NanoVNA Custom Software

[joeqsmith]

I have a whole bunch of fair-rite very tiny balun cores, beads, etc. I have several materials.
..

Can I attempt to replicate what you are trying to do which I am guessing likely involves either a 1:1 or fixed ratio transformer (1:4?) to match the nanoVNA to the "transmission line" on the breadboard?
..

I am guessing that you are talking about making that crystal test jig on a breadboard which seems like it could work.  You may get better results building it up on perfboard.   I have never looked at any of the other software for the NanoVNA but assume what ever you are using supports the measurement.   

Sweeping 20KHz to 100M, still no transformer and unit is not calibrated.    Plot A, starting with a tant for the bulk. B, adding a 0.1.  C adding a 0.001uF.

Should be fine at 100 MHz.

   

The quote you mention from the following post is testing out some new routines to look at the impedance of the power rails.   The mentioned transformer  is not the impedance matching bifilar transformer used for the crystal test fixture.  Instead its a coaxial transformer used to break the ground loop when making shunt through measurements at low frequencies.   The breadboard allows me to build up simple networks used to test the software.   

https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3559860/#msg3559860

[joeqsmith]

I doubt it would have been a problem.
Jeff Bezos to the rescue.
Personally, I prefer to give my dosh to the little guy.

No worries.  I have no plans to pursue it further.   

[purpose]

That'll do me.

[ars_ha3hz]

Hello joeqsmith,
I read earlier (in 2019) in the forum group, a lot of changes have happened since then. Yesterday I downloaded the program and tested NanoVNA v1.08. I attach the difference between the current firmware 1.0.62 (not yet released by DiSlord) and the shell commands in version 0.8.0. I hope you find it useful. I find the program interesting, but I have a lot to learn about using it. A small spelling: there is an unnecessary 'i' in the text in the info window: 'Th (i) e quick brown fox jumps over the lazy dog' telex test text in which all letters are included.

Gyula

[joeqsmith]

I wouldn't be surprised if you found other spelling errors in the software or documentation.  Please feel free to post them as you find them and I will correct them on the next release.   

Radiolistener's firmware running on the original hardware from a more than a year ago seems to be the only stable combination I have found.   

I think I am up to four software problems with 1.08 so far.  I have no plans to fix any of them.  If I get something I feel is useful working with the PDN measurements, I will release 2.0 for both the original hardware and the V2Plus4.   If you have read the last few pages, this new software will most likely be even less compatible with the newer versions of firmware and various hardware that pops up. 

[joeqsmith]

My latest version of software for the original NanoVNA only uses three commands.  One of those is obviously Freq where I only use two of the sub commands.  After working with the V2Plus, I settled on a much simpler approach. 

[DiSlord]

For NanoVNA version greater 0.7 you can use one command
scan start stop points 7

start - start frequency in Hz
stop - stop frequency in Hz
points - points count (up to 101 for H version, up to 201 or 401 (on my fw) for H4)

It return all measured data after
Example see screenshot
Command return list of measured data
freq s11_re s11_im s21_re s21_im

PS you can use faster exchange mode im my fw (in binary format as vs V2)
scan start stop points 15
In this case returned binary
uint16_t mask;
uint16_t points;
struct {
 uint32_t freq;
 float s11_re;
 float s11_im;
 float s21_re;
 float s21_im;
}data[points]

[purpose]

Mr. DiSlord,
Спасибо for the great firmware. 

[joeqsmith]

DiSlord,

Hopefully my last response solves your regional settings problem.  Make sure you read it carefully. 

Without having the H4 hardware, there is little reason for me to try and run your firmware at this time.   The goal would have been to replicate Purpose's setup in an attempt to help them out.  They have some flavor of an H4 which I was unable to obtain.    Not a big deal as the original hardware will do what they need and it seems that my software at least somewhat works with his H4 setup.   

Again, thanks for providing me with help on the command set.   

[joeqsmith]

Shown with a 7805 regulator, 1.84MHz oscillator, meter and various caps.   The DC blocks and transformer are installed.     

[harrimansat]

Shown with a 7805 regulator, 1.84MHz oscillator, meter and various caps.   The DC blocks and transformer are installed.   

good job!, where can I download VI?

Thanks

[joeqsmith]

Shown with a 7805 regulator, 1.84MHz oscillator, meter and various caps.   The DC blocks and transformer are installed.   

good job!, where can I download VI?

Thanks

I have never released the source code (VIs), only the EXEs.  You will find the link to my Github account in my signature.

