The sound of PSU

[rinox]

Maybe you know (but I can't find other thread about) that the really sound of every amp coming from PSU ???
Try to connect a switching PSU to your amp (whichever amp) and you' ll be suprised!!
I made a lots of experiments with the support of the guys from Audison (makers of car amps) and the result is that the same amplifier powered by switching PSU sound really surprising.
In dettail we made a try with an old NAD 3020, removed the original PSU, connected the smps to a car battery and voilà...... unbeliveable sound it's not the same amplifier.

The secret of the sound is in PSU!!!!!!!!!

Now the problem is to make a 220/120 Volts SMPS, if someone have some ideas to build it (PC PSU are not good = low voltage) let us know, thanks.

[analog_sa]

Try to connect a switching PSU to your amp (whichever amp) and you' ll be suprised!!

I have been surprised. And not in a pleasant way.

[peranders]

Originally posted by rinox
The secret of the sound is in PSU!!!!!!!!!

Congratulations to your new discovery. Not many have made it before you.

[thomas997]

Originally posted by rinox
Now the problem is to make a 220/120 Volts SMPS, if someone have some ideas to build it (PC PSU are not good = low voltage) let us know, thanks.

Someone did some modifications to a computer psu to tie the rails together to get +/-/gnd and to boost the voltage a little.

It might have been posted on here.

[HDTVman]

Originally posted by rinox
Maybe you know (but I can't find other thread about) that the really sound of every amp coming from PSU ???
Try to connect a switching PSU to your amp (whichever amp) and you' ll be suprised!!
I made a lots of experiments with the support of the guys from Audison (makers of car amps) and the result is that the same amplifier powered by switching PSU sound really surprising.
In dettail we made a try with an old NAD 3020, removed the original PSU, connected the smps to a car battery and voilà...... unbeliveable sound it's not the same amplifier.

The secret of the sound is in PSU!!!!!!!!!

Now the problem is to make a 220/120 Volts SMPS, if someone have some ideas to build it (PC PSU are not good = low voltage) let us know, thanks.

The key to performance of any amplifer is the power supply as you point out. I would not how ever say that switch mode supplies are the ultimate. The biggest factor as far as I can tell in the output impedance across the power bandwidth of the amp. Any power supply that has an output impedance at least 100 times lower that the load on the amp across the entire bandwidth will let that amp sound as good as it can. (8 ohm load need PS with output Z=.08 ohms or less. My rules for my stuff.)

The issue with switchers is noise at the switching frequency and harmonics. That's not a problem if the amp is bandwidth limited but can be very bad with a wide bandwidth amp.

If you read the threads on "chip amps" a lot of what goes on with these amps is power supply tuning. I seems like everyone is tuning the caps. They all seem to be looking for the part that give the magic touch to a mass produced IC amp. by playing with the power supply impedance across the frequency spectrum.

Just my 2 cents

BZ

[HDTVman]

Any comments on my power supply impedance v frequency ideas?

BZ

[carlosfm]

Originally posted by HDTVman
Any power supply that has an output impedance at least 100 times lower that the load on the amp across the entire bandwidth will let that amp sound as good as it can. (8 ohm load need PS with output Z=.08 ohms or less. My rules for my stuff.)

Originally posted by HDTVman
Any comments on my power supply impedance v frequency ideas?

Idealy you need a low impedance PSU, extending to very high frequencies.
The impedance of a PSU varies with frequency, as a speaker's impedance changes with frequency too.
Specifying a speaker as "8 ohms" can be as misleading as yor "Z=.08 ohms" impedance for the PSU.
You can't use that "rule" of yours.
Just imagine your rule applied to a PSU for a signal op-amp driving an 100 kohm load.

[HDTVman]

Originally posted by carlosfm






Idealy you need a low impedance PSU, extending to very high frequencies.

Yes that is correct.

The impedance of a PSU varies with frequency, as a speaker's impedance changes with frequency too.

Yes, that is what I was refering to as power supply tuning.

Specifying a speaker as "8 ohms" can be as misleading as yor "Z=.08 ohms" impedance for the PSU.
You can't use that "rule" of yours.
Just imagine your rule applied to a PSU for a signal op-amp driving an 100 kohm load.

That was intended as an example of the minimum ratio and although I didn't say so that impedance would be for the entire bandwidth of the amp. A ratio greater than 100 / 1 is not excluded by what I said, in fact that would be better. The key to this is that the power supply impedance needs to be low for the entire frequency spectrum that the amplifer will be expected to deal with.

It's my opinion that the cap swapping that goes on is to pick a part that gives the power supply a impedance v frequency relationship that pleases the experimenter. I build the power supplies for my equipment so that the power supply output impedance is very low more that 2 octives beyond the bandpass of the amplifer in question. I try to make this impedance at least 100 times lower than the speaker loads minimum impedance.

BZ

[rinox]

The problem of a normal PSU is that it is connected to main power (120/220 Vac), that isn' t so "clean".
If you have an oscilloscope, you can see that the 50 Hz (I think also 60 Hz) have many mid and high frequency spikes that in some manner "dust" the sound of PSU and of course the sound of the amply connected, this is a fact .
There are many commercial main conditioners, they also works, but they are not capable to totally eliminate mid and high frequency spikes in the udible spectrum.
An SMPS work at higher frenquencies (30-40 Kz) so you simply can't hear these "rumors".

Another good try is to put some inductances in series of the out of PSU like a psu for tubes amps.
A good PSU tube amp is plain of inductances because tubes are very sensible (much more than tansistors) at certain frequencies.
But is quite impossible to make a good inductance when whe speak about some Amperes (tubes use some milliamperes.)
"The sound of tubes" isn' t only for the tubes itself but for the circuit around them.

Meditated people, meditated !!!

[rinox]

Only a note....
I wrote abot Bob Carver in another post.
May be you don't know, but many years ago, there was a "battle" between Mark Levinson (now Red Rose) and Bob Carver.
In an audiction room more then 100 persons (audiophilies) was invited to hear, in the same conditions, the same source, the same loudspeakers etc.... two differents amps.
Results... nobody heard differences in sound. (Mark Levinson costed 30 times more then Carver).... maybe PSU ??
I will try to search for this article but I don' t know if I'll find it, we speak about old things (more or less, 20-25 years ago).

Meditate folks, meditate.