NanoVNA-Saver 0.1.0


Rune Broberg
 

I just released 0.1.0, grab it here:
https://github.com/mihtjel/nanovna-saver/releases/latest

This release primarily includes a number of improvements to the stability
of the software. Users should see fewer crashes after upgrading, and
several parts of the code have been improved to make them easier to
maintain and keep free of errors.

New features include the ability to save calibration standard sets, for
those users who have characterized sets, as well as the ability to save
images of charts with data directly by right clicking.

The save/load functionality for calibration files has been reworked, and
there is now a field for notes, which are saved and loaded along with the
file.

The phase charts now support display limits on the data and frequency span.

Touchstone files from RFSim99 should now be imported correctly, and I am
told there is also a method available to export compatible Touchstone files
from Elsie.

I look forward to hearing the feedback and suggestions from the community!

Particular thanks go to those who tested my alpha versions! I hope this
brief alpha test has helped the quality a little, even though I wasn't able
to address all the comments I received ;-)

--
Rune / 5Q5R


Bo, OZ2M
 

Thanks for the new release Rune. It is great to be able not only to sweep stuff but also compare vs theory/simulation.

Loading Elsie S2P Files

After some experimentation I think I am able to load Elsie S2P files into NanoVNA-Saver in the following way:

In Elsie select Menu | Tabulate | Write | Write S-parameter file | S2P [2-port] | MA [volts/volt] | MHz | Version 1.0 | Continue with SnP operation.

Now load the file in NanoVNA-Saver | Files | ... navigate to the file | Load reference / Load as sweep.

Bo


Nick
 

Hi Rune

thanks very much for the new release and fixing the buttons in the cal window.!

I like the object picker for calibration files, but cannot save or load calibrations.

I did a new cal, saved as bnc_4 to ~/nanovna-saver, but it has not saved the file there. Tried ~/config/NanoVNASaver but it's not there either!

73
Nick


Rune Broberg
 

That sounds odd. I definitely tested saving of calibration data. Did it
write anything to the console about the save failing? Did you check that it
didn't save as "bnc_4.cal"?

--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 15:13, Nick <g3vnc@uk6.net> wrote:

Hi Rune

thanks very much for the new release and fixing the buttons in the cal
window.!

I like the object picker for calibration files, but cannot save or load
calibrations.

I did a new cal, saved as bnc_4 to ~/nanovna-saver, but it has not saved
the file there. Tried ~/config/NanoVNASaver but it's not there either!

73
Nick




Norbert Kohns
 

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration right on the 0 dB line and on the current version somewhat below the 0 dB line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN


bill wright
 

Is there a trick to get the 0.1. version .exe to work.  I am on a win 10. machine. Is thisa linux .exe?  the other two version unload and install and work well w10 no.I tried to unload the zip file but do not find an .exe fileplease adviseBill kd5yyk

On Wednesday, October 2, 2019, 10:06:42 AM CDT, norbert.kohns@gmail.com <norbert.kohns@gmail.com> wrote:

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration right on the 0 dB line and on the current version somewhat below the 0 dB line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN


Brian Garber
 

Hi Rune,

Thanks for your development efforts! I have a quick question about saving calibration data; can you read from / save to memory locations on the NanoVNA?

73, Brian WB8AM


Rune Broberg
 

Hi Bill,
The .exe file should work just fine on Windows 10. First startup can take a
little while, maybe 10-20 seconds, but after that it should run fine.

What happens when you try to start it?

--
Rune / 5Q5R

On Wed, 2 Oct 2019, 17:19 bill wright via Groups.Io, <chief6226=
yahoo.com@groups.io> wrote:

Is there a trick to get the 0.1. version .exe to work. I am on a win 10.
machine. Is thisa linux .exe? the other two version unload and install and
work well w10 no.I tried to unload the zip file but do not find an .exe
fileplease adviseBill kd5yyk
On Wednesday, October 2, 2019, 10:06:42 AM CDT,
norbert.kohns@gmail.com <norbert.kohns@gmail.com> wrote:

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN






Rune Broberg
 

Hi Norbert,
thanks for the bug report! I can tell you that the data points are drawn
where they should be, but the reference lines are 10 pixels offset. Whoops!
I've fixed it on GitHub, and the fix will be in the next release as well.

