¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. Nanovna-Users
  2. Messages
Expanded Topics Messages
Date
Re: That pesky 50 to 75 ohm conversion.

 

On Tue, Jul 13, 2021 at 02:49 AM, Victor Reijs wrote:

Can you just change in the code this 50 into 75ohm?
It will not be perfect since as you pointed out the hardware (bridge, ...) is 50ohm but at least if you calibrate with a 75ohm reference all values displayed will be shown as referenced to 75ohm
at the center of the smith chart. While looking through the code I also noticed that one has to change the saving of touchstone files as well since that was hard-coded to 50ohm as well.

I have attached my quick and dirty patch (untested!!!) for a reference impedance change from 50ohm to 75ohm. One sets the reference impedance in NanoVNASaver/RFTools.py, line Z_0 = ...
Then when you restart NanoVNAsaver all values displayed should be relative to whatever Z_0 value you have set there.

50_to_75ohms.patch
50_to_75ohms.patch

Re: That pesky 50 to 75 ohm conversion.

 

Just to be sure. Can you just change in the code this 50 into 75ohm?
Everything (S11 and S21) in the NanaoVNA is measured using a 50 ohm
(internal in the device), Perhaps one could recalculate to 75ohm, but one
has to assume that all measured items are liniar towards the load. Is that
indeed true?

All the best,

Victor


Op di 13 jul. 2021 om 10:59 schreef Uwe Lange <uwe_lange@...>:

addendum to my previous reply ...

For those who are using NanoVNAsaver, I had a quick look into the source
code for place(s) of hard-coded Z_0 = 50 ohms.
I believe (but have not tested ...) that if you change all occurences of
50 to 75 in the list below that all values and calculations in
NanoVNAsaver would be done with a reference of Z_0 = 75 ohm:

NanoVNASaver/Calibration.py: g = (Zsp / 50 - 1) / (Zsp / 50 +
1) * cmath.exp(
NanoVNASaver/Calibration.py: g = ((Zop / 50 - 1) / (Zop /
50 + 1)) * cmath.exp(
NanoVNASaver/Calibration.py: g = (Zl / 50 - 1) / (Zl / 50 + 1)
* cmath.exp(

NanoVNASaver/RFTools.py: def impedance(self, ref_impedance: float = 50)
-> complex:
NanoVNASaver/RFTools.py: def shuntImpedance(self, ref_impedance: float
= 50) -> complex:
NanoVNASaver/RFTools.py: def seriesImpedance(self, ref_impedance: float
= 50) -> complex:
NanoVNASaver/RFTools.py: def qFactor(self, ref_impedance: float = 50)
-> float:
NanoVNASaver/RFTools.py: def capacitiveEquivalent(self, ref_impedance:
float = 50) -> float:
NanoVNASaver/RFTools.py: def inductiveEquivalent(self, ref_impedance:
float = 50) -> float:
NanoVNASaver/RFTools.py:def gamma_to_impedance(gamma: complex,
ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py:def impedance_to_norm(z: complex, ref_impedance:
float = 50) -> complex:
NanoVNASaver/RFTools.py:def norm_to_impedance(z: complex, ref_impedance:
float = 50) -> complex:
NanoVNASaver/RFTools.py:def reflection_coefficient(z: complex,
ref_impedance: float = 50) -> complex:

NanoVNASaver/Windows/CalibrationSettings.py:
self.load_resistance.setText(str(self.app.settings.value("LoadR", 50)))

NanoVNASaver/Windows/TDR.py: self.step_response_Z = 50 * (1 +
self.step_response) / (1 - self.step_response)
Re: That pesky 50 to 75 ohm conversion.

 

On Tue, Jul 13, 2021 at 01:59 AM, Uwe Lange wrote:


I had a quick look into the source code
code obtained with:
git clone

Re: That pesky 50 to 75 ohm conversion.

 

addendum to my previous reply ...

For those who are using NanoVNAsaver, I had a quick look into the source code for place(s) of hard-coded Z_0 = 50 ohms.
I believe (but have not tested ...) that if you change all occurences of 50 to 75 in the list below that all values and calculations in
NanoVNAsaver would be done with a reference of Z_0 = 75 ohm:

NanoVNASaver/Calibration.py: g = (Zsp / 50 - 1) / (Zsp / 50 + 1) * cmath.exp(
NanoVNASaver/Calibration.py: g = ((Zop / 50 - 1) / (Zop / 50 + 1)) * cmath.exp(
NanoVNASaver/Calibration.py: g = (Zl / 50 - 1) / (Zl / 50 + 1) * cmath.exp(

NanoVNASaver/RFTools.py: def impedance(self, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py: def shuntImpedance(self, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py: def seriesImpedance(self, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py: def qFactor(self, ref_impedance: float = 50) -> float:
NanoVNASaver/RFTools.py: def capacitiveEquivalent(self, ref_impedance: float = 50) -> float:
NanoVNASaver/RFTools.py: def inductiveEquivalent(self, ref_impedance: float = 50) -> float:
NanoVNASaver/RFTools.py:def gamma_to_impedance(gamma: complex, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py:def impedance_to_norm(z: complex, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py:def norm_to_impedance(z: complex, ref_impedance: float = 50) -> complex:
NanoVNASaver/RFTools.py:def reflection_coefficient(z: complex, ref_impedance: float = 50) -> complex:

NanoVNASaver/Windows/CalibrationSettings.py: self.load_resistance.setText(str(self.app.settings.value("LoadR", 50)))

NanoVNASaver/Windows/TDR.py: self.step_response_Z = 50 * (1 + self.step_response) / (1 - self.step_response)
Re: That pesky 50 to 75 ohm conversion.

 

On Fri, Jul 9, 2021 at 04:18 PM, Mikek wrote:


After calibrating OSL with 75 ohms at 500MHz, I connected a 50 ohm, 75
ohm, 100 ohm and 150 ohms terminations.

50 ohm = 34.5 ohm and 2.7nH.? 75 ohm =49.9 ohm 305pf. 100 ohm = 68.9 ohm
3.5nH. 150 ohm = 102 ohm 165pf.
those values are correct, your VNA will display values in a 50 ohms world. It doesn't know that you
transformed to 75 ohms and calibrated with 75 ohms at the other end of the pad. What you have to
do is multiply all resistance values by the 1.5 ratio of your matching pad:
50 ohm => 34.5 ohm x 1.5 = 51.75 ohm
75 ohm => 49.9 ohm x 1.5 = 74.85 ohm
100 ohm => 68.9 ohm x 1.5 = 102.75 ohm
150 ohm => 102 ohm x 1.5 = 153 ohm

There might be an alternative firmware out there which could allow you to change the nominal impedance
(center of smith chart) from 50 ohm to 75 ohm so all the values (resistance and reactance) would be
properly displayed in terms of 75 ohm at the center.

Uwe.
Re: Any more "development " in bluetooth? #bluetooth

Anne Ranch
 

Roger that...
..and 10-4
It is obviously my fault - but you did not say that
If it bothers me that "Johnny still cannot read " I should NOT be participating in forum .
I believe that was a title of some book concerned with
how education system is failing to teach "reading is fundamental " - or something similar to that, do not quote me.

So for the record - there are no visible means of identifying my hardware - no "nanoVNA" ,
no "nanoVNA by XYZ", nothing, nada, nichevo, nic, zero.

