On Wed, 03 May 2006 20:46:06 -0700, Roy Lewallen
wrote:

Gary Schafer wrote:
. . .
The length of the 50 ohm line is 9 ¾" from shielded area to shielded
area (not counting the center pin exposed when shell is pulled back).
With the barrel installed it makes it 10 ¾" in length that is
shielded.
. . .

By my reckoning, a free space quarter wavelength at 145 MHz is 20.335
inches. Assuming solid polyethylene dielectric coax, this would be about
13.4 inches of coax. 10-3/4 inches would be about 72 degrees.

The probes have a shunt C of 2.5 pF, resulting in a Z (-X) of about 440
ohms at 145 MHz, not 100k ohms. If you parallel that with 50 ohms and
terminate a 72 degree 50 ohm cable with it, the voltage delay in the
cable is 78 degrees -- that is, the imperfect termination actually
increases the cable delay by 6 degrees. If the cable Z0 is 45 ohms
instead of 50, the delay is a couple of degrees less with that load; if
it's 55 ohms, the delay is a couple of degrees more. Maybe your 50 ohm
termination isn't quite what you think it is -- as you can see, it
doesn't take much to substantially change the cable's delay.

As I've said often, most people don't realize how difficult it is to
make accurate RF measurements.

Roy Lewallen, W7EL

Measuring RF is difficult! I can't seem to get any two methods to come
out in the same ball park.

If I just have the two 75 ohm lines hooked up with 50 ohm terminations
on the ends and measure with the probes I get 0 degrees phase.

I can hook one of the 75 ohm lines only with a 100 ohm load and one
probe at the generator end and a 50 ohm load at the other end with a
probe and see near 90 degrees.

I can do the same with the 50 ohm line with a 50 ohm load on each end
and see near 90 degrees.

But when I have the two 75 ohm lines and add the 50 ohm line in one
leg I get only around 60 degrees. By pulling the probe of the
connector (B channel) at the end of the 50 ohm line, and keeping the
center pin close but not in contact so there is only light capacitive
coupling the phase shift increases by around 10 degrees or so.

If I add a few elbow adaptors to the 50 ohm line to increase its
length to arund14 ½" inches or so then I see close to 90 degrees.

So it would seem that my 50 ohm line is too short but when measuring
it without the 75 ohm lines it looks to be the right length.

I built a couple of different 90 degree phasing setups several years
ago using two 75 ohm ¼ wave length lines, a spdt relay and a 90 ohm ¼
wave length line all hooked to the relay to be able to reverse
polarization.
I measured that setup in a similar manor only the center of the relay
went to the signal generator and the 50 ohm loads and probes went at
the ends of the 75 ohm lines. I get 60 degrees of phase shift on that
setup too.

I don't remember how I trimmed the 90 ohm line at the time but I do
remember I tried several different methods to try and measure what was
happening and had the same problems I am having now. I didn't have the
8405A then.
It seems I am doing no better with it. :)

I would like to find a repeatable method that will work when adding a
relay and or connectors to be able to know pretty close what the
actual delay is.

Thanks for the help.

Gary K4FMX