Thanks again for your help!
--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 17:06, <norbert.kohns@gmail.com> wrote:

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN




Bryan, WA5VAH
 

Rune,

Thanks for the new release. Keep up the great work.

I see the same issue with S11 return loss in dB.

I do not see it in the phase plots.

--
Bryan, WA5VAH

On Wed, Oct 2, 2019 at 09:06 AM, <norbert.kohns@gmail.com> wrote:

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN


Rune Broberg
 

Hello Brian,
thanks for trying out the software! :-)

I do not read from or save to the NanoVNA calibration locations. Currently,
my recommendation is that users calibrate the NanoVNA across the full span
of 50kHz-900 or 1500MHz, depending on version, with whatever attached
hardware (cables, adaptors) they are going to use, and save this as
calibration memory 0. This will ensure that the data transmitted to
NanoVNA-Saver is at least reasonable looking, and won't fall foul of the
bits of code that try to detect transmission errors.

Calibration within the application itself can be saved to calibration
files. Currently, the calibration set used (for those users using custom
characterized sets) is not saved in the calibration file, but has to be set
manually in the interface.

This method allows the use of many more data points for calibrating within
the application. I've used as many as 5050 (= 50 "segments") at one point.
This higher resolution cannot be achieved using the NanoVNA's built in
memories.

I hope this explains why it's currently not possible?

Thanks again for taking the time to try out my software :-)

--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 17:22, brianbgarber via Groups.Io <brianbgarber=
yahoo.com@groups.io> wrote:

Hi Rune,

Thanks for your development efforts! I have a quick question about saving
calibration data; can you read from / save to memory locations on the
NanoVNA?

73, Brian WB8AM




Rune Broberg
 

Hi Bryan,
thanks for the feedback! The code used is the same for S11 and S21 charts
of the same type, such as the return loss/gain charts - so the errors end
up being the same ;-) Thanks for checking in any case!

Early aim for next release is early next week. It depends how many features
I try to push into it. ;-)
--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 18:15, bryburns via Groups.Io <bryburns=
aol.com@groups.io> wrote:

Rune,

Thanks for the new release. Keep up the great work.

I see the same issue with S11 return loss in dB.

I do not see it in the phase plots.

--
Bryan, WA5VAH

On Wed, Oct 2, 2019 at 09:06 AM, <norbert.kohns@gmail.com> wrote:

Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after
calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN



Nick
 

Here's the console output when when I hit the load cal button...

nick@LM19-2:~$ python3.7 ./nanovna-saver/nanovna-saver.py
NanoVNASaver 0.1.0
Copyright (C) 2019 Rune B. Broberg
This program comes with ABSOLUTELY NO WARRANTY
This program is licensed under the GNU General Public License version 3

See https://github.com/mihtjel/nanovna-saver for further details
Settings: /home/nick/.config/NanoVNASaver/NanoVNASaver.ini
Gtk-Message: 17:32:47.985: GtkDialog mapped without a transient parent. This is discouraged.

I get the same Gtk-Message when I save a cal with or without the .cal extension.

The cal files become visible if I give them a .cal extension. Files without the .cal extension are not visible in the object picker even when All Files is selected e.g. bnc_1 and bnc_2 are not visible to the object picker because they do not have the .cal extension.

It is possible to save a cal file without the .cal extension in which case it is hidden from the object picker.

Perhaps the save dialog should force a .cal extension e.g. by adding it automatically to the Name field?

73
Nick


Rune Broberg
 

Hi Nick,
this looks ... interesting.

