 to 1.1 ohms into a 4 ohm speaker will result in a -0.2dBV change.
unclejed613 said:
SS amps like the QSC and it's clones also use the PS caps as part of the output circuit. i've seen a competitor (read "knockoff") of QSC
this thread seems to be burning up a lot of bandwidth
stupid question...is the QSC, or it's clones, single ended?
stupid question...are you saying that a single naked transistor is more linear than a triode?
the QSC is a class B solid state amp that has the outputs tied to ground. the power supply center tap is left floating, and becomes the speaker output. the amp works by "walking" the power supply rails, and the power supply center tap voltage "walks" with them, providing the output.
i know the amp fizzard is talking about. similar design, but done with a switching supply, but i actually had their original "heavy iron" amps in mind.
the QSC amp has two design advantages:
1> the output collectors are grounded, so heat sink insulators aren't needed
2> the amp can be driven by an input stage and VAS operating from +/-15V rails
in the picture, pay close attention to where the ground is:
i know the amp fizzard is talking about. similar design, but done with a switching supply, but i actually had their original "heavy iron" amps in mind.
the QSC amp has two design advantages:
1> the output collectors are grounded, so heat sink insulators aren't needed
2> the amp can be driven by an input stage and VAS operating from +/-15V rails
in the picture, pay close attention to where the ground is:
Attachments
fizzard said:
damping factor not only has an effect on amplitude, but how the amp and speaker interact. which has a much larger effect in the frequency response curve of the system.
unclejed613 said:
Once you're above a certain damping factor the speaker dynamics are taken care of. It's much more important that output impedance remain constant in magnitude over the frequency range.
unclejed613 said:
Crown Audio does this too, if I'm not mistaken. Who came out with it first? You refered to a "knockoff" a few posts back.
Andre Visser said:
...not sure what you're getting at, but if it's equivalent energy storage, a 100uF capacitor with 600volts across it stores the same amount of energy as a 100,000uF cap at 18volts.
fizzard wrote:
"Crown Audio does this too, if I'm not mistaken. Who came out with it first? You refered to a "knockoff" a few posts back."
American Audio is the one of the "knock-off" manufacturers. Crown does mostly standard output stages, but i wouldn't be surprised if one or two models of their amps do the QSC thing as well.
i've also seen an American Audio amp that was a class-G version of the QSC-like output.... that was a beast to troubleshoot.
i am pretty sure QSC came out with it first. it's a simple variation on class-B that has good reliability and safety benefits (you could, with such a design actually put the output devices on the outside of the case without needing the plastic covers usually needed on exposed TO-3 cases). the only trouble is that you have to kind of "think backwards" when troubleshooting it since the power supply rails float with the audio. it kind of threw me for a loop the first time i repaired one, but once i understood what was being done, i was fine (what? they're "driving" ground????????)
"Crown Audio does this too, if I'm not mistaken. Who came out with it first? You refered to a "knockoff" a few posts back."
American Audio is the one of the "knock-off" manufacturers. Crown does mostly standard output stages, but i wouldn't be surprised if one or two models of their amps do the QSC thing as well.
i've also seen an American Audio amp that was a class-G version of the QSC-like output.... that was a beast to troubleshoot.
i am pretty sure QSC came out with it first. it's a simple variation on class-B that has good reliability and safety benefits (you could, with such a design actually put the output devices on the outside of the case without needing the plastic covers usually needed on exposed TO-3 cases). the only trouble is that you have to kind of "think backwards" when troubleshooting it since the power supply rails float with the audio. it kind of threw me for a loop the first time i repaired one, but once i understood what was being done, i was fine (what? they're "driving" ground????????)
- Status
- Not open for further replies.