Class/type of amp ?

Class "G" maybe, where the supply on the output devices doubles 'on the fly' depending on current demand (as in amperes)? Lots of amplifiers are now. The supplies are often designated "VH" and "VL". First units I ever came across which used it, were an Aiwa series of hifis, which did the supply switching with a pair of FETs. I think this covers the principles

Arfa

the

fly'

The

across

switching

SC

I don't have the schematic or even saw the track-side of the Mackie board but that explains the presence of power FETs

Do you know if the pdf of that OCR'd doc (so no schema) is publicly available, I could not find it explicitly on that page.

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

Indeed, the SRM450 Bass amp has 2 supplies : +-45v and +-80v. The higher voltages are switched in with IRFP150 and 160 Fets.

The HF amp is conventional, using just the +-45v supply.

Gareth.

Don't know, to be honest. I just Googled "class G amplifier" to see if I could find you anything on the principles, and that document seemed to cover it pretty well. I'm sure that there must be something else out there with diagrams. If not, let me know, and I'll scan one of the Aiwa schematics for you. They were reasonably straightforward, as I recall.

Arfa

**There are only Class A, Class A/B, Class B and Class D amplifiers used in audio. Anything else is just marketing bullshit. What you have is a Class A/B amp, with a switched rail power supply. Class H, Class G, et al are just marketing terms.

--
Trevor Wilson
www.rageaudio.com.au

Well, you could say that about almost anything. There are many manufacturers that would disagree with you that it is just marketing bullshit. In fact I can't remember ever seeing anywhere that a piece of regular Joe hifi has ever been marketed as class G - or even class A/B. I see nothing wrong at all with giving a derivative of an existing class, a new letter. Whilst class G is indeed a switched rail class A/B amp, it never-the-less is different from a fixed rail class A/B amp. Based on what you're saying, you might as well say that class D is an invalid term, as class A and class B and class A/B (and for that matter class C at RF as well) refer to the point that the output devices are biased to in normal operation, whereas class D refers to an entirely different concept of waveform reconstruction by power device switching i.e. the fully digital output stage.

Arfa

**Wrong. Classes of amplifiers are clearly and explicityly stated.

There are many manufacturers

**Of course.

In fact I

**Here's where I get to say: COMPLETE bullshit. Technics, Yamaha and others have claimed Class A & Class A/B operation for their consumer (as opposed to audiophile) grade components.

I see nothing wrong at

**Me either. Trouble is, it is POWER SUPPLY switching. The fundamental Class of the amplifier's operation remains Class A, Class A/B or Class B in all such cases. Rail shifting schemes are not alterations of amplifier Class of operation.

Whilst

**No, it is not. The amplifier is STILL a Class A/B (or whatever) amplifier, with a rail switching scheme attached. Of course, that does not suit marketers, who dreamt up fancy new terms.

Based on what you're saying, you

**Fair comment.

as class A and class B

**Indeed. The term: Class D has always troubled me. It does not fit with the accepted Class of operation of an amplifier.

--
Trevor Wilson
www.rageaudio.com.au

I hear what you're saying, Trevor, but it seems to me that we are basically just dancing around semantics. Granted, it is still a basic class AB or whatever amplifier, but power supply switching or not, there is still additional circuitry to detect when the higher rails are required, and it could be argued that this circuitry is part of the power amp and its overall design concept.

Given that you accept class D, but are not easy with it, what other designation would you use to identify the switched rail concept as something which 'broadly fitted in with the scheme' and allowed engineers to at least know what it was that they were looking at ? As soon as you start giving design concepts fancy names, every manufacturer will pick his own, and no one will know quite where they are at ...

For sure, it's not ideal, and it does fly in the face a little, of what the original concept of the class lettering system was about, but times move on, and I think that for clarity, issuing this concept with a new letter, is acceptable in practice, if not in theory, for the clarity it brings with it.

Arfa

The original concept of "class" related to the fraction of a cycle the device was conducting.There's A (all), B (half), AB (more than half but less than all), and C (less than half). I can't think of any other meaningful fractions.

There are no other classes. To call switching amps "class D", or to create new designations for stepped B+ or stepped-bias designs thoroughly confuses the original meaning.

They'll do it anyhow, for marketing. If Hitachi has a class-G amplifier, then Toshiba, even though using the same circuit, will call it class H, simply to look original.

How about just _saying_ what it is, in simple language? That would clarify things for the technician, in a way that tacking on a marketing-department-selected letter would not.

Again William, I hear what you're saying, and I am in broad agreement. However, when the class lettering system was first used, the world of amplification was a much simpler place. No one would ever have conceived of fully digital amplifiers, or ones whose rails switched 'on the fly' as a result of output stage demand. The class G concept has been around for a while now, and I don't think that manufacturers have, in general, gone down the route of all having their own name for it.

As we are all fully aware, language and linguistic interpretation changes and develops all the time. It is a fact of life that we all accept, otherwise we would all still be saying "thee" and so on, and "gay" would still mean carefree and happy. The same is true of electronics. Meanings change. The world moves on. "Class G" seems to have been accepted pretty generally by manufacturers as the designation for the type of output stage topology under discussion, just as "Class D" is now accepted as a fully digital amplifier, where bias points don't come into it at all, unless you consider 'hugely on' and 'hugely off' to be valid examples of the term.

I think it is just a case of the system being expanded and adapted to encompass new ideas, and on that basis, other than for want of being an historic purist, I really don't have a problem with it, nor do I see why it should be such a huge problem for others.

