开云体育

Thanks for the response guys.
Please forgive me, as I am a little weak using linux.
Can you elaborate more on how you got it to work?
As mentioned it used to work until ubuntu upgraded itself (with my permission yikes) to 20.04.
When I click on the exe file I get no response.
Thanks again all.

The dfu tool and STMCube are interchangeable - Cube just has a lot more features (and runs on a lot more platforms). I have had zero issues using them interchangeable on my -H4 . . .

--
Sent from my Android device with K-9 Mail. Please excuse my brevity.

This gave me an idea.

Does the reverse apply? If the manufacturer used STMCube, could I have successfully loaded an old NanoVNA dfu with DfuSE-Demo which ran with anomalies but then when I tried to load a different version the computer would not recognize the device being plugged in when booted with jogwheel pressed?

This is great !!!?Thank you my friend, for the files with full dump of GS-400 which is just what I needed.
You're right, I destroyed the bootloader so I cannot use either NanoVNA-App or QTbut I already verified that I can communicate to the MCU via STLinkso I'll go ahead and flash it keeping my fingers crossed....I'll return with good news of success.

Jim

I calibrate a coax pig tail that connects to my antenna system (50ft + UnUn
+ Wire + Counterpoise). That way, I eliminate a known good out of my
measurement by “calibrating out” the known good. I have also worked my way
out of diagnosing the antenna system by measuring the window connection,
coax then with unun + antenna + wire. Not sure if this is best practice or
where I learned it. Just seems to make logical sense. How are others doing
it? From posts I have read, it seems people just wing it without any
logical step thru of the antenna system!

On Mon, Jan 30, 2023 at 9:14 AM Jim Lux <jimlux@...> wrote:

On 1/30/23 3:42 AM, PDXer wrote:

Does it need new calibration every time when different frequency range

is selected?

Or is it enough just one initial full freq. calibration?

That depends on the level of accuracy you want. If you do a full range
cal, and then set a narrower range for the measurement, it interpolates
from the full range cal. That may or may not be great (particularly at
the points where it changes which harmonics to use, at N*300 MHz points)

If you cal over, say, 0-900, and there's 100 points, that means there's
a calibrated point every 9 MHz. Leaving aside the "transition points
between harmonics" it's likely that the device's internal errors (what
you're calibrating out) smoothly vary in amplitude and phase between
points.

OTOH, if you are calibrating with a fixture or coax in the system (e.g.
you've got 50 ft of coax, and you're doing the cal at the end of the
coax, to measure your antenna at the feedpoint) then you potentially
have a problem -That 50 ft of coax has multiple wavelengths (and a
different number of wavelengths, depending on frequency) - and 9 MHz
between calibration points may have one or more phase wraps. Now,
linear interpolation might not work so well.

The other thing where you definitely want to recalibrate for the "range
of use" is where you want the best possible measurements. Why
interpolate if you can use real data?

For what it's worth, I cal 0-100 MHz (because it's easy) when working
with HF stuff.

And, if you're using an application (like NanoVNA-Saver, and others)
that does multiple sweeps to get more points in the range, then you can
calibrate with fine resolution over a wide range.






--

73, ko4aqf@...

Depending on your firmware version, you can save multiple sets of calibration data.

I saved all five sets in my nano, specifically, 3 Mhz to 15 Mhz, 15 Mhz to 30 Mhz, 3 Mhz to 30 Mhz, 143 Mhz to 149 Mhz, and 420 Mhz to 450 Mhz,? I then load the calibration data as needed for the antenna span being checked.

David
AK9F

On 1/30/23 3:42 AM, PDXer wrote:

Does it need new calibration every time when different frequency range is selected?
Or is it enough just one initial full freq. calibration?

That depends on the level of accuracy you want. If you do a full range cal, and then set a narrower range for the measurement, it interpolates from the full range cal. That may or may not be great (particularly at the points where it changes which harmonics to use, at N*300 MHz points)

If you cal over, say, 0-900, and there's 100 points, that means there's a calibrated point every 9 MHz. Leaving aside the "transition points between harmonics" it's likely that the device's internal errors (what you're calibrating out) smoothly vary in amplitude and phase between points.

OTOH, if you are calibrating with a fixture or coax in the system (e.g. you've got 50 ft of coax, and you're doing the cal at the end of the coax, to measure your antenna at the feedpoint) then you potentially have a problem -That 50 ft of coax has multiple wavelengths (and a different number of wavelengths, depending on frequency) - and 9 MHz between calibration points may have one or more phase wraps. Now, linear interpolation might not work so well.

The other thing where you definitely want to recalibrate for the "range of use" is where you want the best possible measurements. Why interpolate if you can use real data?

For what it's worth, I cal 0-100 MHz (because it's easy) when working with HF stuff.

And, if you're using an application (like NanoVNA-Saver, and others) that does multiple sweeps to get more points in the range, then you can calibrate with fine resolution over a wide range.

