Power Supply Resevoir Size

[liching1952]

Originally Posted by Terry Given

liching1952, that is just trolling. you are at least trilingual, and from what you've posted so far your english is more than sufficient for the task at hand - Ni Shuo Huai Hua. besides, with your MSc you should be able to follow the schematics and plots. that you have not and will not do so suggests you are either lazy or stupid. I'm assuming lazy.

As MagicBox points out, the thread is about transient response.

Why low ripple? stupid question. have you ever heard of PSRR? what about Early Voltage? cascoding?

It becomes more and more interesting, the whole scheme what you called a Cap multiplier is to minimize ripple now you mention transient. Explain, what which transient. No I dun know what PSRR or early voltage is, as said me Inglish is poor, so explain. And what has cascoding to do? I think that you want to make confusing because you recoq the stupidity but want to cover up. So explain instead calling for nonsence.
Yes its not nice to get critical critism from a low level trilingual new comer isnt it.

[MagicBox]

Originally Posted by liching1952

It becomes more and more interesting, the whole scheme what you called a Cap multiplier is to minimize ripple now you mention transient. Explain, what which transient. No I dun know what PSRR or early voltage is, as said me Inglish is poor, so explain. And what has cascoding to do? I think that you want to make confusing because you recoq the stupidity but want to cover up. So explain instead calling for nonsence.
Yes its not nice to get critical critism from a low level trilingual new comer isnt it.

The big point with the cap multi is that you do NOT use a zener and keep the Vds/ce drop as low as possible since the goal is not to build a crude lineair regulator but to increase the virtual capacitance the output devices see.

[liching1952]

Originally Posted by MagicBox

The big point with the cap multi is that you do NOT use a zener and keep the Vds/ce drop as low as possible since the goal is not to build a crude lineair regulator but to increase the virtual capacitance the output devices see.

Thats right you minimize the lost but thats not my question, why this effort to minimize ripple. A well designd amp doesnt need ripple free supply.
And, what you all dont see, the thir so called cap multiplr introduce a lot of other effects that lowered the overal performance of the amp thats why I called it nonsence but to be unpolite, its stupid.

[MagicBox]

To produce audio correctly and unaltered, a good amp has to work at and operate at frequencies up to 10 times the audio band. That's where transient response comes into play, most notable the "T=0" current as I call it. The initial current that needs to be drawn as a result of a change in the system. Propagation delay, parasitic inductance all limit the T=0 current resulting in a voltage sag that otherwise should not sag but deliver the associated, demanded current.

People here are not fighting ripple, PSRR does that like you say. We're past that. We're not analyzing ripple. We're analyzing T=0 current supply any moment the output demands it. And this does not apply to the 'slow' audio signal, but to the HF loop frequency of the circuit.

At regulatory frequences, the AC output impedance of a MOSFET / Transistor is rather low; any HF signal on the output transistors 'walks' right through the output device, severely destructing PSRR. Luckily the amp must be stable and as such will not have HF AC regulatory swing (if it does, it can't stable out) for audio signals. But to keep that audio signal in perfect shape, you no longer have to view an amp as an LF amp but as an MF/HF amp operating in the 100KHz - 1MHz area.

Edit: Basically the amp has to be fast enough to compensate for its inherent slowdowns in the circuit.

[liching1952]

Originally Posted by MagicBox

The big point with the cap multi is that you do NOT use a zener and keep the Vds/ce drop as low as possible since the goal is not to build a crude lineair regulator but to increase the virtual capacitance the output devices see.

No the amp dun see a bigger cap. There is less ripple but by no mean seeing a large cap. If you use a zener the ripple is zero. Do the amp sees an Infinity cap? Of course not. An infinit cap should supply the amp with current for invinit time which is certainly not the case. What the amps sees is only the small cap at the output. It even doesnt sees the rectifier cap since this blocked by the transistor.

[Nico Ras]

Liching1952,

What does a cap multiplier have to do with this discussion anyway? You were the one that brought it up so why don't you answer your own question.

