[R-390] URM-25D recap.

[Jim Brannigan]

FWIW
I've checked my URM-25 output against my HP 410-C and Tektronix scope.....
various frequencies, etc.  The numbers all sorta' agree....close enuf for
government work...
I agree with Roy.  I have a shop, not a lab.  I'm more interested in
relative readings than absolutes. Repeatability and comparison from a
shielded controlled source, at a known resistance is important.
For example:
If a "2" on the URM-25 voltmeter produces the same output on two different
receivers, they have the same gain.  If a "2" at the RF stage produces the
same output as a "1" at an IF stage, I have gain.  The amount of gain can be
worked out with a step attenuator.
Whether the sensitivity is .8, .008 or .0008 is immaterial at HF because of
atmospheric noise.

Jim


> At 02:15 PM 6/24/02 -0400, JM/CO wrote:
> >Yes, an "RF" voltmeter is required here.
>
> URM-25 Calibrators,
>
> I agree, but there is more.
>
> >   An excellent meter for
> >this purpose is the Hewlett Packard model 410-C,
>
> Indeed it is, but what if such a thing is not available?  All is not
> lost.  Lets make some assumptions:
>
> 1) High accuracy is not required.
>
> 2) The voltmeter circuit in the URM-25 measures accurately enough for our
> purposed across the frequency range of the generator.
>
> 3) If an AC meter is available that can measure well enough at a few
> hundred KC or even much lower, then we can use it to set the URM-25
> calibration.
>
> 4) The manual for any meter that is available will give some idea of the
> accuracy to expect at a few hundred KC or lower.
>
> If we can accept these ideas, then the procedure would be as follows:
>
> 1) Get whatever AC meter is available, and figure out what accuracy you
can
> expect from it.
>
> 2) Calibrate the URM-25 with that meter.
>
> 3) Call the project done.
>
> A couple of alternate methods come to mind:
>
> Method A:  Find some recommended AC voltage detector circuits useful at
the
> frequencies of interest.  Build one and assume the measurements are close
> enough.  This method may well involve very few components (one diode, one
> resistor, one capacitor and a way to mount them), and a DC voltmeter of
> moderate accuracy.   If a circuit reported to have good performance is
> duplicated carefully, the results for these purposes should be quite good
> enough.
>
> Method B: Build such an AC voltage detector and mail it so someone who
does
> have the likes of an HP 410C for calibration by comparison at some
> frequency(ies) of interest. Upon its return, use it to do the calibration.
>
> Notes:
>
> 1) I seem to remember that the output calibration of the URM-25 depends on
> setting the signal generator section output, at the input to the
> attenuator, to 2 volts rms.  (I believe that the GR 1001A and the HP 606
> are similar.)  Making an ac voltage detector that works well at 2 volts
and
> over a limited range of low to moderate frequencies would be a rather
> simple job.  The URM-25 output voltage specifications is based on at last
> the following points:
>   A) Calibration of the 2 volt point is done as specified,
>   B) The  RF attenuator is working properly.
>   C) The generator output is terminated properly.
>
> 2) Making very accurate RF voltage measurements over a wide range of
> frequencies is quite involved.  Careful study of the specifications of the
> HP RF voltmeters and similar general use laboratory quality instruments
> will give an indication of the accuracy attainable in production
equipment.
> A look at the following URL, which explains the RF voltage calibration
> services available at NIST, reveals that at 100 MC and above, peak to peak
> detectors can be calibrated to 1.2 percent.  No one in the country can do
> better. (Think about that next time someone reports the sensitivity of a
> radio receiver to two or three significant digits.)
>
<http://ts.nist.gov/ts/htdocs/230/233/calibrations/Electromagnetic/Voltage.h
tm#534>
>
> 3) An ac voltmeter that works well up to frequencies beyond the range of
> the URM-25 is simply not necessary.
>
> Roy