I was wondering how commercial equipment go about calibrating the S
meter and whats the best way of building a calibrated S meter that would
be reliable as a professional field strength meter?

Since most S meters just read AGC voltage and it would be easy to build
some interface with a CPU to read voltages and calibrate this way. The
question is this, is a S meter calibrated this way actually reading in
a accurate way whats occurring at the antennas terminal. Since most
antenna inputs are not 50 ohms J0, how would you design a system like a
spectrum analyzer that measures signal voltage at the antenna terminal?.
I would be interested in some ideas, i am homebrewing a shortwave
receiver however i wanted a calibrated S meter in Dbuv, Dbm, S units and
Millivolts. I also want to use a calibrated antenna for a Antenna
factor input to have a meaningful long term view of propagation signal
strength.

Thanks
Will

Replace the meter scale with a white card and calibrate it yourself
with pen and ink. You will also need a signal generator and an
attenuator.

Don't forget, an S-meter is a power or wattmeter. It indicates signal
STRENGTH. S9 = 50 pico-watts.
----
Reg.

wrote
Only by coincidence and only in a 50 ohm system elsewhere
S9 is 50 uV (microVolts) equals 50pW.

==========================================
Allison, I'm afraid you are not quite correct about coincidences.

The beauty of calibrating an S-meter in watts is that the result is
independent of receiver impedance.

Of course, there must be a conjugate match between antenna and
receiver. And there usually is. This is taken care of in the
calibration process.

The received signal STRENGTH is indicated in watts, which is all one
wants to know. When reading the meter, who cares about what impedance
the measuring instrument happens to be?

To sum up : There is a transmitter of given power output. There is a
radio path which is an attenuator, And there is a received signal
signal strength meter which indicates watts.

The overall loss between transmitter and power meter may be deduced in
terms of decibels. Professional radio engineers do it all the time.

It can be misleading to think in terms of S9 = 50 uV when one doesn't
know what the receiver input impedance is.
----
Reg, G4FGQ

Reg Edwards wrote:
Replace the meter scale with a white card and calibrate it yourself
with pen and ink. You will also need a signal generator and an
attenuator.

Don't forget, an S-meter is a power or wattmeter. It indicates signal
STRENGTH. S9 = 50 pico-watts.
----
Reg.


Yeah thats for that tip. But the question is how do you actually build a
piece of electronics that measures signal levels at the receiver
terminals. Or is it reading agc voltage just as accurate. The problem
the method you described is that it will vary from band to band.

I just wonder how field strength meters are designed, especially the new
solid state ones. Spectrum analyzers too have a pretty flat response. I
want to build this ability into my homebrew receiver much the same way
a selective level meter works. Maybe someone has done it.

Will

On Tue, 23 May 2006 12:16:27 +0100, "Reg Edwards"
wrote:

Replace the meter scale with a white card and calibrate it yourself
with pen and ink. You will also need a signal generator and an
attenuator.

If you derive the S-meter reading from the a.g.c. control voltage, be
prepared to perform the calibration at various temperatures, since the
gain of simple IF-strips vary with temperature (and thus need a
different amount of a.g.c.), unless the gain is controlled by passive
components only.

With various front end filter for various frequency bands and
attenuation depending of the frequency, you would have to perform the
calibration on several frequencies.

Instead of a pen and ink, it would be more practical to use a micro
controller with sufficient non-volatile storage for the conversion
tables and use analog inputs for a.g.c. voltage and temperature and
some kind of input for the frequency being received. With these inputs
and the conversion table, a voltage proportional of the log of the
antenna terminal power could be generated.

Paul OH3LWR

"Will" wrote
I just wonder how field strength meters are designed, especially the
new
solid state ones. Spectrum analyzers too have a pretty flat
response. I
want to build this ability into my homebrew receiver much the same
way
a selective level meter works. Maybe someone has done it.

===========================================

The accuracy of S-meters varies from one band to another because the
receiver input impedance varies from one band to another. Even from
one end of a band to the other, especially if the input stage is
tuned.

Also, the stage-gain of an RF stage can vary very much from one band
to another.

To ensure a consistent gain it is necessary to convert to the
intermediate frequency (IF) at the very first stage of a receiver.
Preferably using a high-level, balanced diode-bridge, modulator.

The IF stages can be assumed to have a constant, wide-band gain,
except, of course, for the automatic gain control (AGC) action.

Then ensure that the receiver input impedance is constant (usually 50
ohms) over the whole HF range of the receiver. Remember the S-meter
is a power or watt-meter.

It so happens that with dual-gate FET's, and 3 or 4 IF amplifier
stages, a meter which responds to AGC volts will fairly accurately
indicate S-units and dB above S9 in a linear fashion.

However, below about S4 the linearity of S-meter readings begins to
fail. And you will have to make and calibrate your own meter scale.
There's no way of avoiding it! That is unless you can mentally
visualise what the scale ought to look like as you use it.

Accurate commercial field strength meters use antennas dedicated to
the job, conjugate-matched to the receiver, are insensitive to weak
signals, are narrow band and are very expensive.

To obtain an S-meter scale to be proud of, you will have to use pen
and ink with a signal generator and 100-dB stepped attenuator.
Attempts to calibrate the scale with clever, highly complicated
electronics will get you nowhere in a long time. And will cost you
more than the remainder of the receiver.

