Chapter and Verse on Valves

[Lurcher300b]

Class A/AB makes little sense in a push pull valve amp as has been said, unlike transistors valves don't turn off hard. A Solid stage output stage is normally a couple of emitter followers moving up and down by a single ended drive signal. A valve output stage is a transformer coupled LTP driven by a balanced signal. By definition class A is when devices always conduct, so in that sense all transformer PP output stages are class A. Its just a name. I have to admit though 20Jz to 100kHz is a rather impressive power bandwidth for a pair of transformers. Depends on the definition of flat I guess. Anthony does know what he is doing though, so they will be good amps.

[Vlad_Dracul]

Oh I quite agree , the copper amp appears to be a very well manufactured amplifier of its type, however I do have my doubts regarding the claimed specifications and performance, does this amplifier really conduct through the full 360 degree cycle of a waveform?

[Lurcher300b]

Depends on how pedantic you want to be. if you consider a complementary emitter follower output stage for a solid state amp, if the idle current is 100ma, then once the output swings positive so that the current into the load is more than 100ma then the negative device is passing zero current, and the opposite for the other half. If its biased such that at full output power the current into the load is less than the idle current, then neither device shuts off, and its class A. A valve is not the same, the valve is not directly connected to the load but through the output transformer. If you look at the curves for a valve, they get closer together as it gets near the 0v line, but never fully shuts off. because there is no direct transfer of current as in the SS output stage, you don't get the hard cutoff point in a valve amp. Buts thats where the pedant comes into it. In effect one valve is cut off as its not helping drive the load, but unlike a SS output stage there is not the hard transitions from two devices to one single device, so you dont get the same gm doubling that you do with solid state.

[Dr Bunsen Honeydew]

Now how many of you actually understood that post. I am not a valve man so there is good inf there for me, the effect of the output transformer is massive but I hadn't realised it allowed the PP valves to cross each other. Part of the problem is they are so different, essentially a valve amp uses voltage to drive the load (low impedance coils), hence the need for an output transformer, where as transistor amps are really using current to drive the load. Transistor output devices are there not for voltage gain, as little to none is provided there, they are to provide the current to drive the load. They are in many ways the equiv of a valve amp output transformer but without the inherent non linearities. In many ways a valve amp is more suited to the natural load of an electrostatic loudspeaker, as that requires voltage drive. BUT then they ruin it by putting two transformers in the way, stepping down in the amp than stepping up in the speaker. THOUGH whether a couple of thousand volts on your speaker lines is a good idea is mute

The complexities here are complex which is why marketing men and forum eeediots give you simplistic explanations of half truths, non truths, and bullshits.

[Lurcher300b]

Yes, its a easy mistake to miss the fact that solid state and valve outs stages are entirely different. I made the same mistake from the other direction. What really drives home how a valve push pull stage works is when you draw the composite curves, the effect of the two sets of curves for each of the output valves added together. You get very near straight lines within the overlap region.

You go from a single valve set of curves like this



To composite set of curves that look like this

[Dave Cawley]

So with a push-pull valve amplifier, if you measure the total current drawn, with no output and then full out put, is there a difference ? (not a ~50:1 difference you would get in a Class AB transistor amp?) If not then surely that is Class A ? Or can the design be made Class A "or" Class AB ?

Bizarrely I am very conversant with Class C !!

[Lurcher300b]

Well, the total current draw is a good way, you will find the total current doesn't change that much, not least as the inductance of the primary of the output transformer will keep current flowing. I guess the best way of saying is it class a or AB is if the output stage has a shared cathode resistor or not. Or maybe a better question is if it would work with a shared cathode resistor. If not, then AB is a better description. You know with C how hard the valve has to be driven to cut off, and C is often C2

[Dave Cawley]

you will find the total current doesn't change that much

Roughly how much ?

BTW I used Class C extensively in RF applications where it works well, probably the only place !

[Lurcher300b]

I will do some sims. I was thinking about this this morning. not done any C, I guess in that case you cut the valve off by driving the anode negative?

[Dave Cawley]

Class C was a BTW, another time maybe, but fascinating, 807's and all that !!

But, the real question is

you will find the total current doesn't change that much

Roughly how much ?