On Jan 4, 10:55 am, Richard Clark wrote:
In recent correspondence we have been regaled with the possibilities,
the probabilities, the certainties that some fastidious, and
well-heeled Ham "could" achieve the determination of RF power to
within 0.1dB of its actual value.

At HF and VHF, you should be able to do power measurements to a tenth of
a dB, with moderate care.

1. That budget (with an ironic eye towards the language's claim of
"moderate") is a generous $1000 - with that budget being invested
solely towards the measurement (and not in the capital equipment of
transmitter and antenna) through the costs of ancillary equipment
acquisition and of guarantees of absolute accuracy as further
described below.

2. That frequency is 14.1MHz.

3. That power is 10mW.

4. That application is measuring the full power applied from a 100W
rated Ham grade transmitter/transceiver directly (or through a
directional coupler) to
a. an unloaded quarterwave thin antenna sited over 120 half-wave
radials placed slightly above, upon, or slightly below the ground
(slightly being defined as that recommended by an NEC modeling tool
for ground radial modeling); or
b. an unloaded halfwave thin antenna sited in free space.

5. That fully expressed tolerances for all components, capital and
ancillary equipment involved where the budget also absorbs their cost
of determination of their contribution to error.

6. That fully expressed tolerances for sources of system error.

7. That outcome of accuracy of power determination being applied to
the nominal 100W of power at the load and not the nominal 10mW of
power to the sensor.

* * * * * * * *

This will be a very short thread devoid of conclusion when the
poster(s) suddenly become aware of the expense exceeding the budget.
For those who challenge this artificial budgetary limitation, try
asking Congress for a bailout.

73's
Richard Clark, KB7QHC


No problem. Send money. This is, after all, something I do for a
living. I assume, of course, that your $1000 budget assumes labor is
free? (the usual situation for ham measurements)

As you are well aware, most of the cost of accurate (and perhaps
precise) measurements that are calibrated is not in the equipment, but
in the knowledge and effort in making the measurement and doing the
uncertainty analysis to prove that what you say is the measurement is
within some known distance of the measurand. After all, one could beat
your $1000 budget by saying that you should start with sand and iron
ore, and invest copious effort in smelting, refining, machining,
etc... Certainly, the cost of preparing the uncertainty analysis is
low, being paper and pencil, aside from the labor. The books sold by
Lindsay Books might be an example of how one can start with scrap
metal, and wind up with a machine shop, albeit at the investment of
substantial time.

If one has to pay for the skilled labor, either directly, or
indirectly by buying a piece of equipment designed, manufactured, or
calibrated by someone else, it's another thing entirely, eh?

I will concede that there is a catch.. if one is investing one's free
labor, it is conceivable that the time required to start from first
principles and raw materials would be long enough that your standards
or equipment would drift enough to prevent achieving the measurement
accuracy needed. It's the "you can't use a striped snake as a scale/
ruler, because it keeps moving and growing" problem.

However, from an academic perspective, your challenge is interesting.
It would be interesting to see a decent article in QST or QEX that
would describe readily achievable measurement accuracies for hams (and
do away with that horrible term you see in ads and equipment reviews:
"lab-grade"). Interesting, but not as interesting as other things I
might spend my free time on.

Jim, W6RMK