As in Ncore, the lowpass filter used to reduce the switching frequency signal is part of the feedback loop, and add its own inertia (ie delay, ie phase shift), it is most probable that Bruno had to filter one of the feedback pole at relatively low frequency, in order to fulfill Nyquist requirement and keep its amps stable enough.
Or, just the filter itself, in order to get a good compromise between rejection and low order.
Full of common (good) sense.
Last edited:
Esperado, Class D IS purely analogue amp, nothing to do with digital. please do not sustain this fallacy by contrasting with 'purely analogue' Class A, there is no distinction between the 2 here, one is as analogue as the other. some class D have digital input to the modulator, but thats a completely separate issue and nothing to do with being class D
maybe this was not intentional, but that is how it reads
maybe this was not intentional, but that is how it reads
Last edited:
Hi, qusp, where do you read i print such a stupid thing like Class D is digital. I printed PWM !
like I said, whether you mean it or not, your use of language contrasts Class D with 'pure analog class A' suggesting that class D is perhaps not pure analogue.
see what I mean? your use of the word analog here is redundant, as both are pure analog and it only serves to add to/sustain this long standing confusion that the D is digital, or at least not analog. I wasnt suggesting it was likely YOU misunderstood it, simply that your posts contribute to others continued misunderstanding of it.
Last edited:
Exactly. The nCores measure quite well into 2 Ohms and are rated as stable into 1 Ohm. Series connection introduces a resonant circuit between the driver connections, to a degree related to the always-present non-uniformity of the drivers. That resonance, whose strength varies nonlinearly withe the signal, will not be measurably present in the parallel connection. Sure, series connections can sound OK, but there is always that added distortion. Parallel transducer connections are always better than series if the amp behaves well under the parallel load. Of course we all pine to have an amp channel dedicated to each driver--oh wait, mine already have that! : )
Last edited:
Semantic question.
Is class D (pulses of fixed level whose duration is a function of the level of the input signal) pure analog ? Is FM modulation (where frequency is a function of the input level) pure analog ?
Not in my mind, indeed.
Class D is not far from delta modulation (witch is considered as digital) did-you agree ?
Pure analog amplification applies for me when level of the signal is, at each stage , a function of the level of the original signal.
Well, in a way, after the switching frequency had been filtered, we can consider Class D as a 'pure analog' black box ? We agree on this point.
Did everything else than analog is digital ? I never pretended this.
Well, you are right, i have to find a better way to write all those nuances.
Last edited:
yes.
yes.
by a similar logic, a bicyle can fly because, just as the plane, it has wheels.
digital refers to one thing, and one thing only: discrete states.
one second of RedBook PCM can enconde 2 raised to the power 705600 (16*44100) distinct signals. that is ~4 followed by 2 hundred thousands of zeroes. quite a few of them, but still a finite number.
if one decides to encode the information by a continuously variable frequency instead of continuously variable voltage, it's still an analog system. the same with a continuously variable transition moment (switching of output stage transistors). with class D, the information is converted from voltage to switching time, but there's no such thing as finite numbers of discrete states.
this is not a minor issue and definitely not one of semantics. it has the added "bonus" of feeding meaningless discussions along the lines of "class D is digital and digital is bad, so class D is bad".
Last edited:
Hi,
Delta modulation (DM or Δ-modulation) is an analog-to-digitaland digital-to-analog signal conversion technique used for transmission of sound information where quality is not of primary importance. DM is the simplest form of differential pulse-code modulation (DPCM) where the difference between successive samples are encoded into n-bit data streams. In delta modulation, the transmitted data is reduced to a 1-bit data stream.
Implements features and...caos...the continuos evolution for obtain high fidelity and better sn/r are:
Adaptive delta modulation (ADM), continuously variable slope delta modulation (CVSD) is a modification of DM ..etc..etc.
But (exclude particoular project), rest a fact that 1 bit is obtained and processed in analogic way. as comparator in pwm D Class.
FM is analog but you can develop as digital.
We have to consider "Digital" is a magic word for marketing also, and "caos" a magical circumstance.
as a nfb in D Class, has nothing to do with the pure nfb of an analog amplifier.regrds
Ops!... pic
The first digital record i had made as a sound engineer was a revelation: Hell was away, and evils near dead !
Then with sampling rate increasing and oversampling technologies progresses, even its smell and shadow disappeared. With all those dynamic deteriorations, each time you play an analog tape during re-recording and mixing sessions, those wow and flutter, hum, non-linearities, hisses, distortions, this smoke curtain between instruments and you... All those hours to tune polarisation and response curve at each weel change....
Not to talk about vinyls: I remember all the sadness and despair every time we compared vinyls to our original soundtracks .
I'm just waiting now for a digital electric to acoustic transducer
In between, i dream class D amps to be that bad as any decent digital device
to AP2:
I like (and agree with) this definition.