The Gtk error isn't something I have any control over - it seems to be your
Window Manager (at a guess?) or local Qt implementation that's using Gtk
for its popups. It looks like a bug, but possibly not one I can fix...

The dialog window for saving gets the following instruction in the code:
filedialog.setDefaultSuffix("cal")
This is supposed to set the default suffix to be added if no other suffix
is provided, and it does on Windows at least. ;-)

So sadly it looks a bit like Linux-specific Qt errors. (Or, of course,
that I'm using it incorrectly, and it randomly happens to work on Windows.
Both are likely! ;-)

Thanks for getting back to me on this though, it's a very interesting
problem!
--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 18:49, Nick <g3vnc@uk6.net> wrote:

Here's the console output when when I hit the load cal button...

nick@LM19-2:~$ python3.7 ./nanovna-saver/nanovna-saver.py
NanoVNASaver 0.1.0
Copyright (C) 2019 Rune B. Broberg
This program comes with ABSOLUTELY NO WARRANTY
This program is licensed under the GNU General Public License version 3

See https://github.com/mihtjel/nanovna-saver for further details
Settings: /home/nick/.config/NanoVNASaver/NanoVNASaver.ini
Gtk-Message: 17:32:47.985: GtkDialog mapped without a transient parent.
This is discouraged.

I get the same Gtk-Message when I save a cal with or without the .cal
extension.

The cal files become visible if I give them a .cal extension. Files
without the .cal extension are not visible in the object picker even when
All Files is selected e.g. bnc_1 and bnc_2 are not visible to the object
picker because they do not have the .cal extension.

It is possible to save a cal file without the .cal extension in which case
it is hidden from the object picker.

Perhaps the save dialog should force a .cal extension e.g. by adding it
automatically to the Name field?

73
Nick




Larry Rothman
 

Hi Rune,There is a small bug in the serial connect routine.If I disconnect the Nano while scanning, I get the pop-up error and that's OK.If I then click on the Rescan button, the selected serial port disappears but the button underneath still shows "Disconnect".I think if there is no serial port shown, that should be reset as there is no recovery that I can see - 'Saver just hangs.


Regards,Larry


Rune Broberg
 

Hi Larry,
you're right - the application still thinks at that point that the NanoVNA
is connected - despite having just received an error that it was
disconnected. It should probably just disconnect when it gets the error.

I'll put it on the list of little niggles ;-)

Thanks for the feedback!
--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 19:10, Larry Rothman <nlroth@rogers.com> wrote:

Hi Rune,There is a small bug in the serial connect routine.If I disconnect
the Nano while scanning, I get the pop-up error and that's OK.If I then
click on the Rescan button, the selected serial port disappears but the
button underneath still shows "Disconnect".I think if there is no serial
port shown, that should be reset as there is no recovery that I can see -
'Saver just hangs.


Regards,Larry




Bryan, WA5VAH
 

Rune,

Running firmware NanoVNA-H__900_ch_20190924.dfu from hugen with nanovna-saver.v0.1.0

This firmware has TDR mode built-in to the device. If the TDR Transform is enabled in the device, data sent to the PC is not what you expect. It appears to be the result of the TDR transform in some form. However, frequencies are those most recently commanded to the device. I am not sure how to turn off the transform from the PC.

--
Bryan, WA5VAH


Rune Broberg
 

Hi Bryan,
I'm aware of the issue. I don't know if it can be turned off via the serial
console - thus far, my advice to users has been to not put the device in
TDR mode when using it with NanoVNA-Saver, and to use the TDR move in the
application instead. If you do need to use TDR mode on the NanoVNA,
disconnect it from NanoVNA-Saver first.

