Which NanoVNA?

[xrunner]

That is as long as I remember not to xmit into port 2 with an HT. 

Wow, I missed that!

LOL that was a joke that came from watching a video of a person who didn't understand the VNA very well - not me!

This video

[joeqsmith]

I spend some time trying to locate 1.2.08 you installed to see if it performed any better.  No luck.  You would need to provide a link to where you found it, or just upload the file here.  If you want to try and replicate my setup, start by installing the most recent firmware:

https://github.com/hugen79/NanoVNA-H/releases

I could not get stable firmware for the V2Plus from their normal distribution as it seems they do not archive them.  I basically started Google searching for every version I could find that people had uploaded and started running them one by one.  What's really bad is OWO had asked me to try the most recent one they had and again it broke.   So they had it fixed at one time then managed to break it again.   In that case, my only criteria was firmware that didn't lock up requiring the VNA to be power cycled to recover.   Imagine if your car had to have it's battery pulled every now and then to reset the firmware.

[xrunner]

I got the file from the files area at the group nanovna-users@groups.io.

https://groups.io/g/nanovna-users

The zip file is -

NanoVNA fw pack H_H4_LiteVNA_V2_V2Plus_V2Plus4

I'll attach it here if it will let me.

[joeqsmith]

Installing your firmware onto my H4 resolves the error at 116MHz.   Now, mind you it took me a few minutes to download the firmware and install it into the VNA.  That's all it took to find this problem.   The fact the current release has this sort of problem says a lot about the firmware's state. 

I mentioned 1.2.14 allows 1.6kHz to 2GHz.    v.1.2.08 allows 800Hz to 2.147483647GHz.   With the H4 being as old as it is, funny that they can't even settle on something basic like this.      My guess is they expect it increases sales.  Like Keysight with their specs for their high end handheld meters.   Hams can say, I have a 2GHz antenna analyzer to look at my 2 meter antenna (while connecting Port2 their HTs output).       Similar to I have a Bird meter so I know my readings are perfect.       I have no idea what would have possibly drove the 800Hz.  If you could get good data off it, it would be nice to have for PDN measurements.   

With the firmware for the original NanoVNA being open source, I guess everyone felt they should start to have a whack at it.  The code was very unstable and updated more than Dave's 121GW.  Much like the 121GW, it was total hit and miss what would work on each new release.   To save me the time of trying to test each version in my search for something stable,  I ended up writing a simple automated regression test.   While I would expect anyone developing code would have a similar setup, I think the fact that we are seeing this problem with the latest firmware, it continues to be the wild west of coding.  Throw something against the wall and see what sticks.  Let the end users test our code.  It complied without errors, so it's good to go.   I didn't change anything that would have effected that...   Same old song and dance

To run it against 1.2.08 requires a few changes.   I've gone ahead and made those and have it running. It will require several hours to complete but I will upload the report once it is finished.   If I had any plans to continue to support the older protocol, I would just start over and include it in my software.   

[thinkfat]

There is an interesting secondary use for the H4: phase and frequency analysis. Local forum user @erikka has made an alternative firmware available that turns the nanoVNA-H4 into a phase and frequency analyzer with quite interesting properties. You can find that over on tinydevices.org.

Not trying to derail the thread, sorry. Just mentioning that the H4 is an interesting piece of test gear beyond the "entry level" VNA.

[joeqsmith]

Sorry, the report was too large to attach but it did complete and no problems were found.   Like I posted earlier,  don't consider this a stamp of approval as I only use a very small subset of the firmware's features but still this 1.2.08 version you found is a very good starting point.   

[bingo600]

That poor guy.  We all make mistakes but that was a doosy. 

I pulled the H4 from the plastic wrapper and I'll be dammed if it didn't still have a charge and powered right up.  Looks like 1.0.38 was the last firmware I loaded.   Looks like they are up to 1.2.14.  I went ahead and loaded it in and fired up my old software.  It seems to connect just fine.   For some reason, they now allow 1.6kHz to 2GHz.  Fairly large error at 116MHz (see Smith chart).   Looking at the last time I had tried to run a regression test on it, the H4 failed the standard deviation test but at a different frequency range.   I would need to go back and read Dislords posts to see if there was a work around.   Guessing the ham antenna analyzer group wouldn't be too concerned as they don't use this frequency.