And as far as it is acceptable to hijack threads (you did not hijack this one),
if it is OK with the rest of forum users - no more comments from me - "resistance if futile ." .
10-4
Re: Is the the only schematic on the web ? #schematic

Anne Ranch
 

On Sun, Jul 11, 2021 at 11:38 AM, Roger Need wrote:


new version of the NanoVNA which has the P3 connector.
So I would have to ask "seller" - is your version "new" and " does it have I/O pins suitable to connect to bluetooth module?"
Years ago I purchased SDR hardware which required soldering to SMD IC pins...
Asked for and received refund.
Not willing to do that again.
Re: New to the Group and Calibration Question

 

On Mon, Jul 12, 2021 at 05:56 AM, Christopher Maness wrote:


Hello, Chris KQ6UP in SoCal.

Another question. When I initially calibrated and stored in 1 of the possible
setting memories, it retained my desired frequency range. However, I am not
sure how to save a new frequency range so that is what comes up when I first
turn on the unit. Do I have to clear out that memory and its saved
calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP

Best to start off reading this excellent tutorial....

/g/nanovna-users/files/Absolute%20Beginner%20Guide%20to%20The%20NanoVNA/Absolute_Beginner_Guide_NanoVNA_v1_6.pdf

Roger
Re: New to the Group and Calibration Question

 

Also remember that you must press the reset menu item each time before a
calibration , otherwise it doesn't take.

On Mon, Jul 12, 2021, 11:39 AM Larry Rothman <nlroth@...> wrote:

If you use the save under the Calibration menu, it prompts you for a
memory slot #.
If you use the save under the Config menu, it doesn't

On Monday, July 12, 2021, 10:17:46 a.m. EDT, Christopher Maness <
christopher.maness@...> wrote:

> On Mon, Jul 12, 2021 at 6:34 AM Larry Rothman <nlroth@...>
wrote:

Chris,
Memory slot 0 is the startup memory that is recalled when you turn the
nano ON.
I did notice that.
There are two locations where you save on the nano: the memory slots
are used to save freq, marker, scale, trace and cal data and the SAVE
button un the Config menu is used to save general configuration data such
as touch cal, colours, graticule type and a few other items.

How do I make sure I am saving to the memory slots and not the general
config menu? I have a feeling that is what is going on.

As for the cable, you need to calibrate the nano with the cable
attached to take into account the reference plane.

Yes, I figured and that is how I did it, but the calibration did not
zero out the cable. It was not in the correct place for an open stub
that had been calibrated out. The problem maybe connected to my other
problem above. I also noticed that running the thru test did not also
stick. I was showing 1dB loss with the coax going from C0 to C1. It
does seem to work in Memory 0, but I could not get the calibration to
stick in Memory 1. That is likely because I am not saving it
correctly.

You can also use the e-delay in the Scale menu to shift the start of
the measurement to the end of the cable with a 50 ohm reference attached. I
slowly set the e-delay >>until the Smithchart shows 50 ohms and then attach
my unkowns.

What is the theory behind the edelay?

Regards,
Larry
Chris KQ6UP
Re: New to the Group and Calibration Question

 

If you use the save under the Calibration menu, it prompts you for a memory slot #.
If you use the save under the Config menu, it doesn't

On Monday, July 12, 2021, 10:17:46 a.m. EDT, Christopher Maness <christopher.maness@...> wrote:

> On Mon, Jul 12, 2021 at 6:34 AM Larry Rothman <nlroth@...> wrote:

? Chris,
Memory slot 0 is the startup memory that is recalled when you turn the nano ON.
I did notice that.
There are two locations where you save on the nano: the memory slots are used to save freq, marker, scale, trace and cal data and the SAVE button un the Config menu is used to save general configuration data such as touch cal, colours, graticule type and a few other items.
How do I make sure I am saving to the memory slots and not the general
config menu?? I have a feeling that is what is going on.

As for the cable, you need to calibrate the nano with the cable attached to take into account the reference plane.
Yes, I figured and that is how I did it, but the calibration did not
zero out the cable.? It was not in the correct place for an open stub
that had been calibrated out.? The problem maybe connected to my other
problem above.? I also noticed that running the thru test did not also
stick.? I was showing 1dB loss with the coax going from C0 to C1.? It
does seem to work in Memory 0, but I could not get the calibration to
stick in Memory 1.? That is likely because I am not saving it
correctly.

