That quote from Wikipedia says that analog systems "use a continuous range of values to represent information". Exactly what PWM does! There is more to a pulse than its height.Charles Darwin said:
PCM is definitely digital, because the original data is turned into discrete numbers.
finally after 4 pages of tech experts yapping on about AB systems and spouting techno-prejudice, someone mentions that the D does not actually stand for digital..... good show, i thought i was going to read the whole thread and have to post it myself and i dont even own one. digital amps exist, some of them are even class D; but class D is very much an analogue technology.
DF96, what kind of transducer would one use to convert the "digital numbers" in a digital amplifier that you describe to something that one can listen too? Would you say that a clock is digital or analogue, it has both frequency and amplitude? If it is divided by N logically, its frequency changes, which is analogue quantity.
You cannot listen to raw PWM it is just noise, PCM is also just noise.
You cannot listen to raw PWM it is just noise, PCM is also just noise.
This is purely about amplitude, not time. A pulse can vary over time as much as it likes it is only ever going to be a multiple of the minimum duration ie a row of 1s of a certain length. It will never be 0.5 or 0.33 or 2.7 or 3736. It is therefore by definition a digital system.
A more technical paper is here DIRECT ACOUSTIC DAC FROM A DIGITAL TRANSDUCER ARRAY LOUDSPEAKER - PhD THESIS JORGE MENDOZA LOPEZ pdf
Well, it took me a while to realise that some people thought it was digital. It never occured to me that people might make this mistake.qusp said:
I did not describe a digital amplifier. By digital system I mean something like CD or DAB radio. You have to use a Digital-to-Analogue converter. For Class D no DAC is needed, because no conversion takes place - you just use a low pass filter. This again demonstrates that it is not digital.Nico Ras said:
Who said anything about multiples of a minimum duration? If Class D had discrete possible lengths of pulse then it would be a digital system, but as far as I know the pulse length can vary continuously from some minimum (possibly zero) up to some maximum (possibly no break between pulses). The fact that the variation is bounded does not make it digital, as a voltage representing an analogue signal is also bounded.Charles Darwin said:
Looong time ago I made a recording amp for a tape recorder, like class D amps are made now. I made triangle shape 120 KHz HF bias signal, and PWM-ed it controlling by signal I recorded. Quality of record was excellent, wideband, low distortion, crisp highs, lower noise. But I had to disassemble it because neighbors wanted to call the police. They heard everything I recorded on their AM radios.
I replaced one bryston 4BST and one 4B with 2 hypex D2.100 class d converting amps for my kef 104/2 in bridged mode and one class d badge amp for the subwoofer, and don't miss the audiophile bull that is made up around the brystons or any "audiophile" amp of whatever ilk.
Class D works for me, I did not need the heat production even on idle of the brystons.
Saves on AC in summer, and better room temp. control in winter...
Class D works for me, I did not need the heat production even on idle of the brystons.
Saves on AC in summer, and better room temp. control in winter...
sorry, just because something has 2 states, that does not make it digital, you are mistaking binary with digital. besides its not even a true binary signal due to amplitude variation in the output (not of the pure pwm)
Last edited:
No I meant Hex, or is it Octal, binary coded decimal, heck I do not know but "digital" electronics operates only with two states, unless you can give an example of something different, maybe a four or more state switch.
For those that are interested I have found an IRS2092 class D audio driver in my stock of components the data sheet is here http://www.irf.com/product-info/datasheets/data/irs2092.pdf and looks interesting enough to keep me occupied for the week-end.
I haven't found time to read it all, but I d/l'd this paper from the Hypex website, written by Bruno Putzeys et al. Its title is All amplifiers are analogue, but some amplifiers are more analogue than others and the first sentence of the abstract is, "This paper intends to clarify the terms “digital” and “analogue” as applied to class-D audio power amplifiers."
Bruno's papers are a good read and his amps a very good choice if you want to find out what Class D can do.
I think the question in the first post is what happened to digitally encoded amps, not what happened to class D or class A, B and C... Switching systems are loosely referred to as digital systems, whether using relays, transistors, valves or light, whatever you please.
PWM is analogue encoded digital, it is not binary encoded digital. Digital does not define it to be numerical, whether you want to assign 1 and 0 or A and B to the on and off states is really your choice.
PWM is analogue encoded digital, it is not binary encoded digital. Digital does not define it to be numerical, whether you want to assign 1 and 0 or A and B to the on and off states is really your choice.
no, again, you are defining binary and even that not very well, anything that also includes amplitude information ceases to remain even classified as binary
Last edited:
Early musical digital delay devices used delta-modulation. What if to use it for recording & reproduction? There would be simple D/A outputs available, capable to drive loudspeakers directly.
binary is but one transfer medium that digital systems use to transfer encoded information but being binary is not enough to define something as digital. digital systems contain a methodology for decoding the various secondary formats from the binary stream, that do not literally represent the signal used to transmit it.
- Status
- Not open for further replies.