Terry,

do you say that the mag current from a rule of thumb should be a specific percentage of output current, or better still you have a rule of thumb for number of turns per volt? However number of turns will affect the inductance, so what is the starting point.

[liching1952]

Originally Posted by MagicBox

To produce audio correctly and unaltered, a good amp has to work at and operate at frequencies up to 10 times the audio band. That's where transient response comes into play, most notable the "T=0" current as I call it. The initial current that needs to be drawn as a result of a change in the system. Propagation delay, parasitic inductance all limit the T=0 current resulting in a voltage sag that otherwise should not sag but deliver the associated, demanded current.

People here are not fighting ripple, PSRR does that like you say. We're past that. We're not analyzing ripple. We're analyzing T=0 current supply any moment the output demands it. And this does not apply to the 'slow' audio signal, but to the HF loop frequency of the circuit.

At regulatory frequences, the AC output impedance of a MOSFET / Transistor is rather low; any HF signal on the output transistors 'walks' right through the output device, severely destructing PSRR. Luckily the amp must be stable and as such will not have HF AC regulatory swing (if it does, it can't stable out) for audio signals. But to keep that audio signal in perfect shape, you no longer have to view an amp as an LF amp but as an MF/HF amp operating in the 100KHz - 1MHz area.

Edit: Basically the amp has to be fast enough to compensate for its inherent slowdowns in the circuit.

Another nonsense consensus. The energy content of music decreases rapidly after 6kHz or so. There is nothink above 15kHz and there are no transients fast enough to translate to more then 15kHz. As a rule of thumb, Transient x bandwidth = 0.35.
And assuming, there are fast transients, what has it to do with a cap multiplier? With a cap multiplier the amp sees only the small output cap, thus less cap then without the circuit.

[MagicBox]

Originally Posted by liching1952

No the amp dun see a bigger cap. There is less ripple but by no mean seeing a large cap. If you use a zener the ripple is zero. Do the amp sees an Infinity cap? Of course not. An infinit cap should supply the amp with current for invinit time which is certainly not the case. What the amps sees is only the small cap at the output. It even doesnt sees the rectifier cap since this blocked by the transistor.

Reading comprehension. We're not talking about the speaker terminal output. We're talking the collector/drain connection points of the output devices. The cap multi is a tool used locally close to the output device power terminals. It's called a "cap multiplier" for a reason. It prevents local sag with a much smaller cap, tapping into the stored current in the PSU buffer caps, though with the nastly L parasitic in between before it gets to your board. The CM transistor actually compensates for the T=0 current draw associated voltage sag.

[MagicBox]

Originally Posted by liching1952

Another nonsense consensus. The energy content of music decreases rapidly after 6kHz or so. There is nothink above 15kHz and there are no transients fast enough to translate to more then 15kHz. As a rule of thumb, Transient x bandwidth = 0.35.
And assuming, there are fast transients, what has it to do with a cap multiplier? With a cap multiplier the amp sees only the small output cap, thus less cap then without the circuit.

You need to learn that what a circuit does is not always what a circuit wants to do. We're not talking audio signal transients. We're talking regulatory frequency transients caused by non-lineairities in devices. When all is good, the regulatory frequency is very high with minimal swing. When things are not so good you have an amp that's outright unstable and if it's on the edge, it'll be very tacky to listen to.

In the ideal situation there is no regulatory swing, it'll be Vpp=0. BUT as soon as a delta appears in the system, the regulatory swing will be back to adapt to the change in the system. The quicker and faster it can do that, the better the amp is.

[liching1952]

Originally Posted by MagicBox

Reading comprehension. We're not talking about the speaker terminal output. We're talking the collector/drain connection points of the output devices. The cap multi is a tool used locally close to the output device power terminals. It's called a "cap multiplier" for a reason. It prevents local sag with a much smaller cap, tapping into the stored current in the PSU buffer caps, though with the nastly L parasitic in between before it gets to your board. The CM transistor actually compensates for the T=0 current draw associated voltage sag.

Wrong. When there is current demand from the amp, without CM, it will draw current from the big rectifier cap. With CM the amp can only draw current from the smaller C at the CM output. The supply will even drop more then without CM.