开云体育

I've had my H4 for a while now, and I really like it.

I've been trying to get an idea of its SWR accuracy, so I compared it to a friend's RigXpert.
I wrote an app that reads the RigXpert's AntennaScope files and NanoVNAServer's Touchstone files, and plots the SWR of each.
The results are not what I expected - pretty different (image attached, RigXpert-NanoVNA-Comparison.png). I didn't expect the results to match exactly, but I thought they would be close.

I've looked over a number of posts here, and haven't really found an answer. I hope these points answer some of the issues I've seen raised for similar questions:
- Both source files have 101 datapoints.
- The SWR plots I generate match the plots I see in the AntennaScope and NanoVNASaver software.
- I connect the cable from the antenna directly to the RigXpert, and for the Nano, I add a 12" SMA-SO239 adapter cable.
- I calibrate the Nano for 144-148 MHz, Open/Short/50 ohm.
- NanoVNAServer v0.3.4
- The H4 version info is in an attached image (NanoVNA-H4-Version.png).

It's certainly possible that the RigXpert is wrong, but that seems like a low probability.
I see there is a newer firmware version, but it seems to be lacking release notes. Will that solve the problem?

Thanks for any insight.

73,
Don Sayler W7OXR

Don,

Have you checked the velocity factor settings on the rig expert?

Was your pc isolated from earth, ie a laptop. I've heard that an earthed computer connection has quite an effect.

Just a couple of suggestions.

Wes G7CHO

On 1/26/21 8:29 PM, dsay2001 wrote:

I've had my H4 for a while now, and I really like it.

I've been trying to get an idea of its SWR accuracy, so I compared it to a friend's RigXpert.
I wrote an app that reads the RigXpert's AntennaScope files and NanoVNAServer's Touchstone files, and plots the SWR of each.
The results are not what I expected - pretty different (image attached, RigXpert-NanoVNA-Comparison.png). I didn't expect the results to match exactly, but I thought they would be close.

I've looked over a number of posts here, and haven't really found an answer. I hope these points answer some of the issues I've seen raised for similar questions:
- Both source files have 101 datapoints.
- The SWR plots I generate match the plots I see in the AntennaScope and NanoVNASaver software.
- I connect the cable from the antenna directly to the RigXpert, and for the Nano, I add a 12" SMA-SO239 adapter cable.
- I calibrate the Nano for 144-148 MHz, Open/Short/50 ohm.
- NanoVNAServer v0.3.4
- The H4 version info is in an attached image (NanoVNA-H4-Version.png).

It's certainly possible that the RigXpert is wrong, but that seems like a low probability.
I see there is a newer firmware version, but it seems to be lacking release notes. Will that solve the problem?

Thanks for any insight.

73,
Don Sayler W7OXR

My first thought would be that there is some "fixturing" issue - that is, if you're connecting the VNA/Scope at the antenna, the analyzer is part of the circuit.


You need to isolate the antenna being measured from the measurement system (and you) -? if the antenna is 100 feet away, and there's 100 ft of coax between you and the antenna, or if you make a metal covered box on which the antenna is mounted, and your analyzer is in the box.


What I would do to compare is to build a test load that is a RLC resonant at 144 MHz (or there abouts) with the right R to get the resonant impedance approximately right.? Probably one of the easiest ways is to cut 1/2 wavelength of coax, short the far end, and put a 50 ohm series resistor.? it will be 50 ohms at resonance (because the coax will look like a short) with SWR 1:1, and gradually worse on either side.

Run that on both analyzers and see how well they match.

Get a known standard, whatever that may be, a load standard, a short
section of characterised coax etc and compare the results.
What you are doing now is if I gave you an unknown weight and you measured
it on 2 bathroom scales.

On Wed, Jan 27, 2021 at 06:43 AM, dsay2001 wrote:

I've had my H4 for a while now, and I really like it.

I've been trying to get an idea of its SWR accuracy, so I compared it to a
friend's RigXpert.
I wrote an app that reads the RigXpert's AntennaScope files and
NanoVNAServer's Touchstone files, and plots the SWR of each.
The results are not what I expected - pretty different (image attached,
RigXpert-NanoVNA-Comparison.png). I didn't expect the results to match
exactly, but I thought they would be close.