I have one of the "New H4's" ... Late 2022/2023
It has a MS5351 as opposed to the older models having a Si5351 (see pict , if it has MS in the end of the model)

The firmware i use is 1.2.19 (latest) , but when loaded it defaulted to a Si5351.

It seems like you can set your SI/MS model in : CONFIG>EXPERT SETTINGS>MORE>MODE
Need save config after change MS/SI mode (CONFIG->SAVE) this allow store settings

See here
https://groups.io/g/nanovna-users/topic/h4_problem_after_firmware/96248196

The selection of the MS if that's what's in there , should correct some "quirks" , and actually use the "better MS scan mode"


Edit:
HO-RO has a git page w. the latest Dislord firmware here
https://github.com/Ho-Ro/NanoVNA-D/tree/current-bin-hex


/Bingo

[joeqsmith]

HW: 4.2.    Looks like I purchased it around June 8 2021

https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3584301/#msg3584301

It appears that while the firmware Xrunner found is fairly stable, the VNA lost some performance in the lower frequencies compared with the previous version I was testing.

Thanks for the firmware link.  If I decide to pull the H4 out of the box again, I will have a look.   

[joeqsmith]

Comparing the latest LiteVNA (HW rev:3.1) with the H4(HW rev:4.2) using Xrunner's firmware.  This is looking at S21 after cal using the same standards and cables.  Cable was then removed and both ports were terminated.   

The older firmware I tied with the H4 improved its lower freq response.  Pick your firmware poison.

Both LiteVNAs had been updated to Dislord's wide range experimental firmware.    Their performance appears very similar.   

[Kahooli]

I own a LibreVNA. It's a nicer device if you plan to use it as a bench device since you have the native software and large screen.
I have compared data results using it Up to 400Mhz and it's pretty decent compared to the E5061B I have at the office.

[joeqsmith]

I own a LibreVNA. It's a nicer device if you plan to use it as a bench device since you have the native software and large screen.
I have compared data results using it Up to 400Mhz and it's pretty decent compared to the E5061B I have at the office.

You should take the time to post some data for your LibreVNA showcasing its performance.   Back a page, I provided a link to some data for various low cost VNAs which included the LibreVNA. 

https://www.eevblog.com/forum/testgear/warning-about-nanovna-clones-and-comparison-with-siglent-vna/msg4626466/#msg4626466

Since then, I have acquired a new version of the LiteVNA which further improves it's performance.   Odd your Keysight wouldn't be dramatically better.  That data included my very old Agilent PNA.   

A few of us were wondering about the high amounts of ripple that several users show with the LibreVNA.   It looks like it has a major problem but it could be the users not knowing how to run it.  I tried to watch a few reviews for it.  Some of talking head hams got the hardware for free to review but outside of unboxing it, that's about all you get.  I've yet to see an actual review from someone who knew what a VNA is.   With you having the Keysight, consider putting something together. 

***
For some background about the LibreVNA's ripple, read this and follow the links provided.
https://www.eevblog.com/forum/testgear/sub-$1000-vna-for-antenna-matching/msg4546259/#msg4546259

[joeqsmith]

Comparing the data provided by one of the LibreVNA owners with the latest LiteVNA hardware.

Calibrating the LiteVNA using sorted Mini-circuits load and the short, open, thru from the V2Plus4 kit. Ideal model.  Both ports terminated measuring S21.

Libre_Lite_PNA_S21term_10GHz_10Avg:  Sweeping the LiteVNA from 100kHz to 10GHz with the leakage term enabled, 2kHz IFBW (default) and 10 averages.  I included my old PNA as a reference.   

Libre_Lite_PNA_S21term_1GHz:  Looking at the data below 1GHz, with no averaging.  Notice how the LibreVNA appears to perform really poorly at the lower end.  I'm sure it meets their spec but we are comparing a $120 vs $730 (current price from Amazon for the LibreVNA).    Not owning a LibreVNA to collect my own data, I can only go by what other users have posted.  You do get a full 2-port VNA with drift and you get to mount heatsinks to it to make it look cool.   Then there is always that question of  where does that ripple come from....

