😕
No, they're defined by the percentage of time they spending processing relative to the time the input signal is present for.
You can not define a Class A stage as one that processes an input signal for a certain amount of time. It is a stage that processes an input signal for 100% of the time the signal is present.
Otherwise you could say, well a stage that processes an input for 2 seconds is a Class A stage. But one that processes it's input for 1 second is Class AB. It doesn't work, because the input does not remain at a constant frequency. What's more, the period of each cycle is usually hundredths of a second, or less.
To account for a varying input signal, the definition can only use a percentage of processing time relative to the input signal it's self.
A system in which both halves conduct for more than 180 degrees, or 50%, of the time never stop processing the input signal of the entire system. It is a Class A process. Processing occurs for 100% of the input signal.
For an entire PP stage to become Class A/B, processing of the entire input signal must be less than 100%. And so, at least one valve in a complementary pair must move into Class C operation (Relative to the entire system, but Class A/B or worse relative to it's own input), and process it's own input signal for less than 50%, or 180 degrees, of it's cycle. In doing so, the entire stage no longer processes for 360 degrees, or 100%, of it's entire input signal.
Overshooting the cross over point means that one valve is essentially working against the other, until it switches off. As it has begun operating in a section of the entire input signal, usually of an opposite polarity, that was supposed to be processed by it's complementary valve. In a PP layout, you can't have both valves pushing and pulling in unison. Amplification is therefore none linear during this period of time. It is distorted.
Except that setting the parameters of the phase splitter incorrectly means that the entire input signal is unfairly distributed between each valve. The time taken for the phase splitter to switch the polarities to their desired location is exactly the same as the time it takes the person in Andy's description to flick the light switch on and off.
Cross over problems will also arise from other parameters of the processing valves themselves being set incorrectly, but the probablity of an error occuring increases as the density of the original information increases. There is more information held within the input signal of an amplifier than the output due to passive filtration and active distortion throughout the system. Not only this, but the output is of a higher amplitude than the original. Perhaps even tens of thousands of times higher? Therefore, the density of the information in the input is substantially greater than that at the output, hence pure silver interconnects for $40k per 8ft, if there can ever be a reason for spending that much a cable! Heisenberg is everywhere, all the time, and he hate's things working properly. 🙂
Granted. I have actually gone as far as learning to play an instrument myself, which is a lot more than the majority of audiophiles seem to bother with. I know a number people who are 14 - 20 and have taught themselves to play instruments without ever seeing a teacher. It requires no special heaven sent gift. By the same reasoning that says it does, someone who is colour blind has no right paint a picture, or enjoy art. Beethoven was deaf when he wrote some of the most beautiful pieces of music ever.
If you want the output to look identical to the input then that's true. But we've already been over this. I asked one very simple question relating to SE stages,
Had I been answered with,
We wouldn't be here now. In fact, I can even go futher back to my very first post, but I am working on the side of caution.
The reason we are here is that I was not given such an answer. This threads first reply, from dhaen, contained the statement,
I'm not trying to be unfair, but more than just one person made such suggestions. Am I wrong to think that such a statement is suggesting that an SE stage may only run in Class A? Had I said that in one of my first replies, I would I have been thought of as arrogant and someone who does not listen to others. I would have also been blindly following one persons advice. Just as I have been told not to do in another thread, I think, by Hans. I asked one question of why this was so. I have receive a number of replies, but all from a very limited group of people.
I can see only two reason for this. One, those who initially made such suggestions think I'm an arrogant person who won't listen to them anyway and so they don't want to waste their time replying. I doubt this, because I know that pretty much everyone on this forum is kind hearted. Or two, those who made the suggestions do not have any reasoning or support for why this is the case.
When I began to suspect this, I tried to tactfully see if anyone would provide me with a reason. No one did. Rather than acusing these people of blindly following the advice of someone else, I simply continued to ask questions relating to SE stages only running in Class A.
The reality is, if someone had the reasoning to support themselves, they would have answered. They didn't, and so I got no straight answer to the question. And now here we are. 😀
Since this is the forum for tubes, many of the members here seem to actually prefer the warm sound of valves. I'm sure that if you where to strip all the problems of amplification away and produce a perfect, true to theory, Class A amplifier, you would loose a massive degree of this warmth.
People who like valves, myself included, do not, in general, want a genuinely perfect amplification process. Otherwise the music seems to become cold and sterile.