At the end of the day, A, AB, C etc are just arbitrary letters to identify particular amplifier topologies, based on the way they are biased. How the letter related to the biasing scheme still had to be learnt, and I really don't see why the system should not have been expanded in the way that it was, to identify other topologies - even if they are just variants or derivatives - based on something other than bias points.

Yes, you could say that this is "A class AB amplifier with switched rails", but how much easier to just say that it's "Class G" ... ??

Arfa

Arfa, please see my other post. Most class D amplifiers are analog, not digital..

rails",

For that to work, you'd have to have some organization -- such as the IEEE in the US -- setting standards as to exactly what class-G topology is.

I used to own Krell amplifiers. They were billed as class A, but they were class A "only" up to about 1/4 or 1/3 full output. Given the crest factor of acoustically recorded music, this means the amplifier will rarely stray from class A. But strictly speaking, the amplifier is AB, with (very) high bias.

Most amplifiers are biased only a little beyond class AB. My Parasound A21 amps are unusual, in that they're still in class A up to about one ampere, which is high for not-horribly expensive amplifier. They are correctly billed as high-bias class AB.

The web in general, would seem to disagree with you on that one, William ...

and many many more examples. The signal in a class D amplifier is only analogue at it's input, and at the speaker terminals after the low pass filter that removes the HF PWM component. Thus, the whole amplifier is fundamentally digital in the way that it amplifies the signal applied to it.

Well, to make it totally 'official', I guess that's so, but again, if you use the web to search for definitions for class G, you will find that pretty much all manufacturers and describers, use the same definition virtually word for word, which would suggest that an unwritten 'standard' for what it is, already exists.

But surely, that emphasises the point that there are no true 'standards' applied to the existing letters ?

Arfa

The concept has been around, and the "class G" terminology has gone with it for awhile, too, at least around 30 years. I have a book entitled "Solid-State Power Electronics," written by one Irving Gottlieb and copyright 1979, that identifies this circuit topology as a class-G amplifier (and also discusses classes A, AB, B, C, D, F, and H). While it may be argued that such a design oughtn't be given a letter designation, it's hardly worth doing so anymore as a practical matter.

--
Andrew Erickson

"He is no fool who gives what he cannot keep to gain that which he cannot
lose."  -- Jim Elliot

...

it.

The Web is wrong. Most switching amps are analog. That is, everything varies continuously, rather than in quantized steps.

By the way, Arfa, you're doing something intellectually invalid -- you're "appealing to authority", rather than thinking for yourself, or explaining what's going on.

For those who would like to read about the correct explanation of "analog versus digital", please refer to the following references. (I can't find my college textbooks, and I don't really think any of these are very good, because the best explanation is graphical.) Sampling is an analog process, that involves multiplying the signal by the sampling function, which produces a convolution in the frequency domain. NO QUANTIZATION OCCURS. If those convinced that sampling = digitization, let them tell me what the bit depth is.

Here's a quote from the last reference. Note especially the third and next-to-last sentences.

"The signals we use in the real world, such as our voices, are called "analog" signals. To process these signals in computers, we need to convert the signals to "digital" form. While an analog signal is continuous in both time and amplitude, a digital signal is discrete in both time and amplitude. To convert a signal from continuous time to discrete time, a process called sampling is used. The value of the signal is measured at certain intervals in time. Each measurement is referred to as a sample. (The analog signal is also quantized in amplitude, but that process is ignored in this demonstration. See the Analog to Digital Conversion page for more on that.)"

The following is directed at everyone in this group -- and is not a rhetorical question -- why is it, that when someone _explains_ to you, in a fairly clear manner, why what you and millions of other people believe to be true, but _is not_ -- you don't believe them? Aren't you able to think for yourselves?

The fact that most people do not understand, and refuse to understand, the difference between analog and digital is, to me, a little frightening, because it touches on the willingness of human beings to believe what they want to believe -- or worse, what "experts" tell them -- rather than the truth.

Disclaimer: When I was a young'un, I thought that if I believed something, it was so. In retrospect, this is ludicrous, but most people are like that. It was many years before I recognized this error of thinking.

I"William Sommerwerck" wrote in news:gig2ai$l71$ snipped-for-privacy@news.motzarella.org:

In general, your argument is correct and Arfa is mistook.

However one quibble: simply chopping the time stream up into integral slices and restricting the level (or width [or frequency {or phase}]) of an output signal to specific levels does not "really" 'digitize' the data. Doing an analogue to digital conversion and transmitting the data in the form of a stream of digitized information (as opposed to one out of 1024 possible levels, for example) is essential in order for the information to be 'digital' [in my mind].

Although restricting the data to particular levels might be a quantum leap forward in efficiency, it doesn't really make the processing of the data 'digital' {although one could argue that it HAS been digitized into 'BASE

1024', I think that is really cheating}.

In any case, it has been an interesting discussion to follow. Thank you BOTH.

--
bz    	73 de N5BZ k

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

bz+ser@ch100-5.chem.lsu.edu   remove ch100-5 to avoid spam trap

Class G or H. It's supply switching to reduce dissipation. Done it myself. Big amps may have 3 rails.

Graham

So what would you call them ?

WRONG. Class G and H use quite different circuits.

Graham

This arbument has come up before. Some people think that the quantized data must be represted as a "number" in order to be truly digital. Am moment's thought will show this is not so.

The "number of bits" is determined by the number of levels. Indeed, we could quantize at non-binary increments, if we wanted, and the data would still be "digital".

You're welcome. Thank you for reading and thinking.

create

confuses

I wouldn't call them "classes", just a name describing how they work or what they do.

They might be true, but that wasn't the point I was making. There is such a thing as "product differentiation", and you don't make yourself look different by appearing to copy someone else's feature.