开云体育

It would be helpful if you described the specific frequency span you are

using, the number of segments,

and the frequencies where the "errors" occur. If there is some sort of

bug in the software algorithm,

that would help to see where it originates.

** Hello
I am posting an example. The calibration file (.cal) corresponds to the
end of a 50 m coaxial cable (50 Ω). The frequency sweep range is 1 to 20
Mhz in 20 segments.

Follows the description of each sheet in the .xls file

.cal
---
Raw data from the .cal file converted through a .csv file

Garden 1-20 MHz 20 sec avg 25-6
-------------------------------------
Data from the .cal sheet

Algo
----
* Column A to G: the input data. It is "by hand" that I zoned in yellow
the cells identified as "in error".

* Column I to N: a first search for errors with formulas of the kind
=IF(ABS((D19+D21)/2-D20)>$J1;1;"")
In cell J1: the value of the maximum difference before being declared
abnormal. In my example 0.005 (0.5%) works fine

* Column P to U: precise identification of the cell in error by a formula
like
=IF(AND(K21=1;K20=1;K22=1);1;"")

* Column W to AB: correction of values ??coming from columns A to G with
formulas like
=IF(R13=1;(D12+D14)/2;D13)

* CSV-final
Corrected data formatting. The first three lines are special to take into
account that the algorithm does not know how to process the first three
lines of the original file
This sheet will be saved in .csv in order to produce the corrected .cal
but do not forget, beforehand, to save the workbook if you want to keep
track of the calculations. I added in this sheet the graph of one of the
shortR column to show that the evolution of the values of sinusoidal form

But it is also true that the nanoVNA hardware also has limitations and

boundaries,

and it is likely that the "error" points you are seeing in your

calibration is in fact a true

measurement, due to a condition in the hardware that happens at the

subject frequency. And so such an "error"

** It is very unlikely since in this case, it is a calibration file (.cal)
following the operation (closed, open load). I don't have access to the
built-in average function in nanaovna-saver. I'm just seeing the obvious
errors easily identify in the produced .cal file. The errors that my
algorithm locates are big errors that produce straight line segments in the
graphs with the worst effect. There may be other finer errors that I can't
locate.

--
F1AMM (Fran?ois)