Have amplifiers finished evolving?
- Thread starter Tim F
- Start date
Shadders
pfm Member
Hi,
If you examine the following :
https://en.wikipedia.org/wiki/Digital_geometry
and
https://en.wikipedia.org/wiki/Discrete_space
The fact that the implementation uses a low pass filter (integrates) on the PWM signal to create what is seen as an analogue signal, does not negate that the switching function is part of the above references. The PWM signal is either a positive or negative voltage - a digital or binary signal. (https://en.wikipedia.org/wiki/Digital_signal)
Regards,
Shadders.
Shadders
pfm Member
Hi,
Yes - they are - they are either on (binary state 1) or off (binary state 0). Don't confuse the action of the switch and the effect it produces on the light bulb, with the energy dissipated by the light bulb.
Regards,
Shadders.
sq225917
Bit of this, bit of that
Except lights come with dimmers. A bit like this thread.
Class D is PWM switching, the switching is controlled by an analogue sourced signal, ergo it's an analogue amplifier.
If the switching was controlled directly by digital data then we'd all be in agreement.
Class D is PWM switching, the switching is controlled by an analogue sourced signal, ergo it's an analogue amplifier.
If the switching was controlled directly by digital data then we'd all be in agreement.
clivem2
pfm Member
I was only talking about the switch though.
C'mon now...a DAC has a true digital section to it. A Class D amp is far more analogue in its design.
Shadders
pfm Member
Hi,
The PWM signal is a digital signal. It meets all the criteria mathematically and engineering wise for a digital signal. It's output before the integrator is therefore a digital signal.
Regards,
Shadders.
Cereal Killer
432
Never realised PFM had so many ultra experts involving all things audio, why do all these multi million pound R&D programs exist.... all they need do is consult the PFM collective few.
clivem2
pfm Member
Many dimmers are digital too then? They use mark space...
Shadders
pfm Member
Hi,
Yes, the dimmer is digital - in that it uses a mark space ratio to chop the mains, but here the output voltage is a gated sine wave, not two binary signals.
So, the signal/mechanism to gate the sine wave is digital, but the gated voltage is a sine wave, which is an analogue signal, with discontinuities introduced.
For class D, it is a PWM signal, that switches two different voltages - it is a binary signal. It has two signal levels with discontinuities at the PWM switching. It is therefore a digital signal.
An amplifier that has a class A/B output stage which also has a USB input with a DAC is not a digital amplifier, it is still an analogue amplifier since its output stage is analogue. The converse is true for class D - the output stage is digital.
Regards,
Shadders.
sq225917
Bit of this, bit of that
And that's where we disagree.
Class D amps receive an analogue signal from an analogue pre-amp.
That they chop em up to a1-bit PWM signal doesn't make them a digital amp, it makes them a switching amp.
Show me a switching amp that takes a raw digital input from a digital source and doesn't convert it to analogue before making the 1-bit that drives the output stage and I'm in agreement 100%, that's a digital amp.
No one calls class D amps, digital, they all call them switching amps. It's a nomenclature difference for sure, but it's an important distinction.
Class D amps receive an analogue signal from an analogue pre-amp.
That they chop em up to a1-bit PWM signal doesn't make them a digital amp, it makes them a switching amp.
Show me a switching amp that takes a raw digital input from a digital source and doesn't convert it to analogue before making the 1-bit that drives the output stage and I'm in agreement 100%, that's a digital amp.
No one calls class D amps, digital, they all call them switching amps. It's a nomenclature difference for sure, but it's an important distinction.
Shadders
pfm Member
Hi,
The output stage waveform of the class D amplifier meets every criteria (mathematical, engineering) of a digital signal. The output of the output stage is a digital, binary signal.
Switching amplifier defines the topology only. Not the classification of the signal output from the output stage, which is digital.
Regards,
Shadders.
radamel
Music Fiend
Cereal Killer
432
IMHO a Power DAC is the correct description, and how i see the likes of Devialet.
Shadders
pfm Member
Hi,
You could call a class D amplifier as a mixed signal system - a SAR ADC has a digital output, but the input is analogue and has analogue components onboard, where as a DAC has digital inputs and an analogue output.
Regards,
Shadders.
Cereal Killer
432
Hi - If you amplify during the conversion you get a powered DAC. Thats how i understand it, and is what Lyngdorf also refer too (IIRC) http://lyngdorf.com/true-digital-amplification/
Jim Audiomisc
pfm Member
So the answer to the conundrum is that such devices are a form of ADC. Thus *both* 'analogue' and 'digital' at the same time. If we add in that the signal then given to the loudspeaker has to go though an 'analogue' integrator / low pass filter to give the speaker what it needs, it become an ADC followed by a DAC. Even more mixed.
Personally, this argument about classification has always puzzled me a bit. Reminds me of arguments about how many angels can dance on the head of a pin. 8-]
darrenyeats
pfm Member
Jez,
The class-A-versus-class-A/B and feedback things. Are you saying that class A is better because it needs less and/or different feedback? Thanks.
The class-A-versus-class-A/B and feedback things. Are you saying that class A is better because it needs less and/or different feedback? Thanks.
Devialet work entirely digital. Analogue signal path is about 2cm at the end.