NanoVNA Custom Software

[joeqsmith]

Wanted to get an idea how the V2+4 filter compares with the Lites.  Using the same narrow band setup on both VNAs.  Changing the Lite's averages from 1,2 & 3.

[realfran]

Your conclusion is?

[DiSlord]

LiteVNA copy V2Plus4?

[joeqsmith]

Your conclusion is?

Phase trimmers have a use and VNAs are fun. 

[realfran]

Is good I got 4 line stretchers from Hirose and one 21cm Airline the airline is calibrated HP/Agilent 11566A AIR LINE 10 CM. The Airlane is god for the TDR test?
I need to learn to use with the VNA the line stretchers.
The Lite-VNA work flawlessly, Dislord make good work to put a lot for use portable but I use the lite only with PC.

[joeqsmith]

There was that person posting about using 18 data points to increase their TDR edge rates.  They were also using a metrology grade Beatty standard that was 10cm.   I really have no idea what their or your goals are for buying them. 

For the phase trimmer, I am not sure what you are trying to learn.  Imagine a section of coax that you can adjust the length.  As you do, the phase changes.  Are you expecting some sort of magic?  You may want to post what your thoughts. 

[DiSlord]

I found this PDF for Unknown THRU (also known as SOLR) calibration

[gf]

Wanted to get an idea how the V2+4 filter compares with the Lites.  Using the same narrow band setup on both VNAs.  Changing the Lite's averages from 1,2 & 3.

At least with the original NanoVNA V2 firmware, the exact shape near the center is not so important, since the IF freqency resulting from mixing stimulus with LO always hits the peak at the center of the filter exactly (+/- phase noise) once the PLLs are locked, due to the chosen frequency plan ¹⁾. An important property of the filter ²⁾ are however the zeros at f_center +/- N * f_IF, where N is an integer >= 1, in order to cancel IF harmonics completely, which emerge as mixing products of the the (square wave) harmonics of stimulus and LO. These zeros are therefore important in order to deal with the square wave drive.

¹⁾ Stimulus, LO, and ADC clock are derived from the same oscillator and are therefore frequency locked to each other. Stimulus and LO frequency are always integral multiples of the minimum frequency step, IF frequency is an integral multiple of the minimum frequency step, too, and ADC clock is an integral multiple of the IF frequency. Since the ADC clock is frequency locked, too, and an integral multiple of the IF frequency, the IF frequency can't miss the center of the digital filter once the PLLs are locked.

²⁾ The IF filter is not a filter in the traditional sense, but the frequency response is eventually an implied property of the (digital) synchronous detector (mathematically the detector is equivalent to a windowed DFT, but calculated only for a single frequency bin). In addition to the synchronous detector, there is just an analog lowpass pre-filter between mixer and ADC, obviously acting as anti-aliasing filter.

Vector averaging of multiple acquisitions inevitably changes the effective frequency response of the detector, since the effective window (over which the DFT is accumulated) becomes a repetition of the primary window function then (possibly even with zero-valued gaps between the repetitions if there happen to be time gaps between the acquisitions which are averaged). Your screenshot did nicely visualize the change of the frequency response . The averaging strategy of the Nano and the Lite obviously deviate from each other. The peak at the center and the zeros at f_center +/- N * f_IF (which are the properties of importance) are not affected by averaging and still retained, though.

[DiSlord]

>Vector averaging of multiple acquisitions

Lite not use vector average. It continue put to DFT bin xN periods (example for AVGx1 use 30 IF periods, for x2 used 60 and so on). Value named as average - different freq range use different IF, and as result have different RBW for some periods.

Lite use int64 for collect measure (V2Plus4 use int32 and if use big avg can overflow, by this reason i limit max AVG count to 20 in my fw for V2Plus4, original fw use vector average and not have this limit)

[gf]

Lite not use vector average. It continue put to DFT bin xN periods (example for AVGx1 use 30 IF periods, for x2 used 60 and so on). Value named as average - different freq range use different IF, and as result have different RBW for some periods.

My feeling (according to the screenshot posted by Joe) is also that the Lite uses a rectangular window, and not a specific window function (as the original NanoVNA V2 firmware does), is this correct?

