开云体育

The voltage of the reflected wave at a short circuit must cancel the voltage of the incident wave so that zero potential exists across the short circuit (equal to and 180? out of phase with). In other words, the voltage reflection coefficient must be -1 : |1| at an angle of 180?. Angles are measured from the positive real axis, which is horizontal. The pure short circuit location is at the left. (Only applies to voltage, not current.) So the measured response of the reflected wave at the plane of the VNA must be 180? out of phase with transmission.

The reflected voltage of an open circuit is equal to and /in phase/ with the incident voltage (reflection coefficient of +1) so that the open circuit location is on the right.

The ideal is not realized in real life. If you use the NanoVNA calibrated, you can watch the Smith chart marker move from 50 + 0i to something moving left and up to something moving right and down, as you change from a 50 ohm load > short > open.

~R~

--
On the banks of the Piscataqua
Rich NE1EE

The ideal is not realized in real life. If you use the NanoVNA calibrated, you
can watch the Smith chart marker move from 50 + 0i to something moving left
and up to something moving right and down, as you change from a 50 ohm load >
short > open.

With proper nanoVNA calibration and sweeps < 900MHz,
open and short should stay on Smith real axis.

If, after calibrating, you attach another length of coax,
then you can see at higher frequencies
phase shifts != 180 degrees by so-called fringe reactance effects,
e.g for 2 meters LMR-400 with 550-600MHz sweep open vs short:

You are correct. I seemed to be having difficulty calibrating with NanoVNA-Q-0.4.2-bf9c4ba, so I flashed with NanoVNA-H_v0.2.3-2-g8ac9166_20191018.

I got what seemed to be a good cal. I then saw (range 130-230MHz) open, 50 O, and short stayed on Smith real axis, exactly where I thought they should be (R, C, L).

--
On the banks of the Piscataqua
Rich NE1EE

On Tue, Oct 29, 2019 at 08:05 PM, Rich NE1EE wrote:

You are correct. I seemed to be having difficulty calibrating with
NanoVNA-Q-0.4.2-bf9c4ba

That's very interesting. Calibration with NanoVNA-Q should be a little better than with NanoVNA-H firmware.

Can you show me please screenshot of NanoVNA-Q with CH0 LOGMAG and CH1 LOGMAG after CAL RESET (with no calibration) for 50 kHz - 1500 MHz range?

The same screenshot (CH0 LOGMAG and CH1 LOGMAG after CAL RESET) for other devices are also welcome, it will allows to get statistics about RF frontend on different devices

On the Smith Chart, the short is at 9 o'clock and the open is at 3
o'clock. This is the way it is supposed to be. 180 degrees.

Mel, K6KBE

CH0 50 OHM
CH1 50 OHM

larry36, at a glance looks good, but it seems that your hardware has some issues above 1200 MHz.

Here is mine:

10k - 1500M

The same screenshot (CH0 LOGMAG and CH1 LOGMAG after CAL RESET)
for other devices are also welcome,
it will allows to get statistics about RF frontend on different devices

nanoVNA "worse" clone. This is the first time it has 50 Ohms directly to its SMA since July.

version
0.4.2-bf9c4ba-release

trace
0 LOGMAG CH0 12.000000000 8.000000000
3 LOGMAG CH1 12.000000000 8.000000000

Look at the purple line in this chart
/g/nanovna-users/attachment/5975/1/fit_end.PNG
It's the deviation from 180 degrees versus frequency after optimizing the fringe capacities

On Tue, Oct 29, 2019 at 09:38 PM, Oristo wrote:

nanoVNA "worse" clone. This is the first time it has 50 Ohms directly to its
SMA since July.

Did you performed CAL RESET before screenshot? It looks not good. With such hardware you may have worse dynamic range.

It looks that original hugen79 hardware has really better performance.

On Tue, 29 Oct 2019 at 18:22, Mel Farrer, K6KBE <farrermesa@...>
wrote:

On the Smith Chart, the short is at 9 o'clock and the open is at 3
o'clock. This is the way it is supposed to be. 180 degrees.

Mel, K6KBE

180 degrees phase difference between open and short gives you the best
stability in the calibration. However, the inductance of short is
negligible, but the fringe capacitance of the open is not. So you will *not*
get 180 degrees if you short and open a bit of coax. It will be close, but
certainly not 180 degrees.

Despite all these idiots on YouTube and elsewhere telling you that the
short needs to be as short as possible, the delay of the short should be a
little longer than the open, in order to get as close as possible to the
180-degree phase difference.

Picking one of the cal kits I have, the Agilent 85052B 3.5 mm, which is
nearest to SMA, the offset delays are

* Shorts = 31.785 ps
* Opens = 29.243 ps


--
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892

Tel 01621-680100 / +44 1621-680100

Did you performed CAL RESET before screenshot?

Yes, and also disabled CAL-CORRECTION

It looks not good. With such hardware you may have worse dynamic range.

Yes, Hugen warned of such, but note SCALE is 12dB/div with 0db at 8
I have been going to run some noise abatement experiments.

It looks that original hugen79 hardware has really better performance.

"...they [clones] have failed to understand the role of bridge and shielding,
and made some bad modifications to facilitate manufacturing."
/g/nanovna-users/message/158