开云体育

Hi Dave (W0LEV),

Your measurements of your improved test fixture are impressive. I seem to
recall your original fixture had a single PCB shield between the alligator
clips? Now I'm curious about your thought process as you added bits and
pieces to that original fixture and found improved isolation between the
ports. While I understand adding more PCB walls, the addition of wire
bridges, and particularly the wires paralleling the alligator clips,
intrigues me. My sense is that you have created a transmission line on
each port comprising the alligator clip and its paralleled 'shield' wire,
and the shield wire helps terminate fields and weaken those fringing fields
that couple to the other port. Is this an accurate understanding of your
progression? Do you recall the order in which pieces were added and the
incremental improvement in port isolation you obtained at each step along
the way?

Thanks, and 73,
Mike, K8CN
------------------------------
*From:* [email protected] <[email protected]> on
behalf of W0LEV via groups.io <davearea51a@...>
*Sent:* Tuesday, May 20, 2025 3:38 PM
*To:* NANO VNA <[email protected]>; Ham-Antennas <
[email protected]>; Ham Antennas <[email protected]>
*Cc:* Kenneth Wyatt <emc.guru@...>; AA0RS@... <
AA0RS@...>; David Eckhardt <davearea51a@...>; David Feldman <
wb0gaz@...>
*Subject:* [Ham-Antennas] FIXTURE for XMSN MEASUREMENATS 2.0


CAUTION: This email originated from outside of the University System. Do
not click links or open attachments unless you recognize the sender and
know the content is safe.

As promised, I made some rigorous measurements of my home brew fixture for
making transmission measurements of various HF circuit elements using a
"professional" instrument. I could have used the NANOVNAs, but using HP
might add some credence to the results ????? Subjects for the fixture
might include common mode chokes, ugly baluns (which I do not use), current
baluns (which I use wound of RG-400 on toroids), filters, and other active
and passive circuits where a fixture is required.

I took the time to calibrate my HP8753C using HP standards with a full
2-port cal from 1 to 30 MHz to make these measurements of the fixture.
Please have a read of the attachment for the results. While the fixture is
essentially free for a few inexpensive components and my time (retired, so
that's free, too). While it's not up to the old HP standards, not bad for
a home baked project which will lead to other projects.

Oh, yes,..... any typos are also free.

If you attempt duplication based on the images, YMMV. Dimensions can be
scoped from the standard alligator clips which are 1.3-inches tip-to-tip (±
amateur measurement relying on a non-RF ruler).

Measure everything. Let the data speak for itself, *not *opinion, snake
oil, or sorcery.

Dave - W?LEV

Thanks, Dave - the incremental improvements are less critical than the
final result, of course. I need to make up a similar test fixture and
assess its residual coupling with nothing on the alligator clips.

You're fortunate to have the HP 8753C to use at home. Had one at our
university lab, but I use a DG8SAQ VNWA at home, and the university got a
better Agilent VNA with wider frequency coverage beyond 6 GHz if I should
do more microwave work in the future.

73,
Mike, K8CN
------------------------------
*From:* David Eckhardt <davearea51a@...>
*Sent:* Tuesday, May 20, 2025 7:11 PM
*To:* Michael Carter <Mike.Carter@...>
*Cc:* NANO VNA <[email protected]>; Ham-Antennas <
[email protected]>; Ham Antennas <[email protected]>;
Kenneth Wyatt <emc.guru@...>; AA0RS@... <
AA0RS@...>; David Feldman <wb0gaz@...>
*Subject:* Re: [Ham-Antennas] FIXTURE for XMSN MEASUREMENATS 2.0


CAUTION: This email originated from outside of the University System. Do
not click links or open attachments unless you recognize the sender and
know the content is safe.

I can't give you numbers of improvement as I didn't record those. But I
can give you the order of additions.

1) Lengthened the shield between the ports. It was originally only the
width of the main board. supporting the BNCs.

