class A power amps and how they work?

[#12]

Referring to post #8, You have heard class a amps. Fender Champ and Supro are 2 examples.
I can't name which artists used those amps, but I'm sure you've heard them if you're more than 12 years old.
I suspect you are going in very much over your head, but even if you fail, you will learn a lot.
Good luck.

 

[Nigel Goodwin]

Referring to post #8, You have heard class a amps. Fender Champ and Supro are 2 examples.
I can't name which artists used those amps, but I'm sure you've heard them if you're more than 12 years old.
I suspect you are going in very much over your head, but even if you fail, you will learn a lot.
Good luck.

I don't really see the relevence of class A single ended valve amps to this thread?.

 

[Jon Wilder]

As has already been stated Class A for anything high power would be highly inefficient and impractical. That being said -

The things that need to be considered when designing ANY kind of amp are -

The load impedance you plan to drive
Voltage required on said load to get the power
Power supply current
Voltage/current handling of the output devices you plan to use

If the power supply doesn't have the required voltage/current you're not gonna get the power. If the output devices cannot handle the current when on, the voltage when off, and have the power dissipation rating sufficient enough to dissipate the amount of power they will be dissipating during the "transition period" (i.e. the period of time in which the output signal is transitioning from positive to negative and vice versa), you're gonna blow stuff up.

A lot of people seem to think that the output devices are what make the power while the power supply just serves to turn the output devices on. Not the case. An audio amplifier is just a modulated DC power supply...an AC power inverter if you will. Via the output devices, you're using an input signal to convert the DC from the power supply into a higher voltage/higher current copy of itself. Ultimately the output power comes from the power supply...the output devices just use the input signal to pass and control this current to the load.

So...this means you'll need to design the power supply to supply the required peak voltage to the load to get the power as well as deliver the current that the load will pull at maximum output voltage. Then you need to select output devices that can handle the supply voltage AND the load current. If one alone won't handle all the current, you strap multiples in parallel with low value resistors for balanced current sharing on each side of the amp. Most modern discreet component amplifiers use high current MOSFETs for the output devices.

The most ideal is to have a dual rail/dual polarity supply so that you can DC couple the speaker to the output (you may need to incorporate a way to adjust the zero offset into your circuit to ensure that the output is at 0V when idling). If you go single rail, you'll need a rather large value capacitor to block the DC at the output from the speaker (this was done on the Acoustic 470 guitar amplifier...very old design method).

 

[#12]

Nigel: Cracken mentioned a distortion pedal and a Celestion speaker in this thread, both of which are popular with guitar players. These things lead me to the idea that cracken is thinking about guitar applications. Cracken also said he has not heard a class a amplifier. I believe he has, but doesn't know it. I know of a few guitar amps that were class a, but they all used valves. I can not think of a class a transistor amp that was used for guitars, so I could only refer him to valve amps if he wants to hear an example of class a amplifiers.

If you know of a class a transistor guitar amp, I would like to have that information, and cracken could go listen to a class a amplifier for his own education. Please enlighten us.

 

[audioguru]

A good modern solid state amplifier has a very wide frequency response and distortion so low that it cannot be heard. So a good class-A amplifier sounds exactly like a good class-AB amplifier. Why heat the world with a class-A amp? Why listen when there is no difference?

 

[Jon Wilder]

The whole "Class A" thing came about as a sales pitch from the valve amp guru/cork sniffer community right along with terms like "sag", "OT saturation", "crossover distortion", "carbon comp resistors", etc etc. Funny thing is that they talk about Class A amps and how they "sag" yet Class A valve amps don't sag at all.

Class AB all the way for me.

 

[Nigel Goodwin]

Nigel: Cracken mentioned a distortion pedal and a Celestion speaker in this thread, both of which are popular with guitar players. These things lead me to the idea that cracken is thinking about guitar applications. Cracken also said he has not heard a class a amplifier. I believe he has, but doesn't know it. I know of a few guitar amps that were class a, but they all used valves. I can not think of a class a transistor amp that was used for guitars, so I could only refer him to valve amps if he wants to hear an example of class a amplifiers.

Yes, he's talking guitar amps.

If you know of a class a transistor guitar amp, I would like to have that information, and cracken could go listen to a class a amplifier for his own education. Please enlighten us.

There are no class A transistor guitar amps, what would be the point? - and it's pretty pointless listening to a class A valve amp, because it's highly distorted and low quality anyway.