I've looked over a number of posts here, and haven't really found an answer. I
hope these points answer some of the issues I've seen raised for similar
questions:
- Both source files have 101 datapoints.
- The SWR plots I generate match the plots I see in the AntennaScope and
NanoVNASaver software.
- I connect the cable from the antenna directly to the RigXpert, and for the
Nano, I add a 12" SMA-SO239 adapter cable.
- I calibrate the Nano for 144-148 MHz, Open/Short/50 ohm.
- NanoVNAServer v0.3.4
- The H4 version info is in an attached image (NanoVNA-H4-Version.png).

It's certainly possible that the RigXpert is wrong, but that seems like a low
probability.
I see there is a newer firmware version, but it seems to be lacking release
notes. Will that solve the problem?

I own a RigExpert and a NanoVNA-H4. The results are almost identical when I make measurements if I do things correctly.

I think your problem is that you have a common mode issue on your antenna system. If you do not have a proper balun, unun or RF choke at the antenna feedpoint the outer surface of the coax shield will form part of the antenna. Anytime you change the length of the coax or how it is grounded at the radio end you are effectively changing the antenna feedpoint impedance and will get slightly different SWR measurements. For VHF this is very common with antennas like a dipole, J-pole or a whip.

Here is a simple test to see if this is the case. Measure the antenna with the RigExpert in stand-alone mode and then use the Save feature to store the results on the RigExpert analyzer. Then connect it to your computer with a USB cable, measure again and save these results in another file on the Rigexpert. Then start AntScope 2 and connect to the RigExpert. You can import both files into AntScope2 (bottom right corner of AntScope2). These are RigExpert files not Touchstone. Both will plot on the screen and you can see if there is any difference. Now save both files on your PC. Exit AntScope.

Next step is to disconnect the RigExpert from the antenna and POC. Calibrate your NanoVNA for the desired frequency range and connect to the antenna with your pigtail. Connect it to the PC and start AntScope2. Go into settings and you will see an option to connect to your NanoVNA (latest AntScope2 supports the -H4). Connect and then exit settings and do a sweep. It will now plot on the screen. You can now import the 2 files you saved earlier and compare all 4 measurements. I suspect only two will closely match and the standalone RigExpert will be different.

Roger

I assume you are calibrating the nano with the pigtail installed? If not, you are comparing apples with oranges. Unless the load is a pure restive load with no inductive or capacitive components the added length of the pigtail will cause a change in the vswr as it acts like a transformer of the impedance and can very well result in the different curves you are seeing. If you don't have a way to calibrate the nano at the end of the pigtail you can add an equal electrical length of coax cable to the rig expert setup.

Instead of using an antenna which does capture ambient fields as your
comparison standard, find and use non-reactive resistors as you 'gold
standard'. You can also install series/parallel reactive components of
well known values around the resistive elements. That way, the potential
of interfering ambient fields are out of the picture.

Dave - W?LEV

On Thu, Jan 28, 2021 at 5:03 PM John Cunliffe W7ZQ <n2nep@...>
wrote:

I assume you are calibrating the nano with the pigtail installed? If not,
you are comparing apples with oranges. Unless the load is a pure restive
load with no inductive or capacitive components the added length of the
pigtail will cause a change in the vswr as it acts like a transformer of
the impedance and can very well result in the different curves you are
seeing. If you don't have a way to calibrate the nano at the end of the
pigtail you can add an equal electrical length of coax cable to the rig
expert setup.

--
*Dave - W?LEV*
*Just Let Darwin Work*

On Thu, Jan 28, 2021 at 09:03 AM, John Cunliffe W7ZQ wrote:

I assume you are calibrating the nano with the pigtail installed? If not, you
are comparing apples with oranges. Unless the load is a pure restive load with
no inductive or capacitive components the added length of the pigtail will
cause a change in the vswr as it acts like a transformer of the impedance and
can very well result in the different curves you are seeing. If you don't have
a way to calibrate the nano at the end of the pigtail you can add an equal
electrical length of coax cable to the rig expert setup.