You can also use the e-delay in the Scale menu to shift the start of the measurement to the end of the cable with a 50 ohm reference attached. I slowly set the e-delay >>until the Smithchart shows 50 ohms and then attach my unkowns.
What is the theory behind the edelay?

Regards,
Larry
Chris KQ6UP
Re: New to the Group and Calibration Question

 

On Mon, Jul 12, 2021 at 6:34 AM Larry Rothman <nlroth@...> wrote:

Chris,
Memory slot 0 is the startup memory that is recalled when you turn the nano ON.
I did notice that.
There are two locations where you save on the nano: the memory slots are used to save freq, marker, scale, trace and cal data and the SAVE button un the Config menu is used to save general configuration data such as touch cal, colours, graticule type and a few other items.
How do I make sure I am saving to the memory slots and not the general
config menu? I have a feeling that is what is going on.

As for the cable, you need to calibrate the nano with the cable attached to take into account the reference plane.
Yes, I figured and that is how I did it, but the calibration did not
zero out the cable. It was not in the correct place for an open stub
that had been calibrated out. The problem maybe connected to my other
problem above. I also noticed that running the thru test did not also
stick. I was showing 1dB loss with the coax going from C0 to C1. It
does seem to work in Memory 0, but I could not get the calibration to
stick in Memory 1. That is likely because I am not saving it
correctly.

You can also use the e-delay in the Scale menu to shift the start of the measurement to the end of the cable with a 50 ohm reference attached. I slowly set the e-delay >>until the Smithchart shows 50 ohms and then attach my unkowns.
What is the theory behind the edelay?

Regards,
Larry
Chris KQ6UP
Re: New to the Group and Calibration Question

 

Thanks Larry for clarifying. I am going to play with it in a minute.

Chris KQ6UP

On Mon, Jul 12, 2021 at 6:34 AM Larry Rothman <nlroth@...> wrote:

Chris,
Memory slot 0 is the startup memory that is recalled when you turn the nano ON.
There are two locations where you save on the nano: the memory slots are used to save freq, marker, scale, trace and cal data and the SAVE button un the Config menu is used to save general configuration data such as touch cal, colours, graticule type and a few other items.
As for the cable, you need to calibrate the nano with the cable attached to take into account the reference plane. You can also use the e-delay in the Scale menu to shift the start of the measurement to the end of the cable with a 50 ohm reference attached. I slowly set the e-delay until the Smithchart shows 50 ohms and then attach my unkowns.
Regards,
Larry

On Monday, July 12, 2021, 8:56:43 a.m. EDT, Christopher Maness <christopher.maness@...> wrote:

Hello, Chris KQ6UP in SoCal.

I wanted to do a calibration with some short coax going to a test fixture for a LPF. I was surprised that calibration did not zero out the capacitance as the short open coax (stub) acts like a little capacitor. I found that I was better off just measuring the components 1 by 1 directly at the SMA and trusting the calculations. The LPF did end up working nicely in situ, but I would have rather tuned it in the text fixture. This could be pilot error, or maybe the calibration will not zero out that much capacitance. The lead was short and the frequency was low (14Mhz) though, so I was surprised.

Another question. When I initially calibrated and stored in 1 of the possible setting memories, it retained my desired frequency range. However, I am not sure how to save a new frequency range so that is what comes up when I first turn on the unit. Do I have to clear out that memory and its saved calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP
Re: New to the Group and Calibration Question

 

Chris,
Memory slot 0 is the startup memory that is recalled when you turn the nano ON.
There are two locations where you save on the nano: the memory slots are used to save freq, marker, scale, trace and cal data and the SAVE button un the Config menu is used to save general configuration data such as touch cal, colours, graticule type and a few other items.
As for the cable, you need to calibrate the nano with the cable attached to take into account the reference plane. You can also use the e-delay in the Scale menu to shift the start of the measurement to the end of the cable with a 50 ohm reference attached. I slowly set the e-delay until the Smithchart shows 50 ohms and then attach my unkowns.
Regards,
Larry

On Monday, July 12, 2021, 8:56:43 a.m. EDT, Christopher Maness <christopher.maness@...> wrote:

Hello, Chris KQ6UP in SoCal.

