Talk:Power amplifier classes

Peer review of Amplifier

Archived at [[1]]. --Wtshymanski (talk) 21:25, 9 May 2017 (UTC)

Who?

I wonder who first classified amplifiers this way. Who hands out new class letters? Or was this another gift of the saucer people? (So little history on WP, so many data sheets....) --Wtshymanski (talk) 16:17, 28 June 2017 (UTC)

As far as I know, it goes back to at least vacuum tube days, at least for A, AB, B, and C. Gah4 (talk) 06:04, 2 May 2023 (UTC)

Is it "class A" or "class-A"?

I really don't know which is right, but all over wikipedia, we use both and I'm guessing we need to establish that one is preferred over the other, or at least establish that both are fine but be consistent in each article in some way. Dennis Brown - 2¢ 16:23, 18 October 2017 (UTC)

IEEE 100 under the entry "Amplifier class ratings" likes the hyphen (or maybe it's an endash?) but I've gotten into so much trouble quoting IEEE standards as authorities that I'm loathe to proceed on the hyphen (or endash?)-inserting campaign. My advice is to go along with what the claque recommends...it may not be "right" but it will be "Wiki-right" and that's the only thing that matters here. --Wtshymanski (talk) 18:23, 18 October 2017 (UTC)

I'm fine with that, but it all too often varies within an article. We need an RFC, if only because I haven't seen a drama fest over punctuation in a long time ;) Dennis Brown - 2¢ 19:03, 28 October 2017 (UTC)

I would like to see a resolution. My guess would be it should be without a hyphen. But what would the supporting argument be? Some references:

MOS:HYPHEN (Wikipedia style manual)

Hyphenated compound modifiers

How to use hyphens and dashes

I couldn't find a matching rule for this case.

First task: What is the classification of the noun "class A"?

--Mortense (talk) 21:29, 13 November 2020 (UTC)

Image incorrect?

The image given as a schematic for a class-H amplifier appears to be incorrect. It does not work correctly if put in a circuit simulator, and on top of that it doesn't make sense to me as an electrical engineer; the diodes don't actually do anything that I can tell. Can someone source this design and identify the problem, or explain how it works if there is no problem? 104.167.148.149 (talk) 18:20, 23 October 2018 (UTC)

Class B Image Misleading

The image shows correct waveforms, but connecting a BJT like that is class C. I had someone argue that a zero biased BJT is class B and used this wiki as a reference. Conduction angle alone determines class, not any particular device bias method. I suspect it is only intended as simplified, but it has obviously misled at least one poorly informed electrical engineer, or so he claimed. Perhaps it should show undefined devices instead of BJTs, Regards, -- Steve -- (talk) 17:58, 5 August 2020 (UTC)

I don't agree as the diagram is of an "ideal" class B. Class C uses a low-side transistor and (usually) a tank circuit on the high side, it's not push-pull. Practical circuits use an offset in the driver to account for the imperfect gain curve (similar to AB)

Class C is wrong

Class C amps are *always* tuned circuits. If there is a resistor in the collector line then it's a class A amp, whether or not it amplifies the full wave. The idea of a class C output stage is to use the tuned circuit to recover the missing half of the waveform, thus making it more efficient and using the tuning (as stated) to remove unwanted frequencies. Generally the class of amp is defined specifically by the output stage. 89.104.38.178 (talk) 08:36, 2 September 2020 (UTC)

I am sorry, but while some of the things you point to (such class C using tuned circuits) are / may be true or common, this is contrary to the long standing amplifier class *definitions*.

Amplifier class is defined by operating point, ONLY, not bias method nor collector interface. A web search will turn up any number of the same definitions.

Class A conducts the whole cycle and can have a transformer primary in the collector even though a resistor may be common in smaller signal applications and where no DC isolation is needed with the next stage or load. A Class A stage can drive a class AB output stage through a transformer and this is common.

Class B conducts for 180 degrees, Period. The bias must be set up for that for the specific device to get that 180 conduction. These will usually be push-pull, but due to the likelihood of crossover distortion in audio, Class B is not used there with AB guaranteeing there is none under all operating conditions and transistor characteristics. Class B is rare because of the crossover issue easily solved with class AB biasing.

Class AB conducts between 180 and 360 and it is most commonly just slightly more than 180. These are commonly used in linear power amplifiers for Audio output and Radio (Single SideBand and AM) Transmitters. In radio transmitters it can be single ended or push-pull with a tuned output circuit required for harmonic suppression. The AB gives higher efficiency than class A and better linearity than class B. Whereas, in audio it must be push-pull to an output transformer.

Class C conducts for less than 180 and is used pretty much exclusively for RF specifically because of the highest efficiency and will, therefore, have a tuned output for harmonic suppression (and most commonly tuned input).

Regards, -- Steve -- (talk) 15:30, 2 September 2020 (UTC)

Also, each and every stage of a multi-stage amplifier string has some class, not just the output stage. -- Steve -- (talk) 15:36, 2 September 2020 (UTC)

Class H

Class H is confusing. Maybe it's meant to be, I have a basic understanding of electronics although the maths is a bit beyond me. One problem that I don't think is clearly explained, is how does the power supply itself stay efficient? As the article mentions, it's essentially an amplifier powering another amplifier. So how does that solve a problem? And surely any noise in the power supply would come straight through the actual amp's output? So what are you gaining putting a little amp in a big one? If you can explain this like it's Lego or something it would be appreciated!

Class H could also do with being split up from one giant block of text. A step by step explanation of what happens at each stage and why would be great.

84.70.145.70 (talk) 17:53, 8 March 2021 (UTC)

I agree that the section on class G and H is very confusing. Bubba73 ^{You talkin' to me?} 04:15, 12 July 2022 (UTC)

A switching power supply is commonly an oscillator running an amplifier and driving a ferrite core transformer. In the usual case, one wants a constant output voltage, but that isn't required. Also, a linear amplifier with proper negative feedback can work over a wide range of supply voltage. That is, noise (voltage variations) in the power supply do not come through. This is given by the Power supply rejection ratio. So, all you need to do, is configure the switching power supply to increase when more output voltage is needed, and decrease when it isn't. Gah4 (talk) 21:09, 12 July 2022 (UTC)

That is if you are an electrical engineer. The section needs to be written for people who aren't electrical engineers. Maybe some of those details can be in a subsection. Bubba73 ^{You talkin' to me?} 23:39, 12 July 2022 (UTC)

I was only commenting on the how does the power supply itself stay efficient? No comment on the need to change the page. Gah4 (talk) 06:29, 2 May 2023 (UTC)

vout?

"Vout" is used several times in the article. Is it the output voltage? This needs to be clarified in the article. Bubba73 ^{You talkin' to me?} 03:57, 12 July 2022 (UTC)

Class F needs diagram

There's a lot of verbose descriptions of waveforms, which is quite easy to get lost in. Nothing wrong with the content, but a diagram will make it go over much more smoothly. 101.98.178.115 (talk) 20:59, 7 February 2023 (UTC)

Class AB diagram

The 'Ideal class AB amplifier' diagram appears incorrect. Surely the input signal should be applied directly to the base of the lower transistor? The diagram should also include a resistor from the base of the upper transistor to the positive supply, as is the conventional arrangement. It seems to me that applying the input signal to the junction of the two diodes will prevent both transistors from turning on. Rdphillips (talk) 19:31, 5 March 2024 (UTC)