[Amps] Re: AB1 v. AB2

[bear]

Rich wrote:

> >Rich wrote:
> >
> >> >> >You can build an audio amp or modulator that is capacitively coupled
> >> >> >to the grids of the output tube and there is nothing you can do to drive
> >> >> >it into AB2 (grid current).
> >> >>
> >> >> How is DC grid bias applied to the grids?
> >> >
> >> >Separate minus voltage power supply fed via high impedance (resistors) to
> >> >prevent loading down the drive voltage...
> >> >
> >> **  What stops DC grid-current from flowing through such resistors
> >> whenever the grid is driven positive?
> >
> >The grid current that flows is so insignificant that no appreciable current
> >*can* flow, so the tube does not reach beyond AB1 in practical terms.
> >
> **  Eimac rates the maximum permissable grid bias ps source resistance
> for its tetrodes and pentodes.  This determines how much grid-current can
> flow.  In my experiences with AB1, there is a tradeoff in IMD when grid
> current flows -- i.e., the amplifier is driven into AB2.
> -  The argument that high resistance is used to prevent loading down the
> drive voltage does not hold water because, in RF service, the typical
> grid-terminating resistance is 50-ohms (or 200-ohms through a 1:4
> transformer) in order to present a 50-ohm termination for the radio.

That is not the *only* way to present a 50 ohm termination to the amp's input and
still not provide
a low Z impedance to the grid.

Typical is not terribly relevant to the discussion of amplifier class
definitions...

Grid current drawn = AB2 not AB1.

>
>
> >In practice the output signal is merely clipped when it reaches the point
> >where in an AB2 amp grid current would start to flow.
> >
> **  Would ?   The reality is that some grid current Does flow and IMD
> increases.

Sure IM increases, if you've just clipped the amp! The amount of grid current in
the situation I mentioned is so infintesimal as to not make any practical
difference -
the tube is not going to conduct any more due to that small an amount of current.

>
>
> >A good example of an audio amp like this is the good old Dynaco Stereo
> >70... the schematic of it and the MkII, MkIII and Mk IV which are all
> virtually identical
> >circuits are on the
> >web I am sure. Guarranteed this amp will not run into AB2 no matter how much
> >signal you put in the front end... : -)
> >
> **  I can not believe that no grid current will flow when the amplifier
> is overdriven.

Ok. :- )

But current isn't going to be sunk by the driver in that circuit.

>
>
> >If I get around to it, I guess I'll have to look and see what Bill Orr had to
> >say about this topic with respect to RF amps... I'm guessing there is some
> difference in the
> >circuit requirements in general, but honestly I can't recall anything about
> it at the
> >moment.
> >
> ** I recall nothing Orr wrote about this subject.

Whatever - I'm sure he mentions amplifier classes and operation... if not there
then
in another text.

>
>
> >In the end, if the amp can be run into grid current, then it is by
> >definition an AB2 amp,
> >no matter how you cut it.
> >
> -  indeed

Which is my point. AB1 != AB2.

>
>
> -  Rich..., 805.386.3734, www.vcnet.com/measures.

--
         _-_-bear
        http://www.bearlabs.com
      Silver Lightning Interconnects & more...