I understand a Class A amp is always running at its full load (in terms of power consumption). Doesn't matter if no signal, or it is driving speakers hard towards it's maximum output/clipping. However, when you go from no signal to a signal, do other parts of the amplifier come online and start being used more? If I had superpowers and could see electricity/current flowing through the PCB and components, would it change as I added a signal?
Because I know there are a zillion amp designs, thinking about this in the context of something like the Aleph J I just built, or even the ACA (something Pass-related).
Because I know there are a zillion amp designs, thinking about this in the context of something like the Aleph J I just built, or even the ACA (something Pass-related).
When a signal is added and in the absence of clipping, the average DC voltage and currents will remain the same, generally. The voltages and currents will fluctuate with the signal, but these will be symmetrical about the DC levels.
When clipping occurs, the current demands increase and depending on the power supply capabilities, the DC supply voltage may drop and voltages will change. As clipping starts, if it is only on one phase, the voltages will change as the voltage swings are no longer symmetrical, so the average value changes.
When clipping occurs, the current demands increase and depending on the power supply capabilities, the DC supply voltage may drop and voltages will change. As clipping starts, if it is only on one phase, the voltages will change as the voltage swings are no longer symmetrical, so the average value changes.
The way I think of it (probably wrong) is the speakers are in parallel to ground with the output devices. When an audio signal bangs on their gates, current is squeezed out of the path to ground through the output devices and pushed through the speakers to ground. Like squeezing a water balloon or somethin, that pressure goes elsewhere.
This is what happens - I had this simulation setup (just the output stage) on hand:
Current Flows no signal:
Current flows with signal (almost maximum power):
Instantaneous Power Dissipation with signal:
Average Power Dissipation no signal (Red and Blue are on top of each other since they are exact the same)
Average Power Dissipation with signal:
Current Flows no signal:
Current flows with signal (almost maximum power):
Instantaneous Power Dissipation with signal:
Average Power Dissipation no signal (Red and Blue are on top of each other since they are exact the same)
Average Power Dissipation with signal:
If the class-A amp is "single ended" then only one output transistor is driven. The idle current in that transistor is exactly equal to the ~constant current source from the opposite supply voltage so that the idle output is zero. The speaker current is the difference between those two currents when there is signal.
The current source can be 1. a resistor, 2. an inductor or transformer, or 3. a semiconductor current source circuit. A resistor wastes some of the audio signal and half the idle power. An inductor uses half the supply voltage (and power) and very little signal but ~shorts low frequencies depending on how big it is. A current source circuit wastes DC power like a resistor but wastes very little signal.
If the class-A amp is "push-pull" then both outputs are driven, out of phase. Again, the speaker current is the difference but because one output conducts more as the other conducts less, the difference is twice as much.
In any case, in a class-A amp, the output transistor idle current is half the peak value, resulting in a steady power dissipation at least equal to the peak audio output power, 1.4 times the maximum rms power.
The current source can be 1. a resistor, 2. an inductor or transformer, or 3. a semiconductor current source circuit. A resistor wastes some of the audio signal and half the idle power. An inductor uses half the supply voltage (and power) and very little signal but ~shorts low frequencies depending on how big it is. A current source circuit wastes DC power like a resistor but wastes very little signal.
If the class-A amp is "push-pull" then both outputs are driven, out of phase. Again, the speaker current is the difference but because one output conducts more as the other conducts less, the difference is twice as much.
In any case, in a class-A amp, the output transistor idle current is half the peak value, resulting in a steady power dissipation at least equal to the peak audio output power, 1.4 times the maximum rms power.
Last edited:
Yes you would see current starting to pass the load (speaker) instead of being burnt in the power transistor - this implies that the load seen from the power grid and the power transformer in the amp to be constant irrespectable of the input signal (as long as amp is operaring in class A).
//
//
Sorry, this one got lost and I didn't reply. Thanks, very interesting info.
What I think I am hearing is the the different power transistors will be getting more load when in use (even though the overall power consumption of the total amp is roughly the same)? So if you 'believe' burn in, or not, just turning the amp on isn't enough, you need to play a signal to get things working.
What I think I am hearing is the the different power transistors will be getting more load when in use (even though the overall power consumption of the total amp is roughly the same)? So if you 'believe' burn in, or not, just turning the amp on isn't enough, you need to play a signal to get things working.
You’re wrong. They are in series with the output devices.
No you wouldn’t. Current isn’t ‘burnt’. You would see one device conducting more and the other less, and therefore a difference signal passing through the speakers.
Single-ended class A:
One transistor provides gain. The other half is a current source (so, no gain...) that, among other things, produces 0 V at the speaker output.
Once the input signal is present, the middle point is no longer zero. If, at that point, a load is connected towards the common return point, the current will flow through that load (speaker) between the supply potential, through the transistor, and back towards the common return (transformer centre tap). The max current will be equal to the max current through that single transistor.
Push-pull class A:
The other half I referred to above is now an amplifying element, i.e. another transistor. Everything's the same. The current will now be able to source and sink through the speaker load, i.e., we get more current—more power. However, the resulting current through the speaker is no longer sourced and sinked from a single transistor; it is now sourced OR sinked from 2 different sources (2 transistors).
The maximum power will depend on the power supply voltage, the current ability of the power supply - and transistor(s) current gain/ability to stay in a linear area, and the gain of the first voltage gain stage.
One transistor provides gain. The other half is a current source (so, no gain...) that, among other things, produces 0 V at the speaker output.
Once the input signal is present, the middle point is no longer zero. If, at that point, a load is connected towards the common return point, the current will flow through that load (speaker) between the supply potential, through the transistor, and back towards the common return (transformer centre tap). The max current will be equal to the max current through that single transistor.
Push-pull class A:
The other half I referred to above is now an amplifying element, i.e. another transistor. Everything's the same. The current will now be able to source and sink through the speaker load, i.e., we get more current—more power. However, the resulting current through the speaker is no longer sourced and sinked from a single transistor; it is now sourced OR sinked from 2 different sources (2 transistors).
The maximum power will depend on the power supply voltage, the current ability of the power supply - and transistor(s) current gain/ability to stay in a linear area, and the gain of the first voltage gain stage.
Last edited: