开云体育

Hello all XMAS-gift hopefuls,

Is there a way/method/technique for subtracting coax length from readings, as can be done in Rig-Expert's ANTScope, sftwr?

73
Phil, WF3W

On Thu, Dec 8, 2022 at 06:49 AM, Phil, WF3W wrote:

Is there a way/method/technique for subtracting coax length from readings, as
can be done in Rig-Expert's ANTScope, sftwr?

The AntScope2 software by RigExpert allows the user to add or subtract cable and then make a series of measurements. There is a database of cable types (RG-58, LMRxxx etc.) to choose from. You select the cable type and length and it calculates the approximate impedance at the antenna when measuring at the far cable end (subtraction method). If the measurements were made at the antenna and you want to know what happens if you add cable (addition method) it will calculate the approximate impedance at the cable end. The reason I say approximate is the attenuation and velocity of propagation are not exactly known due to manufacturing tolerances and vary with frequency.

The equations to do the impedance transformation are well known. Attached is an excerpt from the ARRL Radio Amateur's Handbook describing how to do it. The equation shown will work for the "addition method". To do the "subtraction method" you need to rearrange the terms so that ZLoad is the desired calculation. The length of the cable, velocity of propagation, and attenuation for the frequency being used are inputs (some unit conversion required).

Roger

Phil,
From what I've seen on the various write ups and Youtubes, you are able to determine the length, loss and any breaks in a coax by shorting the far end and using the 2nd VNA port to determine the coax properties....... from that you should be able to accomplish what you're after.
I've only had one for a few weeks and trying to get my head in gear !
JohnKK4ITX


Click Here for Zephyrhills Area Amateur Radio ClubMany of life's failures are people whodid not realize how close they were tosuccess when they gave up.?????? Thomas A. Edison

The normal way to do it on the nanovna is a bit different. You attach the
near end of the coax to the nanovna, then calibrate (using the
open/short/load calibration loads) at the far end of the cable. Then you
attach the antenna at the far end - all measurements then show the
performance of the antenna itself, having compensated for the coax
properties (primarily length and attenuation). This is more accurate than
the RigExpert's calculation, since as Roger noted, RigExpert uses a data
base of cable types - but any given actual cable will only be 'close' to
the data base values, whereas using the nanovna calibration measures the
actual values.

You can alternately use the "edelay" setting on the nanovna to subtract the
length of the cable, if you look up its properties and enter the correct
value for your length of coax. This also is only approximate, and is
similar to what the RigExpert is doing. You can also use the nanovna to
measure the cable's electrical length, and use that value to set the
edelay, but those are all manual steps, so calibration at the end of the
cable is usually just as easy.

Stan

Yet another method is to measure at the shack end of the feedline. Then,
using SimSmith on your PC, propagate what you measure back to the
feedpoint. You must know the parameters of your feedline: length, Vp, and
type. For HF using anything but RG-174, the loss is pretty
insignificant.

Dave - W?LEV

On Thu, Dec 8, 2022 at 8:02 PM Roger Need via groups.io <sailtamarack=
[email protected]> wrote:

On Thu, Dec 8, 2022 at 06:49 AM, Phil, WF3W wrote:

Is there a way/method/technique for subtracting coax length from

readings, as

can be done in Rig-Expert's ANTScope, sftwr?

The AntScope2 software by RigExpert allows the user to add or subtract
cable and then make a series of measurements. There is a database of cable
types (RG-58, LMRxxx etc.) to choose from. You select the cable type and
length and it calculates the approximate impedance at the antenna when
measuring at the far cable end (subtraction method). If the measurements
were made at the antenna and you want to know what happens if you add cable
(addition method) it will calculate the approximate impedance at the cable
end. The reason I say approximate is the attenuation and velocity of
propagation are not exactly known due to manufacturing tolerances and vary
with frequency.

The equations to do the impedance transformation are well known. Attached
is an excerpt from the ARRL Radio Amateur's Handbook describing how to do
it. The equation shown will work for the "addition method". To do the
"subtraction method" you need to rearrange the terms so that ZLoad is the
desired calculation. The length of the cable, velocity of propagation,
and attenuation for the frequency being used are inputs (some unit
conversion required).

Roger

--
*Dave - W?LEV*


--
Dave - W?LEV

On Thu, Dec 8, 2022 at 06:49 AM, Phil, WF3W wrote:

Hello all XMAS-gift hopefuls,

Is there a way/method/technique for subtracting coax length from readings, as
can be done in Rig-Expert's ANTScope, sftwr?

Phil,

If you like AntScope2 you can use it with your NanoVNA-H4. RigExpert added support for the -H and -H4 some time ago. Go to Settings and manually connect to the NanoVNA Com port. Some features like TDR and Calibrate are not available. You have to calibrate on the NanoVNA for the range of interest and then input the start/stop frequencies. The Cable Subtract/Add feature does work but I am not a fan of it for the reasons I posted earlier. Make sure you use metric settings as that is what is used cable length entry. Screenshot attached.

Roger

On Thu, Dec 8, 2022 at 12:52 PM, Stan Dye wrote:

The normal way to do it on the nanovna is a bit different. You attach the
near end of the coax to the nanovna, then calibrate (using the
open/short/load calibration loads) at the far end of the cable. Then you
attach the antenna at the far end - all measurements then show the
performance of the antenna itself, having compensated for the coax
properties (primarily length and attenuation).

I agree with you that this "de-embedding" the cable method is the most accurate and only needs to be done once for the cable being used. Then the cal file can be saved for later recall.

You can alternately use the "edelay" setting on the nanovna to subtract the
length of the cable, if you look up its properties and enter the correct
value for your length of coax. This also is only approximate, and is
similar to what the RigExpert is doing.

The "edelay" method is the least accurate of all the methods because it only compensates for the propagation delay and not the attenuation. It just adjusts the angle of the "reflection coefficient" and not the magnitude which also needs adjustment.

Roger

On 12/8/22 2:16 PM, Roger Need via groups.io wrote:

On Thu, Dec 8, 2022 at 12:52 PM, Stan Dye wrote:

The normal way to do it on the nanovna is a bit different. You attach the
near end of the coax to the nanovna, then calibrate (using the
open/short/load calibration loads) at the far end of the cable. Then you
attach the antenna at the far end - all measurements then show the
performance of the antenna itself, having compensated for the coax
properties (primarily length and attenuation).

I agree with you that this "de-embedding" the cable method is the most accurate and only needs to be done once for the cable being used. Then the cal file can be saved for later recall.

This is what I do, but I don't store it in the NanoVNA. I store it in a PC program (NanoVNA-Saver) as well as a .s2p file and then use scikit-rf to do the processing.

You can also look at ZPlots, by Dan Maguire, AC6LA, which I personally find more user-friendly, and perform the same analysis.

RigExpert’s AntScope 1, when used in consistent fashion, will provide support fir the earlier RigExpert units.

Hard to believe, but there are still a bunch of internet self-appointed experts who would rather try to cut a half-wave of coax to bring the feedpoint impedance to the shack (or the analyzer) than believe the physics. I have seen their efforts to sadly include bogus Vf use, using coax specs, rather than testing it.

Oh well.

Ed McCann
AG6CX