I wanted to do a calibration with some short coax going to a test fixture for a LPF.? I was surprised that calibration did not zero out the capacitance as the short open coax (stub) acts like a little capacitor.? I found that I was better off just measuring the components 1 by 1 directly at the SMA and trusting the calculations.? The LPF did end up working nicely in situ, but I would have rather tuned it in the text fixture.? This could be pilot error, or maybe the calibration will not zero out that much capacitance.? The lead was short and the frequency was low (14Mhz) though, so I was surprised.

Another question.? When I initially calibrated and stored in 1 of the possible setting memories, it retained my desired frequency range.? However, I am not sure how to save a new frequency range so that is what comes up when I first turn on the unit.? Do I have to clear out that memory and its saved calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP
Re: New to the Group and Calibration Question

 

I did my call from 1-15Mhz I believe. I tried to call for the range
that I was going to be using, but my cals don't seem to be "sticking"
anyhow. Not sure what I am doing wrong.

Chris KQ6UP

On Mon, Jul 12, 2021 at 6:25 AM Jim Lux <jim@...> wrote:

On 7/12/21 5:55 AM, Christopher Maness wrote:
Hello, Chris KQ6UP in SoCal.

I wanted to do a calibration with some short coax going to a test fixture for a LPF. I was surprised that calibration did not zero out the capacitance as the short open coax (stub) acts like a little capacitor. I found that I was better off just measuring the components 1 by 1 directly at the SMA and trusting the calculations. The LPF did end up working nicely in situ, but I would have rather tuned it in the text fixture. This could be pilot error, or maybe the calibration will not zero out that much capacitance. The lead was short and the frequency was low (14Mhz) though, so I was surprised.
That *is* odd - I've done cals with 25 or 100 ft of coax (as well as a
foot) and it seems to work ok. One thing I did get caught, by, though,
was when the cal frequency range didn't match the measurement frequency
range. The firmware tries to interpolate, but with a long coax, and big
frequency steps (say you did your cal 0-300 MHz with 101 steps, so the
steps are 3 MHz), I think the interpolation algorithm doesn't always
work, if your measurement range is narrower (say, 10-30 MHz, with 200
kHz steps).


But with a short cable, it should work just fine.



Another question. When I initially calibrated and stored in 1 of the possible setting memories, it retained my desired frequency range. However, I am not sure how to save a new frequency range so that is what comes up when I first turn on the unit. Do I have to clear out that memory and its saved calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP
Re: New to the Group and Calibration Question

 

On 7/12/21 5:55 AM, Christopher Maness wrote:
Hello, Chris KQ6UP in SoCal.

I wanted to do a calibration with some short coax going to a test fixture for a LPF. I was surprised that calibration did not zero out the capacitance as the short open coax (stub) acts like a little capacitor. I found that I was better off just measuring the components 1 by 1 directly at the SMA and trusting the calculations. The LPF did end up working nicely in situ, but I would have rather tuned it in the text fixture. This could be pilot error, or maybe the calibration will not zero out that much capacitance. The lead was short and the frequency was low (14Mhz) though, so I was surprised.
That *is* odd - I've done cals with 25 or 100 ft of coax (as well as a foot) and it seems to work ok.?? One thing I did get caught, by, though, was when the cal frequency range didn't match the measurement frequency range.? The firmware tries to interpolate, but with a long coax, and big frequency steps (say you did your cal 0-300 MHz with 101 steps, so the steps are 3 MHz), I think the interpolation algorithm doesn't always work, if your measurement range is narrower (say, 10-30 MHz, with 200 kHz steps).