[DiSlord]

Use gaussian window (as do it original fw on V2, V2Plus4 sources closed and i not have any information)
But for faster calculation/small code, gaussian window apply on sin/cos table (table size 2 IF period) for x1 avg used 15x2 periods. So if set span more you can see modulation from multiplie windows.
I have plans rewrite this part.

[gf]

V2Plus4 sources closed and i not have any information

Is the V2plus4 firmware source not included here (i.e. mostly common code with the other V2 models)?

[DiSlord]

>Is the V2plus4 firmware source not included here (i.e. mostly common code with the other V2 models)?

Yes V2Plus and V2Plus4 init and measure code hided in bootloader part. This github contain only V2 sources. But you can build from it V2Plus and V2Plus4 fw, bootloader on startup change interrupt table and do all work for measure.

[Alextsu]

I found this PDF for Unknown THRU (also known as SOLR) calibration

This link had been already posted here
https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3911030/#msg3911030

There exists another textbook, which covers calibration in brief within a dedicated chapter, written by the author of the method, A. Ferrero:
Cambridge isbn 9781107036413.
But once again, very little info is provided from the practical point of view.

[joeqsmith]

I doubt there are too many papers on the internet I have not looked at by now.  Google is pretty good for rooting out the sources.   I tried to contact one of the persons involved with a few of the patents but had no luck.  It's possible they are no longer alive.  I also ran a few more experiments but the results like always, were less than positive and not worth posting.   

My personal ignorance on the subject and lack of education
Lack of material to learn from
Low cost VNAs may not be suited for it (other error sources dominate and nothing/little to gain)
Publications have errors which can waste time
Publications seldom consider the hardware available and look it the problem from a generic view
Publications like to mix terminology which aids in clouding the topic 

[realfran]

nanoVNA-QT use libxavna

inside i found:
        if(matchedThru)
            return "Unknown thru calibration using 3 fully known one port standards"
                    " and a reciprocal thru standard. Thru standard must-have "
                    "an electrical delay between -90 and 90 degrees, should be "
                    "well-matched (or as short as possible), and have less than 10dB of loss. "
                    "Matching error of the two instrument ports is fully removed.";

So you can try research using Unknown thru calibration
https://github.com/xaxaxa-dev/vna/blob/master/libxavna/calibration.C

This Unknown thru calibration doesn't work?
If work there is a need to reinvent the wheel?
There is some problem with originality?
This came from my personal ignorance on the subject.

[joeqsmith]

...
nanoVNA-QT use libxavna
...

This Unknown thru calibration doesn't work?
If work there is a need to reinvent the wheel?
There is some problem with originality?
This came from my personal ignorance on the subject.

I have not tried the QT software since I attempted to use it to demo the VNA.  It sounds like you have it working so there is nothing that prevents you from conducting some experiments to demonstrate the unknown thru.  Looking forward to seeing your findings.

I did notice they have a new firmware release for the Lite but I really have no idea what changed.   Maybe download that along with their latest QT and use that for your testing. 

https://zeenko.tech/litevna?spm=a2g0o.detail.1000023.1.b5a97ff8tHYEOr

[realfran]

Hello, my questions are rhetoric I'm not sufficiently educated on software, I stated before they came from my ignorance on the subject.
This is the Lite-VNA screen shoot. 

[joeqsmith]

I would guess there is a manual for the QT software and you are able to read.  You should be able to setup the unknown thru method and try it.  Looking at the screen shot you provided, I would just select the calibration type.  Total guess on my part but I would expect that pulldown? allows you to select it.    I'm not sure how it will work with your transfer relay but I assume they support something.   I would start by reading the manual for it.   Post your findings here.  I'm interested in seeing what you come up with. 

[DiSlord]

Changes for LiteVNA in last fw:
I add calibration standard correction (allow input all OPEN/SHORT/LOAD/THRU corrections L/C/delay/Z ..) for calibration
Fixed work in CW mode (correct apply power settings)
Always send not calibrated data to CPU (calibrated only if ask)
Added RAW sample show (for debug and ~SA mode)
Enable Harmonic mode, now Lite can work up to ~9G (yes after 6.3G have limits, but usable)
Max points now 1024 (default 1001 for better steps)
For THRU calibration also used ISOLN measure (allow get more linear results)
Allow save 4 trace image
More traces formats

[Mechatrommer]

Always send not calibrated data to CPU (calibrated only if ask)

Enable Harmonic mode, now Lite can work up to ~9G (yes after 6.3G have limits, but usable)

wow! am i listening right? 9GHz? can you ask Hugen or yourself update the NanoVNA-QT to limit up to 10GHz? i asked Zeenko that but Hugen only updated to 6.3GHz limit. i believe its just one line constant edit in QT's code. i dont have Linux to compile. i think xavna's NanoVNA-QT is the most mathematically complete app among all other PC app options. please?

Max points now 1024 (default 1001 for better steps)

can you increase again next time a little bit to 1601 points? because HP and i believe de facto standard by pro grade VNA is saving 1601 points. i dont mind if its slow, i think the LiteVNA is quick enough already if plotting 1001 points.

thanks for your effort on this.

[joeqsmith]

Changes for LiteVNA in last fw:

Thanks.  You may want to consider adding a simple text file like this to the releases. 

I add calibration standard correction (allow input all OPEN/SHORT/LOAD/THRU corrections L/C/delay/Z ..) for calibration
Fixed work in CW mode (correct apply power settings)
Always send not calibrated data to CPU (calibrated only if ask)

I doubt I will do anything to test the calibration but will certainly try the power settings and uncal data.

Added RAW sample show (for debug and ~SA mode)
Enable Harmonic mode, now Lite can work up to ~9G (yes after 6.3G have limits, but usable)

I've been reading some of the emails about this in the public forums.   Last one, someone was attempting to change the hardware with additional shielding.  I suspect with me running it from the USB, it will be worse.  Look forward to trying this out, although I don't have a way to verify results.   I'll try that home made bandpass filter. 

Max points now 1024 (default 1001 for better steps)
For THRU calibration also used ISOLN measure (allow get more linear results)
Allow save 4 trace image
More traces formats

Max points and defaults for me isn't too much of a problem.   The leakage terms made a significant improvement with the V2+4 above 3GHz.     

[realfran]

Hello, my questions are rhetoric I'm not sufficiently educated on software, I stated before they came from my ignorance on the subject.
This is the Lite-VNA screen shoot.

My simple question is direct not to the Nano-QT software but to the original xaVNA-QT software this whose used for one VNA project about 5 years ago this software claim the Unknown thru calibration using 3 fully known one port standards and a reciprocal thru standard. Thru standard must have an electrical delay between -90 and 90 degrees, should be well matched (or as short as possible), and have less than 10dB of loss. The matching error of the two instrument ports is fully removed..
https://github.com/xaxaxa-dev/vna
My first question is this calibration system work?
I don't try to teach anything I stated in my message to be ignorant on the subject.
I'm NOT able to read or writing any software.

[joeqsmith]

Hello, my questions are rhetoric I'm not sufficiently educated on software, I stated before they came from my ignorance on the subject.
This is the Lite-VNA screen shoot.

My simple question is direct not to the Nano-QT software but to the original xaVNA-QT software this whose used for one VNA project about 5 years ago this software claim the Unknown thru calibration using 3 fully known one port standards and a reciprocal thru standard. Thru standard must have an electrical delay between -90 and 90 degrees, should be well matched (or as short as possible), and have less than 10dB of loss. The matching error of the two instrument ports is fully removed..
https://github.com/xaxaxa-dev/vna
My first question is this calibration system work?
I don't try to teach anything I stated in my message to be ignorant on the subject.
I'm NOT able to read or writing any software.

I tried to do a search but didn't see where it was ever tested, or even implemented.  I assume with you asking, you did your own search and found something.    From below, it seems the Nano QT was spun off from it, so you should be able to run the unknown through with your lite, assuming it was not stripped out.   You may have a hard time finding the old hardware from the kickstart but at least for me, the goal would be to target the Lite (and V2Plus4) anyway. 

https://groups.io/g/xaVNA/topics

[gf]

Briefly skimming the source code I cannot see that NanoVNA-QT supports anything else than "SOL (1 port)" and "SOLT (T/R)" calibration, i.e. 1-port and 1.5-port. No full 2-port emulation via DUT flipping. No external relay. The libxavna obviously includes some unknown thru stuff, but this part of the library seems not to be used by the NanoVNA-QT app.