Salas you read my mind ,nice so see your shunt on JC & company thread 
Andrew, came from the wall I don't know what http://www.enel.com/en/ use
Andrew, came from the wall I don't know what http://www.enel.com/en/ use
from JC:
It is composed of 3 passive stages, then 3 active stages of regulation and noise reduction, on each channel and each supply voltage,
------
Iko I thinks we have a lot to learnig ,and an official raw psu+input rfi filter at last all passive for Shunt can be good !
a know trafo (with misured c-l ) and so on ,
I have a question , with split secondary (symmetry psu +0-)is better use double bridge but I have reed that we must balanced the current to avoid current on gnd, ie balenced thr V with resistor ??
Any one can point a pratical book on psu ?
It is composed of 3 passive stages, then 3 active stages of regulation and noise reduction, on each channel and each supply voltage,
------
Iko I thinks we have a lot to learnig ,and an official raw psu+input rfi filter at last all passive for Shunt can be good !
a know trafo (with misured c-l ) and so on ,
I have a question , with split secondary (symmetry psu +0-)is better use double bridge but I have reed that we must balanced the current to avoid current on gnd, ie balenced thr V with resistor ??
Any one can point a pratical book on psu ?
Three passive and three active? I don't have the context, but one question pops in my head: why stop at three? Or why not two? nico, over at tube diy asylum there are endless discussions about the psu, religious wars almost, but also lots of good info. One has to be careful what to dismiss and what not. One of the things that were real "eye openers" for me was looking at the psu with the "noise amp"+scope. It's like looking with a microscope at the surface of a perfect surface, all of a sudden it looks not so perfect anymore. The other was the realization that a smooth V without any regard to the dynamic current draw needs of the source is worthless. As soon as the source starts tugging on the current, all that smooth V starts jumping around like drunk mosquito. Those big guys have been doing this for many years, and they've been doing it professionally, so we have much to learn from their experience. I too noticed that psu active filtering is something that the shunt reg in simulation is excellent, but in reality I'm not so sure. Hence a good passive filter before it might make a lot of sense. You guys with refined listening skills should test. I am not aware of any psu book that would focus on the needs of a picky audiophile.
nicoch46 said:
Hi nico, I'm not sure I understand the question. It is possible that the positive and negative rails see different current drawn by whatever they're powering. Two regulators will balance the V+ and V- but I see no point in trying to balance the currents; for that you would employ two constant current sources, one for + and one for -. It all depends what you're powering with this psu. In general people seem to agree that it's better to have as much separation as possible for the two different rails as well as for the two different channels.
My 2c.
ClassA load and supply rail currents
Hi,
The regulated dual polarity supply will power your ClassA amplifier.
The amplifier in it's quiescent state will draw current from both supply rails. It sources it from the +ve rail and sinks it from/to the -ve rail.
If the amplifier draws exactly the same current for both rails then zero current will flow in the ground in the quiescent state. But I will come back to that.
Now apply a signal to the ClassA amplifier.
That signal generates a current in the output/load.
Where does the output current come from?
On positive parts of the AC cycle the load current is sourced from the +ve rail.
On negative parts of the AC cycle the load current sinks from/to the -ve rail.
The ClassA amplifier achieves this by slightly increasing the sourced current and slightly decreasing the sinked current, when the AC cycle is positive. When the output current is negative the opposite is required to happen, i.e. the +ve rail current decreases and the -ve rail current increases.
The difference between the sourced and sinked flows to the load.
Now we have different currents in the two supply rails and the difference flows through the load and returns through the load ground.
We have the situation of zero output current and may have zero load ground current when there is no input and no input ground current.
I said I would go back.
What if there is a very small amount of noise on the output flowing to the load when the amplifier has zero input signal.
We know that the noise current must be the difference between the +ve and -ve supply rail currents and that this difference flows back through the load ground.
We have a situation where when the ClassA amplifier is in the quiescent state and draws identical current from both supply that the ground/return does have a small noise current flowing in it.
This noise can be around -120dB below the maximum signal level and so the current is very low, but measurable. Most would ignore these nA of noise current, that's why we get away with saying that no current flows in the ground/return when the amplifier is in the quiescent state.
But there is an exception to this.
A very few ClassA topologies do not return the load current through the separate ground line. They flow the output current through the supply rails. These exceptions have a true constant current behaviour even when they are not in the quiescent state.
Hi,
The regulated dual polarity supply will power your ClassA amplifier.
The amplifier in it's quiescent state will draw current from both supply rails. It sources it from the +ve rail and sinks it from/to the -ve rail.
If the amplifier draws exactly the same current for both rails then zero current will flow in the ground in the quiescent state. But I will come back to that.
Now apply a signal to the ClassA amplifier.
That signal generates a current in the output/load.
Where does the output current come from?
On positive parts of the AC cycle the load current is sourced from the +ve rail.
On negative parts of the AC cycle the load current sinks from/to the -ve rail.
The ClassA amplifier achieves this by slightly increasing the sourced current and slightly decreasing the sinked current, when the AC cycle is positive. When the output current is negative the opposite is required to happen, i.e. the +ve rail current decreases and the -ve rail current increases.
The difference between the sourced and sinked flows to the load.
Now we have different currents in the two supply rails and the difference flows through the load and returns through the load ground.
We have the situation of zero output current and may have zero load ground current when there is no input and no input ground current.
I said I would go back.
What if there is a very small amount of noise on the output flowing to the load when the amplifier has zero input signal.
We know that the noise current must be the difference between the +ve and -ve supply rail currents and that this difference flows back through the load ground.
We have a situation where when the ClassA amplifier is in the quiescent state and draws identical current from both supply that the ground/return does have a small noise current flowing in it.
This noise can be around -120dB below the maximum signal level and so the current is very low, but measurable. Most would ignore these nA of noise current, that's why we get away with saying that no current flows in the ground/return when the amplifier is in the quiescent state.
But there is an exception to this.
A very few ClassA topologies do not return the load current through the separate ground line. They flow the output current through the supply rails. These exceptions have a true constant current behaviour even when they are not in the quiescent state.
Very nicely put, Andrew.
and Nicoh,
you can see from the above how trickie it can get as to where the so called "central 0volt" (or star earth) is located and all the different current path that returns to it (or them) does require a bit of sorting out sometimes. Those d. ground loops are quite difficult to avoid sometimes, and the shunt Regs create another path, naturally.
and Nicoh,
you can see from the above how trickie it can get as to where the so called "central 0volt" (or star earth) is located and all the different current path that returns to it (or them) does require a bit of sorting out sometimes. Those d. ground loops are quite difficult to avoid sometimes, and the shunt Regs create another path, naturally.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.