Ian Thompson-Bell wrote
>>> Has anyone thought of using a
>>> grounded grid amplifier as a mix amp?
>>
>> Yes. It seemed the obvious choice for the bus. An
>> alternative would be an ordinary gain stage with a low
>> value
>> grid resistor.
>>
>> I couldn't maintain interest in mixer design beyond
>> getting
>> the ideas straight. It seemed like a fool's errand, and
>> still does. I can't see why a valve mixer is a good idea.
>> A
>> box of mic amps, yes, with a state of the art SS mixer
>> amp
>> and chip earphone amp, maybe.
>>
>
> I guess it is the same reason some people prefer valve
> amps to SS ones - oersonal preference for the 'sound'.
> There are certainly a large number of pro and semi-pro mic
> pres that are aither all valve or use a valve or two
> somewhere in the chain to obtain that elusive sound.
>
> For myself, I have been a keen sound recordist for over 40
> years and even worked at Neve in the 70s designing pro
> mixers. Now that I am retired my main hobbies are music
> recording and audio design. Since I was 'brought up' on
> valves I am having great fun revisiting them and my main
> current project is to build myself a couple of all valve
> mic pres to supplement my existing recording kit
>
>
>> And I could only copy a respected valve mic amp,
>> considering
>> I've never done any recording and have no idea what they
>> are
>> supposed to sound like.
>>
>> And I noticed that no-one else has actually posted very
>> specific details. Coyness, in addition to the usual
>> animosities, was clearly going to frustrate any attempt
>> at
>> serious discourse.
>>
>
> That surprised me too. Maybe it was the thought of making
> money from the project that caused people's reticence. For
> me it is an interesting exercise in design. However, the
> design is just the start of the job. The physical
> implementation is at least as difficult.
>
>> And I don't know enough about noise to adopt a rational
>> strategy, and this particular mixer idea isn't a good
>> learning vehicle because I don't care much for the money
>> or
>> its man, for whom I suggest Eeyore is a good match, which
>> is what I said, and still think. The man doesn't want
>> what
>> he's asked for, and Eeyore won't provide it.
>>
>
> Noise is obviously a crucial issue in mixer design but the
> principles are applicable to any audio design. The first
> stage is the crucial ones and for valves the most
> straightforward way of obtaining a respectable noise
> performance is to use a 10:1 mic transformer which will
> have a NF of a dB or so and nearly 20dB of gain which
> makes the noise performance of the first valve almost
> irrelevant. Thereafter it is mostly a matter of ensuring
> the signal remains adequately above the noise floor.
I can see the reason for a mic amp, as I said. It's the
mixer part that I believe is better done with SS. If the
output of the mic amp is just right, then it will no longer
be right after a valve mixer.
It's interesting to compare cathode input with grid input
for the same stage. For example, an ECC88 with 47k at its
anode and 1.1k cathode resistor has an input impedance to
the cathode of around 650 ohms, and for a 100mA input gives
about 2V output with an impedance of around 20k. So grounded
grid seems a likely candidate if you are looking for a low
impedance input.
The same stage with input to the grid, with added grid leak
of 650 ohms and its cathode resistor bypassed, gives about
the same result, with respect to input impedance, gain, and
distortion, but has an output impedance of around 4.5k.
Seems an anomoly? Where's my mistake? How do they compare
for noise?
The first stage usually is the main culprit, in a circuit
with several cascaded gain stages. But is that still the
case if half way through the circuit the signal is
attenuated to the same level as the input, and then
restored? Is the stage that recovers the gain equally
im****tant?
Ian
|