Sorry to be so despondent.
----
Reg, G4FGQ

"Paul Keinanen" wrote in message
...
On Tue, 23 May 2006 12:16:27 +0100, "Reg Edwards"
wrote:

Replace the meter scale with a white card and calibrate it yourself
with pen and ink. You will also need a signal generator and an
attenuator.

If you derive the S-meter reading from the a.g.c. control voltage,
be
prepared to perform the calibration at various temperatures, since
the
gain of simple IF-strips vary with temperature (and thus need a
different amount of a.g.c.), unless the gain is controlled by
passive
components only.

With various front end filter for various frequency bands and
attenuation depending of the frequency, you would have to perform
the
calibration on several frequencies.

Instead of a pen and ink, it would be more practical to use a micro
controller with sufficient non-volatile storage for the conversion
tables and use analog inputs for a.g.c. voltage and temperature and
some kind of input for the frequency being received. With these
inputs
and the conversion table, a voltage proportional of the log of the
antenna terminal power could be generated.

Paul OH3LWR
===========================================

I don't doubt that the problem is capable of being solved.

Just throw enough electronic hardware and software at it.

The cost can be estimated at 100,000 Euros.

The size and weight of the receiver will be quadrupled.

No amateur would be able to afford buying one.

American tax-payers would prevent NASA from doing the research and
development work. Even Bush would object.

As for me and most other amateurs, I'm quite happy with an S-meter
reading which puts signal strength reports in the right ball-park.
Very strong, very weak, or indifferent.
----
Reg.

Reg Edwards wrote:
"Will" wrote
I just wonder how field strength meters are designed, especially the
new
solid state ones. Spectrum analyzers too have a pretty flat
response. I
want to build this ability into my homebrew receiver much the same
way
a selective level meter works. Maybe someone has done it.

===========================================

The accuracy of S-meters varies from one band to another because the
receiver input impedance varies from one band to another. Even from
one end of a band to the other, especially if the input stage is
tuned.

Also, the stage-gain of an RF stage can vary very much from one band
to another.

To ensure a consistent gain it is necessary to convert to the
intermediate frequency (IF) at the very first stage of a receiver.
Preferably using a high-level, balanced diode-bridge, modulator.

The IF stages can be assumed to have a constant, wide-band gain,
except, of course, for the automatic gain control (AGC) action.

Then ensure that the receiver input impedance is constant (usually 50
ohms) over the whole HF range of the receiver. Remember the S-meter
is a power or watt-meter.

It so happens that with dual-gate FET's, and 3 or 4 IF amplifier
stages, a meter which responds to AGC volts will fairly accurately
indicate S-units and dB above S9 in a linear fashion.

However, below about S4 the linearity of S-meter readings begins to
fail. And you will have to make and calibrate your own meter scale.
There's no way of avoiding it! That is unless you can mentally
visualise what the scale ought to look like as you use it.

Accurate commercial field strength meters use antennas dedicated to
the job, conjugate-matched to the receiver, are insensitive to weak
signals, are narrow band and are very expensive.

To obtain an S-meter scale to be proud of, you will have to use pen
and ink with a signal generator and 100-dB stepped attenuator.
Attempts to calibrate the scale with clever, highly complicated
electronics will get you nowhere in a long time. And will cost you
more than the remainder of the receiver.

Sorry to be so despondent.
----
Reg, G4FGQ


Thanks Reg for the summary.

I think the Flex radio and the Winradio have the ability to be
calibrated accurately via a look up table. I always admired shortwave
receivers like the R&S EK 2000 which had a accurate meter calibrated
in the DbUv scale. I think i will just have to abandon my super
accurate S meter project.

Will

Will take a look at URL:
http://www.ac6v.com/sunit.htm


Maybe the National Semiconductor NE604 IF amplifier IC would be worth
looking into. Claims say it provides an accurate signal strength logarithmic
output that closely tracks the input signal level over a wide dynamic range
that could possibly be used for driving an S meter circuit.

--
CL -- I doubt, therefore I might be !


"Will" wrote in message
...
I was wondering how commercial equipment go about calibrating the S meter
and whats the best way of building a calibrated S meter that would be
reliable as a professional field strength meter?

Since most S meters just read AGC voltage and it would be easy to build
some interface with a CPU to read voltages and calibrate this way. The
question is this, is a S meter calibrated this way actually reading in a
accurate way whats occurring at the antennas terminal. Since most antenna
inputs are not 50 ohms J0, how would you design a system like a spectrum
analyzer that measures signal voltage at the antenna terminal?. I would be
interested in some ideas, i am homebrewing a shortwave receiver however i
wanted a calibrated S meter in Dbuv, Dbm, S units and Millivolts. I also
want to use a calibrated antenna for a Antenna factor input to have a
meaningful long term view of propagation signal strength.

Thanks
Will

In article ,
wrote:

Two caveats, rarely are RX input impedences 50 ohms and there is no
guarentee that the antenna is 50 ohms.

Allison-

Such is life. However you can assume that both are 50 Ohms and use a
resistive pad between the signal generator and the receiver when
calibrating. When making a reading, use a pad between the unknown signal
source and the receiver.

Without a sophisticated system capable of determining impedances while
making measurements, this may be the best you can do.

Fred