Does it need new calibration every time when different frequency range is selected?
Or is it enough just one initial full freq. calibration?

Read better....

You use STM-Cube, so you destroy bootloader (need use NanoVNA-App or QT for update in bootload mode)
Upload this in 0x08000000 address as before
4-inch-GD32F303CCT6-20200619 FULL.bin

One of user made full dump for GS-400

I cut remove bootloader and data, try upload it (hold right button and power on, connect to PC, and use NanoVNA-App or QT for update)

If not help you can try flash all data from dump (but need ST-Link), made full dump for your device

This info was a great help to me, thank you!

On 1/29/23 10:45 AM, Ho-Ro wrote:

On Sun, Jan 29, 2023 at 01:19 PM, Fkawee wrote:

I am trying to reload nanovna-saver following the instructions from this link

The program simply works from the source directory without having to install it. The installation allows you to run the program from anywhere, just have the main program in the path and the modules in the Python module path.
When I started hacking nanovna-saver on my Debian stable two years ago, I did just that:
git clone
cd nanovna-saver
Then I looked at the response of "python nanovna-saver.py" (e.g. missing modules) and installed the missing modules - preferably via apt - e.g.
apt install python3-numpy python3-scipy
until nanovna-save.py was satisfied - DONE!
Martin

I did pretty much the same thing..

On Sun, Jan 29, 2023 at 01:19 PM, Fkawee wrote:

I am trying to reload nanovna-saver following the instructions from this link

The program simply works from the source directory without having to install it. The installation allows you to run the program from anywhere, just have the main program in the path and the modules in the Python module path.

When I started hacking nanovna-saver on my Debian stable two years ago, I did just that:

git clone
cd nanovna-saver

Then I looked at the response of "python nanovna-saver.py" (e.g. missing modules) and installed the missing modules - preferably via apt - e.g.
apt install python3-numpy python3-scipy
until nanovna-save.py was satisfied - DONE!

Martin

dislord maybe can help .... (best firmware modder ever hi hi)

dg9bfc sigi

Am 29.01.2023 um 17:51 schrieb Ray WW0AW via groups.io:

Now I realize I should've saved the original firmware before updating but it's too late for that now,

Yes it is a GD32 - I can read it right on the chip.
However .....
The STM32CubeProgrammer Target information tab says :
STM32F101/F103 High Density
Sevice ID 0x414
Flash size 256KB
CPU Cortex-M3
Bootloader version --
Also there is a 24 Mhz xtal, that appears to feed both the MCU CLOCK and the Si5351 if that is significant
Thank you for assisting to resurect this baby - which WAS working until I made the fatal mistake of trying to update the firmware,
Surely there's someone who can point me to the right firmware to reflash

On 1/29/23 4:19 AM, Fkawee wrote:

I once had the nanovna-saver executable working in ver 18 something. I let ubuntu upgrade to ver 20.04. since then the Nanaovna-saver exe does not work. I am trying to reload nanovna-saver following the instructions from this link

When I run the first command i get this error.
"fkawee@fkawee-Satellite-C650:~$ sudo apt install python3.8 python3-pip [sudo] password for fkawee: Reading package lists... Done Building dependency tree... Done Reading state information... Done Package python3.8 is not available, but is referred to by another package. This may mean that the package is missing, has been obsoleted, or is only available from another source
E: Package 'python3.8' has no installation candidate"
I tried following the instruction to install a "python virtual environment" with the same results.
I am searching for a fix but can't seem to find any. Any tips would help.
Thanks all

Did you apt update (and/or) apt upgrade - the former updates the local list of packages and where to get them; the latter actually brings in the new packages to your system. You might not want the latter, and would rather upgrade selectively.

apt uses a local directory of packages and maybe it doesn't have python3.8 in it.

So I'd do a "sudo apt update" then do the python install.

I once had the nanovna-saver executable working in ver 18 something. I let ubuntu upgrade to ver 20.04. since then the Nanaovna-saver exe does not work. I am trying to reload nanovna-saver following the instructions from this link



When I run the first command i get this error.

"fkawee@fkawee-Satellite-C650:~$ sudo apt install python3.8 python3-pip [sudo] password for fkawee: Reading package lists... Done Building dependency tree... Done Reading state information... Done Package python3.8 is not available, but is referred to by another package. This may mean that the package is missing, has been obsoleted, or is only available from another source

E: Package 'python3.8' has no installation candidate"

I tried following the instruction to install a "python virtual environment" with the same results.
I am searching for a fix but can't seem to find any. Any tips would help.
Thanks all

On Sat, Jan 28, 2023 at 05:52 PM, alan victor wrote:

Along with 30 years of RF DESIGN MAG

Awesome. I have several times looked for RF Design Magazine on the web but never found it. World Radio History has a LOT of good stuff. An excellent resource.

The link...