"Reg Edwards" wrote in message
...
Reg propped up this tar baby and everyone's taken a punch at it.

Perhaps it is time to check in and see if you have your answer yet
Reg.

==========================================
Wes,
Not everybody has yet taken a punch at it. There are several regular
names who are missing.

All I want is a number, eg., of decibels, preferably from a standards
lab.

But it has only been been demonstrated "Measurements" is not a
"Science" - it is an "Art". Perhaps I can clarify my question.

Suppose a customer, perhaps an antenna manufacturer, walks into the
lab wheeling behind him a weird contraption (we've heard of them) and
asks for the forward and reverse gains to be determined and for a
calibration certificate to be issued.

For present purposes actual forward and reverse figures don't matter.

But for the two figures to be of value the uncertainties in the
determination should be stated on the certificate (a legal document).

What are TYPICAL uncertainties, in dB, which appear above the Head of
the Laboratory's signature.



Reg:

I'm primarily doing the US MIL-STD-461E protocol, on widely varying test
specimens (hand-held, man-worn, 2-meter tall racks, 2-meter diameter
parabolic tracking antennas), so I am getting my antennas calibrated for 1 &
3 meter separation distances.

The lab I use has an outdoor range, and I get a Certificate of Conformance
for each antenna which includes the following statements:

"Test and Measurement Equipment used for performance verification is
calibrated with traceability to the U.S. National Institute of Standards and
Technology. Inspection records, test data, and other evidence of conformance
are on file at the sellers facility are are available for inspection on
request."

They further state a "Calibration Uncertainty of (2 sigma) (+/- 1 dB)" and
reference "SAE ARP-958". I then get a listing of the specific instruments
used in the generation of my antenna data, along with their calibration
certificate numbers and the date of each calibration period expiration.

As for the data, I get a tabulation and a plot. For a typical antenna (an
EMCO 3115 double-ridged horn covering 1 GHz to 18 GHz), I get a tabular
array of antenna correction factor, dB return loss, SWR, numeric gain and
dBi, in 250 MHz step increments, across the range. I also get a continuous
swept plot of return loss, and from this, I infer that the lab uses my
unknown antenna as the transmitting element on their range.

This data and the C of C is enough to keep my internal Metrology guy happy,
and it gives me a sufficiently warm feeling. I have been using the same
antenna lab for almost 10 years, and I informally track the calibration data
from year to year on each of my antennas. So far, the data never has looked
"copied", fudged, or otherwise egregious. However, that may just mean I'm
easily fooled.


--
Ed
WB6WSN
El Cajon, CA USA

Reg:

I'm primarily doing the US MIL-STD-461E protocol, on widely varying
test
specimens (hand-held, man-worn, 2-meter tall racks, 2-meter diameter
parabolic tracking antennas), so I am getting my antennas calibrated
for 1 &
3 meter separation distances.

The lab I use has an outdoor range, and I get a Certificate of
Conformance
for each antenna which includes the following statements:

"Test and Measurement Equipment used for performance verification is
calibrated with traceability to the U.S. National Institute of
Standards and
Technology. Inspection records, test data, and other evidence of
conformance
are on file at the sellers facility are are available for inspection
on
request."

They further state a "Calibration Uncertainty of (2 sigma) (+/- 1
dB)" and
reference "SAE ARP-958". I then get a listing of the specific
instruments
used in the generation of my antenna data, along with their
calibration
certificate numbers and the date of each calibration period
expiration.

As for the data, I get a tabulation and a plot. For a typical
antenna (an
EMCO 3115 double-ridged horn covering 1 GHz to 18 GHz), I get a
tabular
array of antenna correction factor, dB return loss, SWR, numeric
gain and
dBi, in 250 MHz step increments, across the range. I also get a
continuous
swept plot of return loss, and from this, I infer that the lab uses
my
unknown antenna as the transmitting element on their range.

This data and the C of C is enough to keep my internal Metrology guy
happy,
and it gives me a sufficiently warm feeling. I have been using the
same
antenna lab for almost 10 years, and I informally track the
calibration data
from year to year on each of my antennas. So far, the data never has
looked
"copied", fudged, or otherwise egregious. However, that may just
mean I'm
easily fooled.
--
Ed
WB6WSN
El Cajon, CA USA

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

Much obliged to you Ed for interesting infomation from a reliable
source.

Including 2-Sigma uncertainty limits of +/- 1 dB.

Thanks.
Reg, G4FGQ

Richard Fry posted a beautiful picture of a Harris / Gates test facility
in which an antenna tower is rotated and tilted up to 90-degrees, I
suppose. Nice way to get the antenna pattern. Hope Harris had a
government contract number to charge that job to.

Best regards, Richard Harrison, KB5WZI

From: "Richard Harrison"
Richard Fry posted a beautiful picture of a Harris / Gates test facility
in which an antenna tower is rotated and tilted up to 90-degrees, I
suppose. Nice way to get the antenna pattern. Hope Harris had a
government contract number to charge that job to.
________________

Azimuth patterns were taken by spinning the antenna+tower around a
horizontal axis centered above the two trestle supports you see in the
scanned photo. The AUT is positioned broadside to the source antenna.

Elevation patterns were taken by spinning the whole assembly in the
horizontal plane, on the horizontal centerline of the antenna+tower
assembly. The trestles sit on a huge wooden beam which itself is supported
by, and centered on a motor-driven turntable -- making that possible.

So both sets of patterns can be taken without needing to put the
antenna+tower in the vertical plane (no tilting to 90 degrees is necessary).

However we had several other positioners for vertical antennas to use when
the measurement of elevation patterns was not required.

Yes, this customer had deep pockets, but was not a government agency. Just
a major broadcast group.

RF