***
The LiteVNA's firmware normally limits the frequency range to 6.3GHz.

[joeqsmith]

Of course, if you want to spend some cash, the Siglent SAV1032x sells for $3,470.   Again a user was nice enough to repeat the test using an IFBW of 10kHz.   The LiteVNA has a maximum IFBW of 4kHz. 

I do like that new LiteVNA hardware.

***
For that price, it looks like Siglent includes one of their famous rotary encoders. 

[exampleguilty]

Basic NanoVNA-F or H is good enough for learning, NanoVNA-F V2 if you want nice mechanical design, a bit more frequency range and better display.

Apart from that, NanoVNAs are build for low cost, low part count and tiny size, people at HP and Rohde-Schwarz did it better, 40 years ago.
Go for like an old HP-8753 if you want lab grade instrument, that doesn't do things like using square waves as test signal, that messes up most active circuits.

If they had increased budget by maybe $50-$100, the could have used something like a local oscillator/mixer design with a couple of ADF4350s in their pure spectrum range between 2.2-4.4GHz and a few switched filters to get a nice sinewave test signal istead of the nasty square waves from SI5351 and divided down ADF4350.

User interface/touch screen is implemented quite nicely, though.

Once I have a firm grasp of the fundamentals of RF, I want to pursue a career in microwave engineering. If you like a great mechanical design, a little more frequency range, and a better display, consider the NanoVNA-F V2.

[Vtech]

Comparing the data provided by one of the LibreVNA owners with the latest LiteVNA hardware.

Calibrating the LiteVNA using sorted Mini-circuits load and the short, open, thru from the V2Plus4 kit. Ideal model.  Both ports terminated measuring S21.

Libre_Lite_PNA_S21term_10GHz_10Avg:  Sweeping the LiteVNA from 100kHz to 10GHz with the leakage term enabled, 2kHz IFBW (default) and 10 averages.  I included my old PNA as a reference.   

Libre_Lite_PNA_S21term_1GHz:  Looking at the data below 1GHz, with no averaging.  Notice how the LibreVNA appears to perform really poorly at the lower end.  I'm sure it meets their spec but we are comparing a $120 vs $730 (current price from Amazon for the LibreVNA).    Not owning a LibreVNA to collect my own data, I can only go by what other users have posted.  You do get a full 2-port VNA with drift and you get to mount heatsinks to it to make it look cool.   Then there is always that question of  where does that ripple come from....

***
The LiteVNA's firmware normally limits the frequency range to 6.3GHz.

Hi!
Looks like you have a large collection of data for different VNAs. Would you be so kind as to post a comparison of LiteVNA, NanoVNA-H4, NanoVNA-V2Plus4 and maybe some others in the frequency range 150kHz - 230 MHz?

I would like to do some impedance measurements in this frequency range (coupling-decoupling networks, CDNs) and maybe some radiated RF disturbance in the 30 MHz - 1GHz range. I'm leaning towards buying LiteVNA based on your positive opinion about it. Am I correct that Zeenko Store on Aliexpress is the best source?

[joeqsmith]

Comparing the data provided by one of the LibreVNA owners with the latest LiteVNA hardware.

Calibrating the LiteVNA using sorted Mini-circuits load and the short, open, thru from the V2Plus4 kit. Ideal model.  Both ports terminated measuring S21.

Libre_Lite_PNA_S21term_10GHz_10Avg:  Sweeping the LiteVNA from 100kHz to 10GHz with the leakage term enabled, 2kHz IFBW (default) and 10 averages.  I included my old PNA as a reference.   

Libre_Lite_PNA_S21term_1GHz:  Looking at the data below 1GHz, with no averaging.  Notice how the LibreVNA appears to perform really poorly at the lower end.  I'm sure it meets their spec but we are comparing a $120 vs $730 (current price from Amazon for the LibreVNA).    Not owning a LibreVNA to collect my own data, I can only go by what other users have posted.  You do get a full 2-port VNA with drift and you get to mount heatsinks to it to make it look cool.   Then there is always that question of  where does that ripple come from....