Sorry but adding a short pigtail will not be an "apples to ranges" comparison. The complex impedance will change when you add a short pigtail to a long feedline but the VSWR change will be minimal. The reason is that the complex impedance change will result in a phase angle change to the reflection coefficient but the magnitude of the reflection coefficient will only be a tiny fraction smaller. You can easily verify this on a Smith chart or by adding a few feet of coax to a feedline and then measuring VSWR.

This is the kind of thing that scares me with the NanoVNA. There are many different "possible solutions/problems" posted above that could all likely be valid issues. So this means you REALLY need to know exactly what you are doing and have a REALLY good understanding of all this to actually successfully use the NanoVNA as a simple antenna analyzer. Somebody like myself who is not very proficient or knowledgeable abotu all this "stuff", will likely have a hard time getting good results that can be counted on. And worse, I have no idea if my results are any good or not! As in the example posted, is the Rigexpert correct, or is the nanovna correct? Or is neither correct....?

--
Regards,
Chris K2STP

As in so many other cases, ignorance is bliss, the only way out is to learn
more.

This is the kind of thing that scares me with the NanoVNA. There are many

On Sun, Jan 31, 2021 at 11:42 AM, Chris K2STP wrote:

This is the kind of thing that scares me with the NanoVNA. There are many
different "possible solutions/problems" posted above that could all likely be
valid issues. So this means you REALLY need to know exactly what you are
doing and have a REALLY good understanding of all this to actually
successfully use the NanoVNA as a simple antenna analyzer. Somebody like
myself who is not very proficient or knowledgeable abotu all this "stuff",
will likely have a hard time getting good results that can be counted on. And
worse, I have no idea if my results are any good or not! As in the example
posted, is the Rigexpert correct, or is the nanovna correct? Or is neither
correct....?

Any Antenna Analyzer has to be used with some knowledge of what you are measuring and if you are doing it correctly. It doesn't matter if you have a $25,000 VNA, a NanoVNA or a basic MFJ-259. One needs to know how to operate the device and have a basic understanding of antennas and transmission lines. A common problem that results in poor measurements and can lead the user to false conclusions is common mode current on the outside shield of the feedline.

A coaxial transmission line will have a differential current mode between the inner surface of the shield and the center conductor. It is also possible for current to flow on the outer surface of the shield as well and this is known as common mode. Some authors refer to the outer surface of the shield as the "third wire". Here is one example of where this can cause problems. If the unbalanced transmission line (coax) is directly connected to a dipole the center conductor will be attached to one arm and the shield to the other. Power is delivered to the antenna by differential mode (center conductor & inner shield surface). Since the outer surface of the shield is connected to one arm of the antenna it becomes part of the antenna and radiates just like the dipole arms. This has several unwanted effects. The first is that the antenna is no longer balanced because the current is not equal in both arms. The second is that the radiation pattern of the antenna will change. The third is that RF will be carried back to the hamshack. And finally the feedpoint impedance of the antenna will change. The last one will result in an SWR reading that is different from what is expected from a basic dipole above ground.

Not only will the SWR (and Return Loss) readings be different they will change depending on what is done at the ham shack end. If you hold the Analyzer tool in your hand you will get different readings than if you set it on a table due to capacitive coupling to your body. If you connect the analyzer to a PC you will have a ground connected and that will give a different reading. When the feedline is connected to a transceiver the termination will also be different so there will be a discrepancy from what you see on the analyzer and the SWR meter of the transmitter. If an additional length of coax is added you also get a different reading because the "third wire" is longer.

How much the readings change depends on a lot of factors including the type of antenna, length of feedline, grounding etc.

One can reduce the level of common mode current through the use baluns, RF chokes and proper grounding which is another topic.

Roger

As mentioned several time in previous posts, don't use the antenna/xline as
the standard. Use KNOWN standards which don't change with the environment
and are not susceptible to external fields - UNLIKE the antenna. Then you
will at least KNOW what you are dealing with.

Dave - W?LEV

--
*Dave - W?LEV*
*Just Let Darwin Work*