¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. Nanovna-Users
  2. Topics

Calibration issue ?

 

Hi, new to this group.

NanoVna ttrftech v. 0.5.4
NanoVnaSaver v. 0.3.7

I'm trying to use a home made fixture for testing chokes via S21 parameters, two short wires exiting VNA ports ending with clips

I've done a S.O.L.I.T. calibration with clips connected each other (thru).

Once calibration is applied and a sweep is done with fixture still in thru position (as per calibration) , software still shows the variable amount of reactance and a slight resistance variation across the sweep caused by the non-50ohm fixture.

Shouldn't I see a flat impedance 50+J0 ohm all along the frequency sweep instead, due to calibration correction factor ??

Am I perhaps wrong in calibration concept ??

Thank you

Federico
ik3umt
 

Post the graphs or the S21 files from NanoVNA Saver.

On Tue, 29 Sep 2020 at 12:48, Federico - ik3umt <ik3umt@...> wrote:

Hi, new to this group.

NanoVna ttrftech v. 0.5.4
NanoVnaSaver v. 0.3.7

I'm trying to use a home made fixture for testing chokes via S21
parameters, two short wires exiting VNA ports ending with clips

I've done a S.O.L.I.T. calibration with clips connected each other (thru).

Once calibration is applied and a sweep is done with fixture still in thru
position (as per calibration) , software still shows the variable amount of
reactance and a slight resistance variation across the sweep caused by the
non-50ohm fixture.

Shouldn't I see a flat impedance 50+J0 ohm all along the frequency sweep
instead, due to calibration correction factor ??

Am I perhaps wrong in calibration concept ??

Thank you

Federico
ik3umt
 

Here they are, fresh from nanovnasaver after SOLTI calibration :
NanoVna has been boxed with a pair of SMA to N adapters.

73
Federico
ik3umt

cal1.JPG
cal2.JPG
290920_1.cal
290920_1.cal
IMG_8309.jpg

 

Hi Federico,

you said above that you did the "through" calibration with the clips connected. It puzzles me, though, that the S21 readout is 1.4 dB. Regardless of how good your fixture or your calibration is, you should always measure a S21 very close to 0 dB it you leave the setup in the "through" calibration position.

As for why you measure the S11 (and impedance) you do: Can you explain more in detail how you did the other calibration steps (open, short, load)? Note that the reference plane for S11 (and thus for the impedance calculated from it) is set by the OSM calibration.

Regards,
Christian
 

Christian, my macroscopic mistake ;-(

I was doing SOLI calibration with standard "N" references connected directly to VNA ports, and the only T calibration applying my fixture with clips connected !!

Once repeated the whole calibration procedure with standards applied between clip and ground (except T) , thus taking into account fixture reactances, all is gone as expected.

Thanks to Piero I0KPT to open my eyes.....


Dragan, thank you for spreadsheet , how to use it ??

I see the columns A to E are pure data, the others are calculation from previous ones

Should I take data as is from .s2p file ??

Thanks

Federico
ik3umt
 

P.S. in the meanwhile I upgraded device to v. 0.8.0 and restart from scratch...

Federico
ik3umt
 

Federico, copy the first 5 columns from the .s2p file produced by nanovna-saver (these column should be FREQUENCY, S11 REAL, S11 IMAG, S21 REAL, S21 IMAG) and past them into the first 5 columns of the Excel spreadsheet from Dragan.

Piero, I0KPT

Il 30/09/2020 10:29, Federico - ik3umt ha scritto:
Dragan, thank you for spreadsheet , how to use it ??

I see the columns A to E are pure data, the others are calculation from previous ones

Should I take data as is from .s2p file ??

Thanks

Federico
ik3umt
 

Amazing !

Now, I don't know if nanovnasaver author is here reading this group: wouldn't be interesting to have R X and Z of a "through" measure to be calculated automatically and eventually displayed by software like S11 already is ??

Thank you again !

73
Federico
Ik3umt

-----Messaggio originale-----
Da: [email protected] <[email protected]> Per conto di Piero Tognolatti
Inviato: mercoled¨¬ 30 settembre 2020 10:37
A: [email protected]
Oggetto: Re: [nanovna-users] Calibration issue ?

Federico, copy the first 5 columns from the .s2p file produced by nanovna-saver (these column should be FREQUENCY, S11 REAL, S11 IMAG, S21 REAL, S21 IMAG) and past them into the first 5 columns of the Excel spreadsheet from Dragan.

Piero, I0KPT

Il 30/09/2020 10:29, Federico - ik3umt ha scritto:
Dragan, thank you for spreadsheet , how to use it ??

I see the columns A to E are pure data, the others are calculation
from previous ones

Should I take data as is from .s2p file ??

Thanks

Federico
ik3umt
 

Just make sure that you use the second version
where ATAN2 is used for phase calculation.

Dragan, thank you for spreadsheet , how to use it ??

I see the columns A to E are pure data, the others are calculation from
previous ones

Should I take data as is from .s2p file ??

Thanks

Federico
ik3umt
 

Well, at last I've loaded my .s2p into excel and generate a graphic of my choke..
I would like to know how reliable is the whole thing.....

Federico
ik3umt

choke_1.JPG

 

The formulas were taken from G3TXQ spreadsheet so compare with it?
If you are good with excel/libreoffice graphs you can mark the regions
where Z > 2K and R > |X| (the desired resistive choke) same as G3TXQ
did in his graphs.

BTW read up on stacked cores on G3TXQ site (if it is down use archive.org,
it gets hacked on a regular basis). Usually it is better to use them in
series.

On Wed, 30 Sep 2020 at 17:45, Federico - ik3umt <ik3umt@...> wrote:

Well, at last I've loaded my .s2p into excel and generate a graphic of my
choke..
I would like to know how reliable is the whole thing.....

Federico
ik3umt





s21_calcs.ods
s21_calcs.ods
choke_impedances_g3txq.png

1 - 11 of 11