Make sure you read the README BEFORE you try and download and install it.   One of the most common problems people ask me is they will download the EXE and then they try and find the runtime for it, then if they manage to find the right version, the software won't find the device because they still haven't installed VISA.      The next problem is people will try and use the wrong software for their device.  They have the V2Plus and are trying to use the software for the original hardware.    I suspect DisLords problem with their regional settings is they didn't change their .INI file covered in the manual. 
There are also cases of trying to use the software with unsupported firmware/hardware.  Most likely 2.x is be be even less compatible.   I suggest archiving the current versions.      If you manage to get it running, the next hurdle will be reading the manual.   

The PDN measurement features are still being developed.  The changes will need to be added to the V2Plus and the manual needs to be updated.  So it's a ways off. 

[joeqsmith]

Using the same setup, the original NanoVNA was swept from 10kHz to 1.5GHz using a thru.  The data was then normalized.   Next both ports were terminated using a common shield.

[joeqsmith]

Looking at the calculated resistance with both ports terminated, it's a bit under 3mOhms from 20kHz to 70MHz.   

The V2Plus4 is limited to 50kHz which we saw from looking at the 100uOhm resistor is already on the high side. 

[joeqsmith]

Using the same setup but removing the homemade common mode transformer.   We wouldn't expect there to be any difference as we are after all testing with the two ports terminated and there should be no ground loop.    More a sanity check than anything.

[joeqsmith]

Using the completed setup to sweep the 100mOhm, 25mOhm and 1mOhm resistors from 20kHz to 70MHz.   Still a few more things to work out but for a $50 VNA, I'm impressed.  The V2Plus4 is limited to 50Khz which is better then my new boat anchor. 

When I was looking for a new VNA I was interested in finding something that would be usable at these lower frequencies.  Asking about what was available in the HPAK group,  some of the feedback suggests not many people have a use for it.   I'm keeping my fingers crossed that the new low cost VNAs are able to measure narrow band devices as well as make some of these low impedance measurements. 

True, but wanting your Microwave VNA to go down to DC is like
wanting your Ferrari to have tank tracks.

It just isn't right.  Better would be to have a Ferrari and
a tank.  They serve different functions.

-Chuck Harris

[joeqsmith]

25, 1 and 0.1 mOhms.   The 0.1 mOhm appears less inductive but we can see its beyond what the original NanoVNA can measure.   

[joeqsmith]

Purple, original setup on the breadboard.   Pink showing the 7805, oscillator and the single alum elec cap.   Green, removing all the caps and cutting the leads of the two tants and adding three ceramics.     Less caps and the impedance is much more stable.     


I came across this video from Keysight talking about the breadboard's claimed 10MHz limit.  I like the little FieldFox but while he says it's 25GHz is overkill for breadboard work, I would imagine it's possible to use a breadboard at frequencies in excess of this     
 

[joeqsmith]

I tried it and sure enough.  No problem though.  Open the .INI file and add the line:

useLocaleDecimalPt=False

Rerun the application.  It will then use a decimal as it should regardless of the regional settings. 

It works, thank you! 

I am curious when you installed the software, did you have the .INI file or if you had to create it?    Member DisLoard appears to be having the same problem but from their latest response it seems they may not have the .INI.   I suspect they may have not installed it from the original installer (which should have had the .INI file).   I tried a clean install for the V2+4 and it shows up in the same directory as the .EXE.  Maybe something else is going on.

The manual includes the contents of the .INI (auto created by LabVIEW during the build) and a copy paste is fine but assuming the install was done in the proper order, you shouldn't have to do this.     I may add more details in the manual.

[cyp_eev]

I am curious when you installed the software, did you have the .INI file or if you had to create it?

I created the NanoVNA.ini file, it was not automatically created after first running the exe.
I thought it's maybe because I did not reinstalled the runtime engine (although it doesn't make much sense). But I haven't done futher tests.

[joeqsmith]

I am curious when you installed the software, did you have the .INI file or if you had to create it?

I created the NanoVNA.ini file, it was not automatically created after first running the exe.
I thought it's maybe because I did not reinstalled the runtime engine (although it doesn't make much sense). But I haven't done futher tests.

Thank you very much.  When you created the file, did you only include the one line "useLocaleDecimalPt=False"?

[cyp_eev]

Thank you very much.  When you created the file, did you only include the one line "useLocaleDecimalPt=False"?

I copy/pasted all the lines from the manual, but i just tested again and it's also working with only the on line in the ini:

Code: [Select]

[NanoVNA] 
useLocaleDecimalPt=False


[joeqsmith]

This is good to know.  Thanks for trying that last test.  I will add more details to the manual about the regional settings.  I have linked these to DiLord's open issue on Gitub.

[cyp_eev]

No problem, you're welcome, anytime.

Thank you!

[DiSlord]

I also create nanovna.ini file in NanoVNA.exe directory, after all work

PS in my fw i also add expert option for change dot or comma as digit separator (sended to console)