Products > Test Equipment

Best VNA for around or under $2000?

<< < (20/23) > >>

knudch:
A few comments

These ADC based devices showing a interesting behavior when you reach the bottom of there dynamic range.
There noise floor seems to "drop" or the accuracy is getting worse...but I guess it is result of digitizing/calculation process.
Values close to "noise floor" seems to be more "guess work" than measurements.
My learning's is to evaluate the "noise floor" don't measure with zero signal but with 5-10dB before the apparent noise floor.

The measurement with Anritsu was added for comparison which of NanoVNA & LibreVNA is more "correct".


Br
Knud 

joeqsmith:
I used two 40dB ZDTECH (cheap parts from another experiment) + 10dB Midwest Microwave. 

LiteVNA 3.2, 50k - 1GHz, 1601 points, normalized using 500Hz IFBW.    Green 90dB attenuator with 500Hz IFBW.   Blue is with both ports terminated.     

Our two H4s, after sorting out the problems,  have a 70dB dynamic range down to around 1.6KHz.  They are both better than what I have been able to achieve with any of the LiteVNAs in this region.

OF course,  1601 points and 500Hz IFBW is slow.  I posted some data where I averaged 3000+ samples with the Lite to get the floor below 110dB but that took several hours to run. 

***

--- Quote ---My learning's is to evaluate the "noise floor" don't measure with zero signal but with 5-10dB before the apparent noise floor.
--- End quote ---
As a general rule, I am wanting at least 10dB of head room. 

H4 dynamic range at lower frequency after repairs/mods.  I doubt if you buy a new one that you would get this level of performance.
https://www.eevblog.com/forum/testgear/best-vna-for-around-or-under-$2000/msg5679611/#msg5679611

H4 measuring a step attenuator from 0 to 80dB and sweeping from 50k to 2GHz.  Also 11th order lowpass filter with 90dB rejection at 30MHz.
https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg5698991/#msg5698991

LiteVNA64 measuring an interdigital with 3500 averages.  Skirts are around 110dB down.   
https://www.eevblog.com/forum/testgear/best-vna-for-around-or-under-$2000/msg5679611/#msg5679611

knudch:
I forgot the LibreVNA settings
Was:
-3 dBm output
1001 point, RBW 1 kHz, sweep time was approx 1-2 sec

301 point RBW 1kHz, sweep time is approx 0,5 sec
401 points RBW 100Hz, sweep time is approx 10sec

No average in any case

NanoVNA sweep time is approx 1-2 sec for 301points, RBW is not known

joeqsmith:

--- Quote from: knudch on December 05, 2024, 01:00:49 pm ---A few comments
...
Values close to "noise floor" seems to be more "guess work" than measurements.
...

--- End quote ---

When I have looked at sub 50kHz with the LiteVNA, seemed non-linear.  As long as I stay in that 50k - 4GHz, 10dB headroom it performs well for what it is.  You can get some useful data down to about 20Khz and up to about 9HGz with it.   

Shown is an old Weinschel 63dB attenuator that I mounted some drivers to.  The plan was to use this to automatically sweep the floor of my original NanoVNA much like you are showing, but I never added software to support it.  For that matter, I am not even sure if this attenuator works.  It's only rated to 2GHz and a 1dB resolution.

LibreVNA is not all that expensive but I think getting to the point where I could automate some of the tests to evaluate it would be a big investment.   It it offered some sort of gains over the LiteVNA/H4/transfer relay/Solver combo, I would have pulled the trigger on it.  The lack of a real review of it, the closed serial interface and then locking me out of their original Groups was more than enough to deter me.  Granted, designer has opened up the serial interface and the groups was a SNAFU on seller's part (why they have two groups now).  Still waiting on an actual review of it beyond an unboxing. 

joeqsmith:
LiteVNA64, normalized with both ports terminated.   Showing 500 and 200Hz IFBW.   I was using the LiteVNA to run some experiments in the X-band by upconverting the 4GHz to 12GHz.  To achieve the highest dynamic range, I was limited to about 10GHz.  The LibreVNA has about the same dynamic range (from data that was provided to me).  It's real advantage is having that 10MHz reference.   For these tests, I had pulled the LiteVNA's internal oscillator and mounted another SMA to the board which I drove from my RF generator (which is locked to the GPS).   This solved the drift problem I was having.   

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod