Home Critic's Corner

Discuss a review. Provide constructive feedback. Talk to the industry.

RE: Corrections

These products re-format a digital input, then send it directly to the output stage (of the amp).

We're not driving a voice-coil with digits -no, but digits are controlling the output stage of an amp. And by doing this, a huge amount of electronics are removed from the signal path.

Do you understand that the 're-format' is an analog process as well as the operation of the output section itself? If not, well: it is. This is why I suggested you try to build one sometime.

Digits are not controlling the output section of the amp! An analog process (PWM) is. This is why I quoted from the manufacturer's websites!

Pulse Width Modulation is entirely an analog process. It was first proposed in the 1940s. Once a signal has arrived in that state it is not digital. Digital is a bunch of numbers.

Class D is an analog process from start to finish. The correct term, if not class D, is 'switching'. The companies that purport to have a 'Digital amplifier' are engaging in a form of shorthand. Its marketing.

This type of amplifier and digital both have distortions that are uniquely their own. For example, digital audio has a distortion that can be called 'inharmonic distortion' and is a form of intermodulation where the intermodulation is between the signal and the scan frequency. The digital world does not like to admit this so they call it 'aliasing' but it is distortion nonetheless.

Its also very audible to the human ear. This is because aliasing artifacts tend to be high frequency in nature and so occur in the range at which our ears are most sensitive. They can't be easily detected using digital sources, but are easily heard if you record and playback using an analog sweep tone). Try it some time.

The distortions of a class D amplifier have to do with the clock frequency and the accuracy of the PCM circuitry. Switching amplifiers have some limitations not unlike digital (bandwidth being one of them); if you really want the amplifier to be low distortion, a higher scan frequency is paramount. This gets to be tricky as high speed power switching is easy enough, but the higher you go the more expensive it gets. For this reason most current class D circuits switch at a frequency that is too low (500KHz is a good minimum but 5MHz is better). The problem is avoiding phase shift in the audio passband; to do that you usually have to have bandwidth to 10x the maximum frequency to be reproduced. That puts the switching frequency higher than nearly all the class D (or so-called 'Digital Amplifiers') made.

Now the switching devices like to stay on for a while after they are turned off, so to prevent a phenomena called 'shoot-through current' which heats up the power transistors pretty fast, a circuit is installed that introduces 'dead time'; a period of time that gives the output devices of one side of the amp time to shut off before the other side turns on. The more dead time the more distortion; dead time is not needed at lower scan frequencies. So there is a bit of catch-22 situation going on there.

This is not to say that I don't see hope for the technology- Class D is clearly the rising star in amplifier technology (we've been working on it for several years now).

I have less hope for the digital audio recording industry as an attitude that its 'good enough' prevails. Your own post OP is an example. As long as that attitude is out there, analog will continue to outperform digital as the incentive to improve simply does not exist.

This post is made possible by the generous support of people like you and our sponsors:
  The Cable Cooker  

Follow Ups Full Thread
Follow Ups


Post a Message!

Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.