If there is a way to distinguish and maybe even reset what mode it's in,
I'd love to know! :-)
--
Rune / 5Q5R

On Wed, 2 Oct 2019 at 19:23, bryburns via Groups.Io <bryburns=
aol.com@groups.io> wrote:

Rune,

Running firmware NanoVNA-H__900_ch_20190924.dfu from hugen with
nanovna-saver.v0.1.0

This firmware has TDR mode built-in to the device. If the TDR Transform
is enabled in the device, data sent to the PC is not what you expect. It
appears to be the result of the TDR transform in some form. However,
frequencies are those most recently commanded to the device. I am not sure
how to turn off the transform from the PC.

--
Bryan, WA5VAH




bill wright
 

RuneSorry for the alert  Win10 was trying to protect me from the download.I deleted  and then reloaded. This time it gave me a functional app.  Now all is well.I will test on thanks you so much for your efforts cheersBill kd5yyk

On Wednesday, October 2, 2019, 10:48:41 AM CDT, Rune Broberg <mihtjel@gmail.com> wrote:

Hi Bill,
The .exe file should work just fine on Windows 10. First startup can take a
little while, maybe 10-20 seconds, but after that it should run fine.

What happens when you try to start it?

--
Rune / 5Q5R

On Wed, 2 Oct 2019, 17:19 bill wright via Groups.Io, <chief6226=
yahoo.com@groups.io> wrote:

  Is there a trick to get the 0.1. version .exe to work.  I am on a win 10.
machine. Is thisa linux .exe?  the other two version unload and install and
work well w10 no.I tried to unload the zip file but do not find an .exe
fileplease adviseBill kd5yyk
    On Wednesday, October 2, 2019, 10:06:42 AM CDT,
norbert.kohns@gmail.com <norbert.kohns@gmail.com> wrote:

  Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN









| | Virus-free. www.avast.com |


Bob Albert
 

Rune, forgive me if you aren't the guy to whom I should ask this.
I purchased a nanoVNA recently and it isn't working as I think it should.  Regardless of terminations I always get a dot at the right edge of the Smith chart.  I always get a flat line for SWR.  However, the frequencies are correct and I can select many menu options.
I don't know if this is a hardware problem (I don't think so).  Is there a way to reset the processor or load some alternate software?  I hate to throw away this lovely little unit, but I may have to get another.
I have calibrated it several times but the problem persists.  At one point I did get a reasonable display but it disappeared and haven't been able to get it back.
For what it's worth, the accuracy of the frequency is amazing.  It's within less than 1 ppm.  For that alone, this unit was worth the price but that's not something I need.  I already have a rubidium standard.
Bob

On Wednesday, October 2, 2019, 10:48:11 AM PDT, bill wright via Groups.Io <chief6226=yahoo.com@groups.io> wrote:

RuneSorry for the alert  Win10 was trying to protect me from the download.I deleted  and then reloaded. This time it gave me a functional app.  Now all is well.I will test on thanks you so much for your efforts cheersBill kd5yyk
    On Wednesday, October 2, 2019, 10:48:41 AM CDT, Rune Broberg <mihtjel@gmail.com> wrote:

Hi Bill,
The .exe file should work just fine on Windows 10. First startup can take a
little while, maybe 10-20 seconds, but after that it should run fine.

What happens when you try to start it?

--
Rune / 5Q5R

On Wed, 2 Oct 2019, 17:19 bill wright via Groups.Io, <chief6226=
yahoo.com@groups.io> wrote:

  Is there a trick to get the 0.1. version .exe to work.  I am on a win 10.
machine. Is thisa linux .exe?  the other two version unload and install and
work well w10 no.I tried to unload the zip file but do not find an .exe
fileplease adviseBill kd5yyk
    On Wednesday, October 2, 2019, 10:06:42 AM CDT,
norbert.kohns@gmail.com <norbert.kohns@gmail.com> wrote:

  Hi Rune,
thanks for the software update and for the additional features!
One issue I recognized: S21 is with the previous version after calibration
right on the 0 dB line and on the current version somewhat below the 0 dB
line.
Am I doing something wrong with the new version?

Kind regards
Norbert, DG1KPN









|  | Virus-free. www.avast.com  |