But with a short cable, it should work just fine.



Another question. When I initially calibrated and stored in 1 of the possible setting memories, it retained my desired frequency range. However, I am not sure how to save a new frequency range so that is what comes up when I first turn on the unit. Do I have to clear out that memory and its saved calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP
New to the Group and Calibration Question

 

Hello, Chris KQ6UP in SoCal.

I wanted to do a calibration with some short coax going to a test fixture for a LPF. I was surprised that calibration did not zero out the capacitance as the short open coax (stub) acts like a little capacitor. I found that I was better off just measuring the components 1 by 1 directly at the SMA and trusting the calculations. The LPF did end up working nicely in situ, but I would have rather tuned it in the text fixture. This could be pilot error, or maybe the calibration will not zero out that much capacitance. The lead was short and the frequency was low (14Mhz) though, so I was surprised.

Another question. When I initially calibrated and stored in 1 of the possible setting memories, it retained my desired frequency range. However, I am not sure how to save a new frequency range so that is what comes up when I first turn on the unit. Do I have to clear out that memory and its saved calibrations in order to retain the frequency settings?

Thanks,
Chris KQ6UP
Re: Any more "development " in bluetooth? #bluetooth

 

On 7/11/21 12:54 PM, Jose Barros wrote:
Why would one place a radio emitter inside a vna case is above my comprehension
The VNA has a fairly narrow band receiver, so it will sweep through the BT frequencies (which are around 2.4 GHz, so well above the original NanoVNA frequencies, but within the range for other similar models).? You *might* see some reduced SNR if you happen to sweep right onto a BT frequency, but overall, I'd not expect to see much of a problem.


given that these are sub $100 devices, a good strategy is "try it and see" - if you were selling them as a product like Keysight or Anritsu, yeah, you'd carefully shield things, and specifically manage the sweep next to the BT frequencies, etc.


OTOH, people operate VNAs in labs all the time, with their mobile phone or pad, blaring away on BT, not to mention WiFi, without much problem. (unless you're working in that band...).? A bigger problem in labs I've been in is USB and the harmonics from it.


Now, what I wouldn't do is put a BT module inside a shielding metal case - That's sort of asking for "very short range only".
Re: Any more "development " in bluetooth? #bluetooth

 

AMEN, Jose!!!!! I've asked myself the same question. Bluetooth has no
place inside the NANOVNAs or the Tiny SA, for that matter.

Dave - W?LEV

On Sun, Jul 11, 2021 at 5:14 PM Jose Barros <josemanuelbarros@...>
wrote:

Why would one place a radio emitter inside a vna case is above my
comprehension





--
*Dave - W?LEV*
*Just Let Darwin Work*
Re: Any more "development " in bluetooth? #bluetooth

 

Why would one place a radio emitter inside a vna case is above my comprehension
Re: Is the the only schematic on the web ? #schematic

 

On Sun, Jul 11, 2021 at 11:07 AM, Anne Ranch wrote:


Thanks, close but no cigars.
My PCB has no P3 , the schematic explains some reasons why.
It looks as whoever cloned my PCB was concerned about battery life span.
Perhaps my hardware was intended for "portable use only " .
Since your board has no P3 connector there is no convenient place to attach the Transmit (Tx), Receive (Rx) power and Ground Connections from a Bluetooth module like an HC05. For your NanoVNA board the only place to get the Tx and Rx signals is by soldering to the SMD pins on the CPU as shown in the attached diagram. These pins are not connected to anything on your version. You said in another post that you do not want to solder to the chips so Bluetooth on your NanoVNA is not possible.

Your only other option is to buy a new version of the NanoVNA which has the P3 connector. If you don't want to do that either then Bluetooth is not possible.

Roger

Pins.png

16861 - 16880 of 39697