Last edited:
I never implied that you said digital is bad. but saying class D is digital feeds the incorrect understanding of others (run a search on the web and you'll come up with tens of discussions where class D is proven to be bad because it's digital, and look where digital brought us and all that typical whining).
I'm rather skeptical we'll ever have a true digital amp, it could create more problems that it would solve and I'm not sure there will be any benefit other that an ideological "purity" of implementation.
it would either be a power multibit DAC (good luck with that) or a PCM to PWM converter. but, unfortunately, you'll still be dealing with output stage imperfections (of analog nature) which you'll have to correct by feedback. and you'll do... what? convert back to digital (by means of an imperfect ADC) and do error correction in the digital domain? and to achieve what? a philosophical purity which does not apply in real life? "purity" is long gone by the time you look at the output stage and the ADC.
I'm rather skeptical we'll ever have a true digital amp, it could create more problems that it would solve and I'm not sure there will be any benefit other that an ideological "purity" of implementation.
it would either be a power multibit DAC (good luck with that) or a PCM to PWM converter. but, unfortunately, you'll still be dealing with output stage imperfections (of analog nature) which you'll have to correct by feedback. and you'll do... what? convert back to digital (by means of an imperfect ADC) and do error correction in the digital domain? and to achieve what? a philosophical purity which does not apply in real life? "purity" is long gone by the time you look at the output stage and the ADC.
Last edited:
sorry to hear that, I still think I'm safe as long as we're not running for elections here.
I initially though we should just refrain from continuing the slew-rate discussion, now I'm almost sure it applies to all subjects.
Class D can and does produce DC levels as well as high frequency levels without the loss in heat, unlike class A and AB or B. It is an excellent way of producing sound at high levels and to drive large servo systems.
Considering the compression and removal of most harmonics with digital recordings, I find it difficult to understand some of these comments that are looking for perfection.
For me, perfection is vinyl record playing on my Thorens deck through my Quad 303 but then I have tinnitus from an old PA injury and wouldn't know any better. Perhaps.
Don't forget, our ears are logarithmic, so any linear sound level is not too impressive.
I just fix them.
Considering the compression and removal of most harmonics with digital recordings, I find it difficult to understand some of these comments that are looking for perfection.
For me, perfection is vinyl record playing on my Thorens deck through my Quad 303 but then I have tinnitus from an old PA injury and wouldn't know any better. Perhaps.
Don't forget, our ears are logarithmic, so any linear sound level is not too impressive.
I just fix them.
Last edited:
You should try politic, indeed, this is an area where intellectual dishonesty, poor understanding and aggressive attitude are not disabilities. No need for real competency neither.
Last edited:
OK I was mistaken, you really do seem confused yourself as to concept and execution in reality being one and the same thing. as I said, the modulator can be controlled with a digital front end, but that has nothing to do with Class D specifically and only applies in some cases. no, I wouldn't call PWM digital, since, well, you know, the pulse width is modulating … the control input may be able to be conceptualised as digital (as in having an on and off), but the process and output is not digital, nor is the feedback.
its not a matter of semantics, its just a matter of correctness
would you call somebody flicking a light switch on and off, mimicking the stream of 1's and 0's he sees on a screen as a digital process?
its not a matter of semantics, its just a matter of correctness
would you call somebody flicking a light switch on and off, mimicking the stream of 1's and 0's he sees on a screen as a digital process?
Last edited:
Julf, please, think twice about servos.
As long as your closed loop amplifier will have a transmission delay (and they all have) and the overall gain is a function of the feedback, you will have distortion.
Attached, bandwidth curves of the same amplifier took inside the loop (input of the second stage). First image in a voltage feedback configuration (222V/µs, 1Mhz) second in current feedback (1200V/µs, 5Mhz).
As you can see, we are far to be flat in the audio bandwitch, even in the second case.
Hmm - but Ncore (as well as most Class D amps) measure flat in the audio bandwidth
So this simple simulation is not capturing what really happens in a Class D amp
What is the exact model you are usingl?
???
If i says i'm living in a city not far from Paris, did-you read i'm living in Paris ?
If i said at the age of 68, i'm not far from death, did that means i'm dead ?
What is the interest of this controversy ?
I imagine we both know exactly how to design a working class D amplifier (that i did to evaluate, at the early beginning of the idea with discreet components)... well, for you i don't know.
EOT form me, as i said, it is just boring and useless.
Don't you had read: "Attached, bandwidth curves of the same amplifier took inside the loop (input of the second stage)" ?
Those pictures are measuring the signal just after feedback substration.
I'm sure qusp or mr_push_pull will provide the same kind of measurements with their Ncore. They are not common, but interesting as they reveal a lot about open loop bandwidth and poles.
The bandwitch of those amps are flat at the output, up to 1Mhz for the voltage feedback version, 5 Mhz for the current one, of course and as indicated.
Last edited:
- Status
- Not open for further replies.