开云体育

On Mon, Jan 13, 2020 at 02:41 AM, Rune Broberg wrote:

0.0.0 means the firmware is *so old*, it doesn't have a "version" command ;-)

and,

On Mon, 13 Jan 2020 at 11:35, Joe Kirk wrote: Thanks Herb.I tried that approach but this is what I got and v0.0.0 made me wonder.
===============================================================

Joe,
Your situation is why Rune and other software developers asked the firmware guys to add a "version" command. Some software features may not be available for certain firmware or for certain models (i.e. the NanoVNA-F) and it allows the software developers to error check for those versions. Sounds like its time to upgrade your firmware.

- Herb

On Mon, Jan 13, 2020 at 04:16 AM, Leif M wrote:

What about the TDR. There is no delay there.
=====================================================

Don't forget you have enter a cable velocity factor or it will use a "canned value" (0.66%??) which may or may not be correct for the cable you are testing.

- Herb

On Mon, Jan 13, 2020 at 04:22 AM, Leif M wrote:

I just checked and my NanoVna_saver TDR shows distance in meters.
=================================================

Leif,
Check message #9483 in this thread. The math to convert from distance to time is probably trivial but not something I do frequently. You might put in a request to Rune to report the cable length in distance and time if pulling out a calculator is something you would rather avoid.

- Herb

I just checked and my NanoVna_saver TDR shows distance in meters.

What about the TDR. There is no delay there.

What you are looking for is a minimum loss situation.

All ferrites are the same but for Al value (always freq dependent) and hysteresis losses. Any low loss ferrite will do for your balun once the optimum wire thickness and winding number is achieved.

Low loss characterization would be easiest when doing a simple 1:1 transformer as all impedances remain (the same) 50 Ohm.

Than just characterize at your frequency of interest, the transfer value as a skalar (no need here for complex measurements)

Then make a 9:1 balun (arbitrary N windings) and use a 400 ohm serial resistor (to make sure it is loaded by 450 Ohm)? to the input of the nanoVNA (400 +50(input nano) = 450) to verify you were looking for.? Alternately you can reverse the measurement by inputting into the "9" side + 400? and loading the "1" side. (Losses shall essentially be the same, if not, set up problem)

Then connect the "1"side of the balun to your receiver (without 400 Ohm) and characterize the "9" side ?? for SWR and having a 450 Ohm real input.

(you might characterize the 400 ohm? resistor first (to be non-inductive) so as to make sure it does not spoil the measurement, and allowing you to measure the transfer losses created by the resistor)

Once you found best ferrite you may experiment with winding numbers (to cope for the specific Al value of your ferrite) and optimize the transfer to your receiver. Experiments may be bifilar winding, other geometries of the ferrite, (though you will find a correlation between ferrite volume and losses,so keep it small) and winding variations and interconnections. The last parts are the fun in this hobby.

If you do a good job, your balun may function from 300-30000 kHz. It is difficult to have more than 1:100 in frequency range for a non tuneable? balun, due to non-symmetries in windings and interconnects.

My 1 cents.....

Gert

On 13-1-2020 12:06, ptapon@... wrote:

Hello everyone,

A newbie set of questions.

On the net, you have as many opinion, way of measuring etc. as you have pages with the word ferrite in it.
Moreover, they are too few discussion to how to characterize a ferrite with a VNA. Bunch of stuff for common mode chokes and the like, but not a lot for baluns
Given we have now a wonderful set of tools, NanoVNA and nanoVNA Saver, I'd like to understand how to make the right choice.

I'm only interested in receiving antennas, therefore no power will go through, no heating concerns etc.
I'm, for now, interested in HF bands with, for now, a long wire.
I'm therefore aiming for a 9:1 balun.
I have a bunch of "no names" ferrites in my junk drawer, some from power supply, therefore likely good for KHz ranges and others from various sources.
As many of us, amateurs, I tend to use what I have handy, and not building, from a web page, the, say, Fair-rite type 43 with 9 turns of 26 AWG, without "fully" understanding what I am doing.

My main question:
What will be the characteristics to look for when comparing ferrites - and choosing the best one- for a 0-30Mhz 9:1 balun?

I have built an enclosure with dual sided PCB and SMB connectors (from another junk drawer :-) ) with a 50R resistor at the input SMB.
Calibration is done with the enclosure. I'm using the same wire I will likely use for the balun, between the input and either the ground or another SMB connected to S2.

I have already zillions of NanoVNA saver screen shots with various ferrites but I'm unable to find the right info to make a decision.
Will it be from the smith chart, SWR, Z, return loss, ??

Is there a way to find the right ferrite without winding a 9:1 balun each time.

When testing a 9:1 balun:
Do I need to load the "output" of the balun with a 450R resistor?
Do I need to connect the output (loaded or not) to the S2 ?
If yes, what will be the right info to look for?

Any help or pointer to the right docs will be helpful.
Thanks in advance

Jean

--

I think I've mentioned previously that I have no way to build the software
specifically for Windows XP - the only supported versions from Microsoft
are currently Windows 7, Windows 8.1 and Windows 10. And Windows 7 is
running out of extended support *tomorrow*. I will continue to build on 32
bit Windows 7 for as long as I can.

