Richard Clark wrote:
On Tue, 30 Nov 2010 13:50:35 -0800 (PST), JIMMIE
wrote:

Owen, the meter is just a 1 to 100 linear scale meter on on 100 uA
movement. Ive used it for years just to "guesstamate" VSWR. Not really
as inaccurately as one may think. My thought was to cal a scale for
the "full scale set" knob so that when the meter is adjusted for full
scale the knob position will indicate power. Part of the reason for
doing it this way is that it is a nice large good quality meter and I
dont want to risk damaging it by taking it apart. I may have other
uses for it later on. Im thinking "linear scale capacitance meter
similar to the heathkit model. I intend to use an NE555 osc instead
of vacuum tube osc should I do this .

Jimmie

Hi Jimmie,

On reading this, several thoughts came to mind. You write about
having used a linear scale to guesstimate SWR. Sounds good and it
immediately leads us to an existential question:
"What value is there in knowing the value of SWR?"

The first motivation following a glance at the SWR meter is to LOWER
the SWR, for whatever value it may reveal. In this sense, the value
is a trivial consideration - relative indications are enough to
achieve the goal.

Using the feed from a Bruene style detector pair into a Log-Amp will
give you a power response in a linear scale.

These days, unless you're deliberately trying to retrofit, you'd be
better off with a scratch design using the very nice Analog Devices RF
power detector chips (huge dynamic range, temperature compensated) and a
microcontroller to do the calibration for directivity. A fancy one
might even attempt measure the frequency and do that part of the cal as
well.

Once you're not trying to go directly from detector device to a meter,
you can do a lot of useful stuff. If you still want a meter needle, have
the microcontroller put out an analog signal.

It would probably even be cheaper than trying to make a really flat
coupler with good directivity.