If people only ever wanted a genuinely perfect Class A stage, there would be one single design for an amplifier and no other method would matter. That's obviously not the case, since most guys don't even seem to be able to settle on which kind of cable to use for interconnects.
My wish to create peak to peak wave clipping distortion is no different to the subtle distortions valves create that make everyone on this particular forum want systems based on them. If people wanted perfect interconnects, cryogenically cooled super conductors would not seem like a joke. Insulated in a vacuum, cryogens can keep an item below their own boiling points for weeks. Yet no one has any cryogenic cabling in their stereos... that I know of... yet!
If all audiphiles strive for is a perfect Class A amplifier, then they'd all, almost certainly, have entirely solid state systems. They don't.
Again, I can do nothing but thank you for actually replying to my posts.
No,, that is not correct, the class of operation is refering to each device, not the complete amplifier as such.
Time, (relative time) and percentage is more oir less the same thing, usually class A, AB, B aso are defined in what the conduction angle is, class A 360 degrees, class B 180 degrees aso.
No, they dont work against each other, they work in phase as seen from the output and the output of the tubes add to each other.
BTW,
Your definition of class A is not correct, class A is defined where the amplifying device conducts all the time during the signal enevelope, it doesn't mean that 100% of the input must be amplified linearly or that the output signal is distorsion free, a class A amplifier with 10% distorsion due to unlinearities i still a class A amplifier.
Regards Hans
Hi,
While I agree that alot of people are attracted by the so called "warm" sound of valves, this warmth is nothing more than distortion.
This is also the same type of distortion that attracts people to SET amplifers, quite probably to compensate (unconsciously or not) for distortion from their source(s).
A well designed amplifier, and I mean one that's truthful to the source not the euphonic type, using valves will sound crystal clear and transparent: warm when the source is, analytical or edgy when the source is and utterly correct when the recording is exactly that.
This goal can be achieved using class A (easier but not efficient), B will show much higher distortion, AB combines pros and cons from both plus it's own artefacts, namely crossover distortion and what have you.
In order to do that one has to be able to understand where this warmth comes from.
As said above this is caused by harmonic distortion typical for a non- global feedback SE stage, apply a NFB loop around it and most of that warmth will be gone.
In a PP Class A amp much of this warmth/distortion is automatically nulled out due to the nature of the PP amplifier itself. (remember the output works in a balanced)
So, at the end of the day an amplifier can be taylored to please every single pair of ears, it's just a matter of balancing the harmonic spectrum of the distortion.
Even an amplifier with very low distortion figures obtained by application of global negative feedback will still show a tonal balance depending on the harmonic spectrum of whatever distortion that was left there prior to closing the FB loop in the first place.
IOW, I still have to hear the first amp that suddenly sounds sweet and rich after application of FB if it didn't sound that way already. YMMV.
An invaluable instrument for playing this kind of games is a spectrum analyser as it gives one a visual clue of what's going on.
Cheers,😉
While I agree that alot of people are attracted by the so called "warm" sound of valves, this warmth is nothing more than distortion.
This is also the same type of distortion that attracts people to SET amplifers, quite probably to compensate (unconsciously or not) for distortion from their source(s).
A well designed amplifier, and I mean one that's truthful to the source not the euphonic type, using valves will sound crystal clear and transparent: warm when the source is, analytical or edgy when the source is and utterly correct when the recording is exactly that.
This goal can be achieved using class A (easier but not efficient), B will show much higher distortion, AB combines pros and cons from both plus it's own artefacts, namely crossover distortion and what have you.
In order to do that one has to be able to understand where this warmth comes from.
As said above this is caused by harmonic distortion typical for a non- global feedback SE stage, apply a NFB loop around it and most of that warmth will be gone.
In a PP Class A amp much of this warmth/distortion is automatically nulled out due to the nature of the PP amplifier itself. (remember the output works in a balanced)
So, at the end of the day an amplifier can be taylored to please every single pair of ears, it's just a matter of balancing the harmonic spectrum of the distortion.
Even an amplifier with very low distortion figures obtained by application of global negative feedback will still show a tonal balance depending on the harmonic spectrum of whatever distortion that was left there prior to closing the FB loop in the first place.
IOW, I still have to hear the first amp that suddenly sounds sweet and rich after application of FB if it didn't sound that way already. YMMV.
An invaluable instrument for playing this kind of games is a spectrum analyser as it gives one a visual clue of what's going on.
