On Mar 25, 10:05=A0pm, legg <l...@[EMAIL PROTECTED]
> wrote:
> On Mon, 24 Mar 2008 15:16:36 -0700 (PDT), Engineer
>
>
>
>
>
> <junk2...@[EMAIL PROTECTED]
> wrote:
(snip)
>
> You already have resistors in the cathode circuit that could be used
> to determine quiescent current balance directly. What you might do is
> adjust static grid bias, as this does not affect AC gain of the
> mismatched devices, and can be achieved using lower-power variable
> resistors.
>
> RL- Hide quoted text -
>
> - Show quoted text -
I think you're suggesting a separate bias adjust like "fixed" bias -
it could work, too! But it needs a bias supply... then you might as
well go for full fixed bias regime and use 10 ohm precision resistors
in each cathode as a sensor with zero differential cathode voltage as
the aim. Gets a bit more B+, too!
I was trying to do this while using simple cathode bias. I conclude
that the balance measurement can't be done at the cathode.
Of course, you can calculate the currents if you know the cathode
resistors but:
1. You don't know them - the 100 ohms pot changes them differentially
each time you adjust it. That's it's job.
2. Even if you knew them (measure each time...) the balancing would be
iterative, a big pain! i.e. measure each voltage, calculate currents,
note error, change 100 ohm pot a bit, determine each new cathode
resistor, measure each voltage, calculate... etc.
I want a direct, one step, differential, centre zero set-up as we can
have for fixed bias. Think I'll just build a bias supply!
BTW, I'm not talking about AC drive balance, that's another topic...
I got lucky, my P-P class AB OPT was driven equally right up to
clipping at DC balance.
Cheers,
Roger
|