 

[sonuv]

what are class A,Class B ,Class AB etc

 

[Nigel Goodwin]

what are class A,Class B ,Class AB etc

The biassing conditions for the amplifier - almost all transistor audio amplifiers are AB - although that's now changing quite a bit to class D, particularly for high power stuff.

 

[Jon Wilder]

and it's pretty pointless listening to a class A valve amp, because it's highly distorted and low quality anyway.

And it is that very distortion characteristic that blues and rock guitarists are after.

Although the distortion characteristics of a valve are much more desirable than that of a transistor.

what are class A,Class B ,Class AB etc

Class A amplifiers are "single ended" (i.e. only need 1 output device) and that 1 output device amplifies the full sine wave. When the amp is in the quiescent state (i.e. when there's no input signal present...which we will call "zero signal condition"), the output device is dissipating its full rated dissipation. When under signal average dissipation drops to about 1/2 what it dissipates under zero signal conditions. Because it is single ended, there is no offset current in the output transformer to keep the idle current from magnetizing the core so the OT must have a sufficient air gap between the coils and the core to minimize idle magnetization. Also, in a typical Class A amp the power supply must be spec'ed to deliver LOTS more power than what the amp will actually put out (something over double the output power) as the amp dissipates its full plate dissipation rating at idle, which ends up being a bit higher than what ends up at the output under signal. Under signal, 1/2 of this power is handed off to the load while the valve dissipates the other 1/2.

Because Class A amps only transfer 1/2 the available power to the load while the valve dissipates the other 1/2, they are considered to be highly inefficient comparted to Class AB and Class B.

Current draw from the supply on a Class A amp is constant. When under signal, the voltage at the valve plate drops, which increases the voltage dropped across the OT while decreasing the voltage dropped across the valve at the same time. This trading off of the voltage drops between the load and the valve "hands off" the power to the load that the output device was dissipating at idle. In other words, you're just changing which device the supply voltage is being dropped across when under signal.

Class AB amps are push pull. Each output device passes current to the load for more than 1/2 but less than the full swing of the sine wave. This means that there will be an "overlap region" where both devices are on at the same time until at some point in the sine wave swing they transition into Class B mode, where one of the output devices turns off (goes into cutoff) while the other output device is still conducting. They alternate on/off at opposite times. On a transformer coupled amp, the center tap of the transformer primary is connected to the + rail while the start/finish wires go to the output devices. Because one side of the primary appears to be wound BACKWARDS from the other relative to the center tap, current through the transformer primary flows in the opposite direction, which reverses the magnetic polarity of the magnetic field, which induces a negative going swing at the secondary while the other induces a positive going swing at the secondary. With both output devices alternating on/off at each end of the transformer primary, this is how you end up with a complete sine wave at the output.

When at an idle state, both sides of the OT primary are pulling current through the output devices. Each output device is biased so that they are only dissipating roughly 70% of their dissipation rating while this is happening. Due to both sides of the OT primary pulling current, with one side wound backward from the other relative to the center tap this causes one side of the primary to be magnetized at an opposite magnetic polarity than the other side, which cancels the core magnetization at zero signal condition, which means no air gap needed on the transformer to reduce idle magnetization of the core.

Under signal at full output, damn near all of the available power is handed off to the load while the output devices dissipate very little at the peaks of the sine wave, which makes for a much more efficient amplifier. Output devices dissipate the full available power 1/2 way up the sine wave swing, which ends up being DOUBLE what the output device is rated to take, but due to being in cutoff for 1/2 the swing the average dissipation ends up equaling the output devices rated dissipation.

Current draw on a Class AB amp is constant when in Class A mode, then increases at the point in the swing where the output devices transition to Class B mode.

Class B is just like Class AB, but with no overlap region. When one output device is on, the other is off, and they switch on/off at opposite times. Each output device only passes current for 50% of the sine wave swing. Of course, on a valve design there still needs to be an overlap region to mask the non-linear region at the bottom of the response curve in order to eliminate crossover distortion.

 

[Nigel Goodwin]

And it is that very distortion characteristic that blues and rock guitarists are after.

Exactly - it's nothing to do with class A - just low quality valve sound.

 

[audioguru]

And it is that very distortion characteristic that blues and rock guitarists are after.

Some old geezers who played electric guitars in 1959 liked the sound from the only available amplifiers (with vacuum tubes) that was played on "horrible-sounding" speakers of the day. The old geezers had their high frequency hearing destroyed by loud sounds from guns of recent war. So they couldn't hear the awful distortion that was produced.

Why do Brits call a vacuum tube a "valve"? A valve is mechanical.
Maybe a "valve" in Britain has no vacuum?