For running from source, the requirements are Python 3.7, PyQt5, pyserial,
numpy and scipy. At least PyQt5 might be difficult to get running on older
versions of Windows, and Python 3.7 is definitely not officially supported.
It's all open source, though, so anyone can have a go at making the
necessary changes.

Running the binaries at least requires a new-ish Microsoft VC++
Redistributable. Installing a very new version of this on Windows XP SP3
might help. But I have no way to test it, and I would trust that anyone who
actually gets it working would support their fellow users of historic
systems in their endeavours ;-)

--
Rune / 5Q5R

Hello everyone,

A newbie set of questions.

On the net, you have as many opinion, way of measuring etc. as you have pages with the word ferrite in it.
Moreover, they are too few discussion to how to characterize a ferrite with a VNA. Bunch of stuff for common mode chokes and the like, but not a lot for baluns
Given we have now a wonderful set of tools, NanoVNA and nanoVNA Saver, I'd like to understand how to make the right choice.

I'm only interested in receiving antennas, therefore no power will go through, no heating concerns etc.
I'm, for now, interested in HF bands with, for now, a long wire.
I'm therefore aiming for a 9:1 balun.
I have a bunch of "no names" ferrites in my junk drawer, some from power supply, therefore likely good for KHz ranges and others from various sources.
As many of us, amateurs, I tend to use what I have handy, and not building, from a web page, the, say, Fair-rite type 43 with 9 turns of 26 AWG, without "fully" understanding what I am doing.

My main question:
What will be the characteristics to look for when comparing ferrites - and choosing the best one- for a 0-30Mhz 9:1 balun?

I have built an enclosure with dual sided PCB and SMB connectors (from another junk drawer :-) ) with a 50R resistor at the input SMB.
Calibration is done with the enclosure. I'm using the same wire I will likely use for the balun, between the input and either the ground or another SMB connected to S2.

I have already zillions of NanoVNA saver screen shots with various ferrites but I'm unable to find the right info to make a decision.
Will it be from the smith chart, SWR, Z, return loss, ??

Is there a way to find the right ferrite without winding a 9:1 balun each time.

When testing a 9:1 balun:
Do I need to load the "output" of the balun with a 450R resistor?
Do I need to connect the output (loaded or not) to the S2 ?
If yes, what will be the right info to look for?

Any help or pointer to the right docs will be helpful.
Thanks in advance

Jean

0.0.0 means the firmware is *so old*, it doesn't have a "version"
command ;-)

--
Rune / 5Q5R

On Mon, 13 Jan 2020 at 11:35, Joe Kirk via Groups.Io <g3zdf=
[email protected]> wrote:

Thanks Herb.
I tried that approach but this is what I got and v0.0.0 made me wonder.
Joe

I opened the web page on two Android devices. They are versions 5.3 and 6.0,
which I know are old, but both worked with a USB memory stick so I know that
the OTG cable was good.

The problem is NOT nanoVNA firmware level.

If you can find an Android terminal program that can access USB serial devices,
it should be able to use nanoVNA shell commands.

Thanks Herb.
I tried that approach but this is what I got and v0.0.0 made me wonder.
Joe

For updating the software, it should be sufficient to just type "git pull"
in the nanovna-saver directory. This updates all the files - and the only
reason for the install command is to fetch the dependencies.

I don't run the software on Linux, and I choose to prioritize developing it
further over getting it running on multiple platforms - sorry. :-)

--
Rune / 5Q5R

Hi Aldo,
thanks for reporting the bug. If possible, please try starting the
application from a command prompt (cmd.exe), and copy or screenshot the
error message that turns up in that window after the crash. This will
pinpoint exactly where the error lies.

I will try to look at it as soon as I can. :-)

--
Rune / 5Q5R

Hi Chuck,

I created a pdf for the solution. Hope you can shape it. I wish you a successful installation.

73, Gyula

On 1/12/20 5:05 PM, Nels Nelsen wrote:

Hi Jim I don't see where my program starts.

???? I haven't yet created a shortcut on the desktop or in the Menu for nanovna-saver.? So, I open a terminal in my home directory, then enter these commands:

cd nanovna-saver

?????? followed by:

python3.7 nanovna-saver.py


???? And away it goes!???? That do it for you?

Pierre Martel wrote:

I used locate and find and I did not found any other place where I

This is almost surely a path and permissions problem. Mostly path. Where things wind up using this or that install routine, versus what the "distro" has mandated as the be-all and end-all of directory structures.

I wish the "high forehead" pocket-protector-flapping, egotistical, IDIOTS in charge of all the Linux distros would pull their collective high-forehead crania out of their asses and decide upon

ONE

directory structure. Right now, I can think of four discrete layouts, with /usr, /usr/local, /usr/etc/local (Believe it or not! But I think this one died a lonely death already - I hope.) and a couple of other variants involving /opt, where things can wind up.