Cheers,😉
Either way, it's exactly the same process, just viewed from another perspective within the amplifier.
Defining an entire system by what an individual unit within it does is not a fair method either. As one tube may be operating Class A whilst the other is operating is A/B, B or even C. In turn, the end result will be an amplifier that does not operate in Class A, since half of it's processing is being done out of Class A.
If I were to describe efficiency, I would do so by using a percentage of desired output to the percentage of usable input. I guess there are a number of ways to think about it. But as soon as something becomes relative to something else, it is only really one step for being a percentage. If you're happy with the digital notation of 0 being 0% and 1 being 100%, this step need not be taken.
Do you not find it easier to deal with percentages of time as opposed to degrees?
My mistake. Sorry EC8010. 🙁
[kicks himself on behalf of EC]
You are correct to critise me for not mentioning distortion caused within a processor that operates for 100%, or 360 degrees 🙂, of it's input cycle.
I wasn't really trying for a definition to cover every possibility, but rather to let people reading my posts understand how I was relating percentages to Classification. I could have said,
"A Class A stage must be perfectly linear and distortion free."
Which has been suggested in discussions on Class A.
Just for the excitment of quoting RCA, below are the defining notes they make on classification.
I appologise for any mistakes I've made. I visited the Solid State forum last week and began reading a thread on LM317 regulators. This thread, when I last saw it, was well over 50 or so page long and composed of lots of short, somewhat meaningless, replies relating to legalities of Copyrights and people's personal problems. I never got past about the 8th page. In attempt to prevent my threads becoming 50 page monsters that don't help anyone, I try to cover as many points as possible in one post.
My last post actually reached the 10,000 character limit. So, some times I can end up drifting if I don't keep a careful watch out for mistakes. I don't know about you guys, but I'd rather get a cup of coffee and read one long post instead of having to continuously return to read one line answers.
I believe this thread has now pretty much covered everything it set out to. I'll start finalising some real world values tomorrow for the stage I would like to build.
Thank you very much to everyone who has helped! I have learnt a lot!
🙂
Definitly. I have a Fluke 123 hand held scope at the moment. I'm thinking of buying the interconnection upgrade for it so that I can run spectrum analyses from my PC. The FlukeView software has quite a nice bar graph method of doing this. I'll have to play around with it a bit more, but, at the moment, I'm not sure if you can change the mode to give a true line graph analysis of the spectrum with a high density, continuous, scale.
Just out of curiousity, I tried searching for spectrum analysers one ebay. 😱
$30,000
But that was an analyser for the GHz bandwidth I guess! 🙂
On an almost remote point, it's quite interesting to just view the waveform from my guitar while an open string is playing. When you touch the first harmonic, the waveform instantly double's it's self, and so on.
If it's of interest to anyone, there is an instrument tuning tool available on the internet called TuneLab. It's free to download. Try searching for it on google.
It gives a line graph spectrum of the input over a frequency bandwidth, to allow instrument technician to tune strange off set, stretch tunings temperaments.
If you set the note of interest quite high, and ask it to display the full spectrum, you can get a graphical view of the input from at least 1 - 10,000Hz. It has a somewhat slow refresh rate, but it would be adequate for simple tests if you didn't want to buy an analyser for just one particular amplifier.
I thought it was quite fun to just talk into it for a while and to how my voice signed it's self. There are some very distinct frequency peaks for certain formations.
Tried a powerbrake?
I have been reading this thread as I am interested in learning how to bias my marshall amp when I revalve it but I noticed that the issue here seems tobe to do with playing big amplifiers at bedroom volumes. I have been using a Marshall Powerbrake for around ten years now. It means I can play my marshall at any volume. I have the mastervolumes set at 8 all the time. This gives me the nice fatness and harmonics generated by the el34 when they are running hot. I wouldn't consider playing without one unless I bought/built a tiny amp. Just an idea!
I have been reading this thread as I am interested in learning how to bias my marshall amp when I revalve it but I noticed that the issue here seems tobe to do with playing big amplifiers at bedroom volumes. I have been using a Marshall Powerbrake for around ten years now. It means I can play my marshall at any volume. I have the mastervolumes set at 8 all the time. This gives me the nice fatness and harmonics generated by the el34 when they are running hot. I wouldn't consider playing without one unless I bought/built a tiny amp. Just an idea!
- Status
- Not open for further replies.