***
The LiteVNA's firmware normally limits the frequency range to 6.3GHz.

Hi!
Looks like you have a large collection of data for different VNAs. Would you be so kind as to post a comparison of LiteVNA, NanoVNA-H4, NanoVNA-V2Plus4 and maybe some others in the frequency range 150kHz - 230 MHz?

I would like to do some impedance measurements in this frequency range (coupling-decoupling networks, CDNs) and maybe some radiated RF disturbance in the 30 MHz - 1GHz range. I'm leaning towards buying LiteVNA based on your positive opinion about it. Am I correct that Zeenko Store on Aliexpress is the best source?

If I provide you with data for the low range you first ask about, and you base your decision on that.  Then you try to measure at the high range and get poor results, what was the point?   As I have written many times,  I like the LiteVNA for experimenting above 300MHz.   Below 300, I still like the original NanoVNA because it is faster, and much more capable of making narrow band measurements.    Once you try to use any of them with harmonics, the performance is total ass. 

They are so cheap, if you really feel you can make use of one, I would just get one of each to cover your bases. 

[Vtech]

They are so cheap, if you really feel you can make use of one, I would just get one of each to cover your bases. 

Thanks!
I was thinking about doing so - buying one of each. Where can I find "the original NanoVNA"? There are so many of them - I'm lost.
I'm mostly interested in the low range and I don't need stellar performance. I was using a borrowed NanoVNA V2 clone and it did the job just fine so I decided to buy one but then I realized that there are so many of them that I don't know which one to choose.

[joeqsmith]

Sorry but I am not in sales and really have no idea where you would purchase one.  Being outside the USA, compounds your question. 

I have an H4 but if you took the time to read this short thread, you know I never used it and didn't find stable firmware for it until recently.   consider too that you may want it to run stand alone, or with one of the open sourced programs, which is not how I run them.  So lots of variables.    My NanoVNA was purchased in 2019 and is still running firmware from that era.

Most people that buy them have never used a VNA or have any idea how they work or what they can be used for.  IMO, most would be ahead by reading and saving their money until they have a better understanding.

[xrunner]

IMO, most would be ahead by reading and saving their money until they have a better understanding.

Where do you come up with these radical ideas?

That's why these things don't come with manuals - because you aren't supposed to read one! 

[Vtech]

Sorry but I am not in sales and really have no idea where you would purchase one.  Being outside the USA, compounds your question. 

This is the problem - the original NanoVNA v1 is nowhere to be found. NanoVNA-H4 seems to be the next best thing but without having one I don't know if it has similar performance in the lower frequency range.

[Solder_Junkie]

I'm mostly interested in the low range and I don't need stellar performance. I was using a borrowed NanoVNA V2 clone and it did the job just fine so I decided to buy one but then I realized that there are so many of them that I don't know which one to choose.

Just get a H4 from one of the official suppliers, for best results use it with the free Windows software.
You can also turn it into a TinyPFA in a matter of a couple of minutes by loading replacement firmware, it can be just as easily returned to a NanoVNA. This dual use only applies to a genuine NanoVNA H4 version.

SJ

[joeqsmith]

Sorry but I am not in sales and really have no idea where you would purchase one.  Being outside the USA, compounds your question. 

This is the problem - the original NanoVNA v1 is nowhere to be found. NanoVNA-H4 seems to be the next best thing but without having one I don't know if it has similar performance in the lower frequency range.

Because you didn't read this post, or because you were asking something else?   If there was a specific test you would like me to run, just ask. 
https://www.eevblog.com/forum/rf-microwave/which-nanovna/msg4703390/#msg4703390

[Vtech]

Because you didn't read this post, or because you were asking something else?   If there was a specific test you would like me to run, just ask. 
https://www.eevblog.com/forum/rf-microwave/which-nanovna/msg4703390/#msg4703390

Oops! My bad. I've seen this but totally forgot about it. Thanks.

[joeqsmith]

No problem.  If there is something else you would like me to try just let me know.   

[Smokey]

Who is the OEM for the LiteVNA64? 
Is it these guys?  https://www.zeenko.tech/litevna