2) Playing with a thin sheet of double sided board, I added the
additional orthogonal shields. I was going to enclose each port except for
the lond ends, but in "playing" with the essentially scalar network
analyzer, I discovered I gained just as much by shielding opposite sides of
each port.

3) That looked pretty good, but again playing with just a piece of AWG #18
stripped solid copper wire, I discovered the added wires on each side of
the alligator clips really bought me a lot. And, yes, I believe your
analysis that the wires form a transmission line which electrically looks
much like a trough line but allows access to the clips.

4) From there, I could make minor improvements in isolation, but all
those severely limited physical access to the alligator clips. The
additional improvements traded off against physical access was clear I'd
stop where I was. The 70 dB (somewhat +) isolation between ports was quite
satisfactory for me. I could dig deeper into the noise by increasing the
output of the source port on the 8753C, but, again, any additional physical
components would limit access to the clips. So I stopped with what you
see.

Now I gotta wonder how high in frequency it maintains the isolation and
good RF performance. Maybe tomorrow? I have too many projects lined up in
retirement. It's the satisfaction and joy of having good RF test equipment
in retirement!

Dave - W?LEV

On Tue, May 20, 2025 at 8:06?PM Michael Carter <Mike.Carter@...>
wrote:

Hi Dave (W0LEV),

Your measurements of your improved test fixture are impressive. I seem to
recall your original fixture had a single PCB shield between the alligator
clips? Now I'm curious about your thought process as you added bits and
pieces to that original fixture and found improved isolation between the
ports. While I understand adding more PCB walls, the addition of wire
bridges, and particularly the wires paralleling the alligator clips,
intrigues me. My sense is that you have created a transmission line on
each port comprising the alligator clip and its paralleled 'shield' wire,
and the shield wire helps terminate fields and weaken those fringing fields
that couple to the other port. Is this an accurate understanding of your
progression? Do you recall the order in which pieces were added and the
incremental improvement in port isolation you obtained at each step along
the way?

Thanks, and 73,
Mike, K8CN
------------------------------
*From:* [email protected] <[email protected]> on
behalf of W0LEV via groups.io <davearea51a@...>
*Sent:* Tuesday, May 20, 2025 3:38 PM
*To:* NANO VNA <[email protected]>; Ham-Antennas <
[email protected]>; Ham Antennas <[email protected]>
*Cc:* Kenneth Wyatt <emc.guru@...>; AA0RS@... <
AA0RS@...>; David Eckhardt <davearea51a@...>; David Feldman <
wb0gaz@...>
*Subject:* [Ham-Antennas] FIXTURE for XMSN MEASUREMENATS 2.0


CAUTION: This email originated from outside of the University System. Do
not click links or open attachments unless you recognize the sender and
know the content is safe.

As promised, I made some rigorous measurements of my home brew fixture for
making transmission measurements of various HF circuit elements using a
"professional" instrument. I could have used the NANOVNAs, but using HP
might add some credence to the results ????? Subjects for the fixture
might include common mode chokes, ugly baluns (which I do not use), current
baluns (which I use wound of RG-400 on toroids), filters, and other active
and passive circuits where a fixture is required.

I took the time to calibrate my HP8753C using HP standards with a full
2-port cal from 1 to 30 MHz to make these measurements of the fixture.
Please have a read of the attachment for the results. While the fixture is
essentially free for a few inexpensive components and my time (retired, so
that's free, too). While it's not up to the old HP standards, not bad for
a home baked project which will lead to other projects.

Oh, yes,..... any typos are also free.

If you attempt duplication based on the images, YMMV. Dimensions can be
scoped from the standard alligator clips which are 1.3-inches tip-to-tip (±
amateur measurement relying on a non-RF ruler).

Measure everything. Let the data speak for itself, *not *opinion, snake
oil, or sorcery.

Dave - W?LEV




--

*Dave - W?LEV *