Don't let any of their BS fool you, it's somebody's personal preference and NOT "security." Any directory on a Linux system is as secure as any other one, if the right options are selected, so /usr and /usr/local are not different in any significant way.

That crap, and the "sudo" insanity pushed by the ScrewBuntu crowd drive me absolutely ape-snot.

All that said, now it just means we need to track down where your system has decided to put the various bits and pieces, kick it in the gonads, and make a few symbolic links to where things -really- are instead.

Check the install logs, and see where the Qt5 stuff actually was sent when you tried to install it. Then we might be able to track down where your distro will accept them and force the correct parts down it's throat.

--
wes will
n9kdy
BSIndEng, MSIT(InfoSec)
(And some other alphabet soup to go with it.)
(All that and a couple dollars get me coffee at the local gas station. I can lose all of the diplomas, but I better have the two bucks.)

Also check /var/tmp

You also mentioned using locate. locate is useless until after the database is rebuilt, which happens in a system cron job or if you run it yourself.

Vince - K8ZW.

--
K8ZW

I've downloaded PyQt5-5.14.1.tar.gz from the link in the error I sent




there is no setup.py in this archive..



Le dim. 12 janv. 2020 à 22:03, Pierre Martel via Groups.Io <petem001=
[email protected]> a écrit :

if I try to install PQty5 by itself

python3.7 -m pip install PQty5
Collecting PQty5
Could not install packages due to an EnvironmentError: 404 Client
Error: Not Found for url:

pierre@workbench:~/nanovna-saver$

Le dim. 12 janv. 2020 à 21:56, Pierre Martel via Groups.Io
<petem001@...> a écrit :

I used locate and find and I did not found any other place where I
could have a tmp directory beside ton one on the root.

I did watched if the sudo python3.7 -m pip install . command did put
some stuff in the /tmp directory, end yes it does but it get erased as
soon as the install fail

Le dim. 12 janv. 2020 à 21:41, N9KDY <n9kdy@...> a écrit :

Pierre Martel wrote:

Yes I do. and there are some stuff left in it

ls /tmp/
mozilla_pierre0
pulse-PKdhtXMmr18n
ssh-IfE43VTJEnve

systemd-private-fb8c2b57f05146519286706d79b30aeb-colord.service-ITCxnV

systemd-private-fb8c2b57f05146519286706d79b30aeb-ModemManager.service-EKXWbl

systemd-private-fb8c2b57f05146519286706d79b30aeb-rtkit-daemon.service-mGzzVF

systemd-private-fb8c2b57f05146519286706d79b30aeb-systemd-timesyncd.service-GHA7MZ

systemd-private-fb8c2b57f05146519286706d79b30aeb-upower.service-69fZ57

Temp-1f3b678f-d226-4b63-964b-d3fca628a08f
Temp-809947be-febe-417e-8b53-4e89a2bb3775
tmpaddon

But notice there is NO pip-install directory.

Check under /usr/tmp and /usr/local/tmp and see if there might be a
pip-install directory in there somewhere.

--
wes will
n9kdy

if I try to install PQty5 by itself

python3.7 -m pip install PQty5
Collecting PQty5
Could not install packages due to an EnvironmentError: 404 Client
Error: Not Found for url:

pierre@workbench:~/nanovna-saver$

Le dim. 12 janv. 2020 à 21:56, Pierre Martel via Groups.Io
<petem001@...> a écrit :

I used locate and find and I did not found any other place where I
could have a tmp directory beside ton one on the root.

I did watched if the sudo python3.7 -m pip install . command did put
some stuff in the /tmp directory, end yes it does but it get erased as
soon as the install fail

Le dim. 12 janv. 2020 à 21:41, N9KDY <n9kdy@...> a écrit :

Pierre Martel wrote:

Yes I do. and there are some stuff left in it

ls /tmp/
mozilla_pierre0
pulse-PKdhtXMmr18n
ssh-IfE43VTJEnve
systemd-private-fb8c2b57f05146519286706d79b30aeb-colord.service-ITCxnV
systemd-private-fb8c2b57f05146519286706d79b30aeb-ModemManager.service-EKXWbl
systemd-private-fb8c2b57f05146519286706d79b30aeb-rtkit-daemon.service-mGzzVF
systemd-private-fb8c2b57f05146519286706d79b30aeb-systemd-timesyncd.service-GHA7MZ
systemd-private-fb8c2b57f05146519286706d79b30aeb-upower.service-69fZ57
Temp-1f3b678f-d226-4b63-964b-d3fca628a08f
Temp-809947be-febe-417e-8b53-4e89a2bb3775
tmpaddon

But notice there is NO pip-install directory.

Check under /usr/tmp and /usr/local/tmp and see if there might be a
pip-install directory in there somewhere.

--
wes will
n9kdy

I used locate and find and I did not found any other place where I
could have a tmp directory beside ton one on the root.

I did watched if the sudo python3.7 -m pip install . command did put
some stuff in the /tmp directory, end yes it does but it get erased as
soon as the install fail