 

[Jon Wilder]

Why do Brits call a vacuum tube a "valve"? A valve is mechanical.
Maybe a "valve" in Britain has no vacuum?

Actually the British valve nomenclature better describes how a valve operates. "Vacuum tube" just denotes one of their construction properties (i.e. they possess a vacuum).

When a valve is sitting at idle, a static DC bias voltage that is negative relative to the cathode is applied to the control grid. The static plate current is dependent on the voltage applied to the control grid.

When you inject an AC signal at the grid, this modulates the DC bias at the control grid above/below its static value, which causes plate current to rise/fall with this modulated DC bias voltage at input signal frequency. By doing this. essentially you are opening/closing a "voltage controlled current control valve", which is exactly how triode, tetrode and pentode valves work.

 

[audioguru]

Actually the British valve nomenclature better describes how a valve operates.

A transistor is also a "valve". Maybe the Brits should have called it a Valve II or a "faucet". The American actress??

 

[#12]

I suspected there are no class A transistor guitar amps because it's so counter intuitive to an engineering brain to try to dissipate that much power from a transistor when it's so much easier and cheaper to build a class AB...but I wasn't sure. (Thanks Nigel.) Besides, the distortion from a class A transistor amp would be so harsh that even a guitar player wouldn't like it!

Meanwhile, I have emailed a musician friend asking for names of popular songs and artists that used class A amps so that cracken might realize he has heard class A amps, or go look up a few songs and educate his ear. I believe that cracken will benefit from knowing what to expect from a class A amp before he tries to build one. That is why I would bother to make this effort.

 

[Nigel Goodwin]

Why do Brits call a vacuum tube a "valve"? A valve is mechanical.
Maybe a "valve" in Britain has no vacuum?

Why do Americans call them 'vacuum tubes'? - it's a British invention, why change the name?.

 

[Nigel Goodwin]

I suspected there are no class A transistor guitar amps because it's so counter intuitive to an engineering brain to try to dissipate that much power from a transistor when it's so much easier and cheaper to build a class AB...but I wasn't sure. (Thanks Nigel.) Besides, the distortion from a class A transistor amp would be so harsh that even a guitar player wouldn't like it!

Distortion from a class A transistor amp would be even LOWER than a class AB one, that's the point of class A - but could you hear the difference between 0.01% and 0.005%, or something equally stupid.

Meanwhile, I have emailed a musician friend asking for names of popular songs and artists that used class A amps so that cracken might realize he has heard class A amps, or go look up a few songs and educate his ear. I believe that cracken will benefit from knowing what to expect from a class A amp before he tries to build one. That is why I would bother to make this effort.

You still seem confused by class A - there are no transistor class A guitar amps, there's no point whatsoever. There are small class A valve amps, but that's because class A is the only way to make a single ended amplifier. Valve amps provide the distortion many guitar players are looking for, class A is no better than class B, it's simply the only way to make a single ended amplifier.

In push-pull valve guitar amplifiers (any decent size valve amp) it's normal to use different anode loads in the phase splitter, this adds loads of extra distortion, over and above that inherent in the valves.

 

[Jon Wilder]

In push-pull valve guitar amplifiers (any decent size valve amp) it's normal to use different anode loads in the phase splitter, this adds loads of extra distortion, over and above that inherent in the valves.

If we're talking about the common long tailed pair phase inverter, that has nothing to do with "distortion".

It is more an attempt to balance the phase inverter output symmetry between the two sides than anything else. The most common setup is the 82K/100K setup with the 82K on the inverting side of the PI to reduce the gain of that side to more closely match that of the other side. Increasing the tail resistor value also helps to balance the output symmetry as well.

Unbalancing the PI causes the power amp to cancel less of the power amp generated even order harmonics. Harmonica players benefit from this more so than guitar players.

 

[#12]

Nigel: I was referring to the guitarists habit of driving every amplifier into distortion mode. If they do this with a class A transistor amp, I'm sure they wouldn't like the sound.

I didn't ask my friend about songs made with a class A transistor guitar amp, as you said, there are none. He will reply with a list of songs made with class A valve amps. However, if cracken says he does not wish to examine some class A guitar sounds, I will stop this apparently intrusive and unwelcome contribution.

 

[audioguru]

Nigel: I was referring to the guitarists habit of driving every amplifier into distortion mode. If they do this with a class A transistor amp, I'm sure they wouldn't like the sound.

Distortion is high frequency harmonics. But most guitar musicians are deaf to high frequencies.
That is why they like the sound that is awful to normal hearing.