The power supply influence sound quality loooong before you reach clipping levels.
Ahh, those anecdotes 'but it sounds better'. Never can win there. Never learn anything there either
jd
So, in comparison, how valuable are your own anecdotes about how your PAX amplifier sounds more like a class A amplifier than a class AB amplifier, along with the insinuation that this is perhaps because there is something magical about HEC over plain old conventional global NFB?
Enter the overly humble club of the masters of feedback!
Last edited:
Ok, point taken. Now, go back to those glowing things with pure conventional NFB and let everybody know how sweet they sound.
Of course you could explain that in more detail, if you wanted?
jd
Because a soft supply can never perform like a stiff one, you will always battle to get the electrons to flow.
Why did Walt and you design a superreg if soft supplies are better?
Because a soft supply can never perform like a stiff one, you will always battle to get the electrons to flow.
Why did Walt and you design a superreg if soft supplies are better?
OK, I agree that power supplies are important to quality sound, but the comparison was with soft. higher-voltage supplies leading to dynamic headroom, and stiff supplies with zero headroom, all other things being equal.
Maybe that is not a realistic proposal.
But on the other hand you can make a supply with for instance a capacitance multiplier that has pretty much the same performance whether it is fed by a stiff or a soft supply.
The point I wanted to make is that good dynamic headroom in itself is definitely going to sound better because it leads to less clipping. But I agree with you that if it leads to worse sound overall it should not be used.
jd
Like all anecdotes; there may be something to it, or maybe not. You know how I think about anecdotes, right?
Also, I admit that if people tell me repeatedly how fantastic my amp sounds, I tend to agree with them. I'm a sissy at heart
jd
I think a lot of people are fooled by their ears.
I hear this all time about people changing speaker leads or PSU capacitors and suddenly the amp is wonderful.
While there is huge amounts of evidence as yet I havent heard much in the way of theory for why things sound so much better.
I can understand minor differences that only a scope or spectrum analyser might pick up but vast changes I really dont believe.
this is where we disagree.
to avoid clipping I use 100W to 170W amplifiers into 89dB to 93dB speakers, usually at an average level below half a watt.
I don't go and buy a high dynamic headroom amplifier that kids us all (nearly all) on that it is better when to me at least it is clearly not better at delivering high current transients.
The supply voltage will drop just as fast with a capacitance multiplier unless the transients are very short. Never tried it so I have no idea what its effect would be on SQ.
I believe if you have a problem with clipping, you need a higher wattage amplifier. To me a good indication of an amplifier's abilities is to see if it can come close to doubling its power into half the load. I hear the differences between PSU's at listening levels way below clipping, don't know why, perhaps the current transients into speakers are much more important for SQ than realised.
You’d have a hard time finding any post of mine waxing lyrically over any alleged magical sonic properties of vacuum tubes.
Well it seems to be that it’s fine to relate your own anecdotes to support a technically tenuous position, but highly annoying for someone else to do so (and worth complaining about).
Dynamic Headroom/Clipping
Hi All,
Janneman - you said that you thought an amplifier with a basic power supply would sound better at 100watts than an amplifier with a regulated supply due to the fact that the latter would clip.
You also said that most of the time you would be unlikely to listen at power levels above 10 watts - that being the case which would be better, the regualted or unregulated supply (assuming the regulated supply would be capable of suplpying all the peaks ever demanded of the amplifier when being driven at 10 watts).
If the amplifier is Class A and draws constant current (which is i believe the case with my Aleph-X) then what difference is there if there are no peaks?
I believe that regulators can affect the sound but what can be done to mitigate the effects? (i have lots of capacitance after the regulator, is there anything else that might be better?)
Hi All,
Janneman - you said that you thought an amplifier with a basic power supply would sound better at 100watts than an amplifier with a regulated supply due to the fact that the latter would clip.
You also said that most of the time you would be unlikely to listen at power levels above 10 watts - that being the case which would be better, the regualted or unregulated supply (assuming the regulated supply would be capable of suplpying all the peaks ever demanded of the amplifier when being driven at 10 watts).
If the amplifier is Class A and draws constant current (which is i believe the case with my Aleph-X) then what difference is there if there are no peaks?
I believe that regulators can affect the sound but what can be done to mitigate the effects? (i have lots of capacitance after the regulator, is there anything else that might be better?)
Getting more vaoltage
Sorry for this question, but i don't understand so i'll ask;
non-regulated supplies are supposed to give more headroom because they will allow the rails to rise a little if there is more power demanded - my understanding was that with a simple transformer, bridge and reservoir capacitor arrangement the rails were more likely to sag as more current is pulled?
Where do these extra volts come from? and how?
Sorry for this question, but i don't understand so i'll ask;
non-regulated supplies are supposed to give more headroom because they will allow the rails to rise a little if there is more power demanded - my understanding was that with a simple transformer, bridge and reservoir capacitor arrangement the rails were more likely to sag as more current is pulled?
Where do these extra volts come from? and how?
[snip]Well it seems to be that it’s fine to relate your own anecdotes to support a technically tenuous position, but highly annoying for someone else to do so (and worth complaining about).
Not at all. It's only if anecdotes are presented as hard, scientific facts, it's annoying. You should really read more carefully here.
best
jd
They will rise a little is LOW average power is asked, so when a peak comes along, the higher voltage is there to reproduce the peak at a higher level and not clip (yet).
jd
Not at all. It's only if anecdotes are presented as hard, scientific facts, it's annoying. You should really read more carefully here.
best
jd
Current works great on driving low impedance drivers where voltage is suited better at driving high impedance drivers.
Either way it should be balanced at what ever your application is.
I would start with at least 750-1000VA for 2 channels if driving full range and half of that if driving mid/highs as voltage becomes more important in the upper frequency ranges where as for bass there more current than voltage required.
Also lets not forget the limitations of our output devices for both current and voltage break down while maintaining them in SOA.
Last edited:
a higher supply voltage delivers larger current transients. The supplys Ri becomes less important and the energy stored in the supply caps grows with U square.
Regards
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.