nanoVNA vs SVA1015X VNA (Need entry level VNA)

[spacehen]

I'm looking for an entry level VNA to add to my RF lab I am setting up. I already have a DSA815-TG spectrum analyzer so my question essentially is: is it worth the extra money to buy the SVA1015X which is also a spectrum analyser with limited VNA capability or should I just buy a nanoVNA. I need to make reasonably accurate S parameter measurements for amplifiers, filters and antennas from 10MHz to 1GHz

Thank you 

[tomud]

IMHO to buy SVA1015X is not very profitable in your case. As for nanoVNA, it depends on the model you want to buy. The first version is not very suitable if we are talking about measurements up to 1GHz (in fact, it works quite well up to 300MHz).

I would consider nanoVna v2 plus4 (original only) or if you have more money to spend, you can buy libreVNA (IMHO A very good option). You can also think about selling the DSA815-TG and buying the SSA3021X Plus which can be converted to SVA1032X.

Although if we're talking about the VNA itself, libreVNA will probably be the best choice (price / features / parameters) - especially since it's a full dual port VNA (gives the possibility of measurements S11/S12/S21/S22).

As for the accuracy of the measurements, I will not answer here, it is too big a topic. Much depends on the device, cables, calibrators. You can find the technical parameters of all devices on the Internet, you have to see them yourself and answer the question if it is enough for you.

[spacehen]

Maybe I should just save up and buy a proper VNA, thanks anyway.

[joeqsmith]

Maybe I should just save up and buy a proper VNA, thanks anyway.

With the very small bit of information you provided, I wouldn't suggest it due to the high costs.  Rather wait until you learn enough to know what you need. 

For learning the basics, the NanoVNA is perfect. If damaged, the low cost allows them to be replaced.   As suggested, the original one can put up some decent data to about 300MHz.   The V2Plus4 is certainly one option to experiment above that.  Personally, I have not used mine since I picked up the LiteVNA.   Personally,  I use a PC to control my lab which includes these low cost VNAs.  While I no longer update the software for the original NanoVNA, I do continue to develop for the V2Plus4 and LiteVNA.   You can find that software here:

https://github.com/joeqsmith/Solver32-LiteVNA-V2Plus4

If you are interested in seeing how they perform, I have created a few videos showing them with my software.   Of course, with low cost we have square wave drive, low dynamic range, lack of features like bias-Ts and attenuators.  Video below talks about the square wave drive.       

     

[graybeard]

I have both a nanovna-f and a sva1015x.

The sva1015x works well from 10MHz to 1.5 GHz.  If i am working in that range it is the one I use.  I especially like the log frequency sweep feature and slightly greater dynamic range.  The interface is much nicer on the Siglent.  It is also faster.

However the nanovna-f works far better at lower frequencies.  The price is much better too.

[nctnico]

I have both a nanovna-f and a sva1015x.

The sva1015x works well from 10MHz to 1.5 GHz.  If i am working in that range it is the one I use.  I especially like the log frequency sweep feature and slightly greater dynamic range.  The interface is much nicer on the Siglent.  It is also faster.

However the nanovna-f works far better at lower frequencies.  The price is much better too.

Any idea about how the LibreVNA would perform? This looks like a very nice kit for the money.

[tomud]

Any idea about how the LibreVNA would perform? This looks like a very nice kit for the money.

It works very well and as I wrote when someone wants a good VNA for little money - this is probably one of the better options. Better than SVA1000X (as soon as we consider using it as a VNA). The SVA1000X has the advantage that the VNA is only one of the functions, unfortunately the poor dynamics of the VNA is a limitation in the case of measurements of filters with higher attenuation.

[nctnico]

Thanks. Can you comment on the lowest frequency the LibreVNA is useful at? The purpose I have in mind right now is looking at PDN (power distribution network) and measuring combinations of decoupling capacitors. Measurements go from sub 1MHz to several hundred MHz (close to 1GHz). I have two VNAs; one goes from 10Hz to 300MHz and the other from 5MHz to 4GHz. There is overlap but for some measurements it sucks needing two instruments.

[tomud]

@nctnico

I had libreVNA borrowed from a friend and unfortunately I did not test below 3.5MHz, I was basically testing in terms of ham bands. However, I have not heard any opinions that it works beyond the declared specification.

Here is a quote about the lower frequencies:

Due to this being an open source VNA, there are some limitations in the design. These remarks are as follows:
Due to some structural limitations the S12 isolation is 10 dB worse than that of the S21 parameter. Above 3GHz the performance decreases with increasing frequency causing the port isolation to only be 50 dB at 6 GHz. Furthermore, below 1 MHz, the performance decreases with decreasing frequency and the effective dynamic range decreases to 70 dB at 100 kHz. The output power may decrease when frequency is lower than 1 MHz or higher than 3 GHz.
With this VNA the use of higher accuracy calibrators allows for higher precision of measurements.

Personally, I was supposed to buy libreVNA ... unfortunately the pandemic and the crisis on the electronic components market hit my finances quite hard and at the moment I am forced to give up all purchases of measuring equipment for hobby applications

[bd139]

I would define what measurements that you need to make carefully before buying one. Write down use cases, examples and test criteria.

The existing DSA815-TG may be adequate if you add a directional coupler to it.

[TopQuark]

I have the nanoVNA V2 plus, the SVA1032X and the LibreVNA.

I'd recommend the LibreVNA (full 2 port S params), but the nanoVNA V3 (aka VNA6000) is apparently just around the corner, so I would wait and see how that thing performs if you only need to measure S11 and S21 without flipping the DUT manually.

I'll be reviewing the LibreVNA in a couple of weeks, but the TLDR is:
- Software is not very polished (features are a bit hidden, may freeze up after a few hours of continuous use for some users), but has some advanced features found only in VNAs many times more expensive (TRL cal, time gating, 2x thru de-embedding etc.)
- Hardware can get hot and readings may drift, but not a problem if you stick a heatsink (optionally a fan) on it.
- 6GHz or even higher if using harmonics, giving good time domain resolution for the price.
- Full 2-port (measures S11, S21, S22, S12)
- I think it is the best value for money in the low cost VNA market right now, open source everything is a big plus.

The SVA1032X IMO is primarily a SA with a VNA tacked on it. The features and usability have improved over time, but is not nearly as powerful as the LibreVNA in terms of being a VNA (T/R vs full two port, advanced de-embedding etc. etc.) . I rarely use the SVA1032X as a VNA after I got my LibreVNA.

Thanks. Can you comment on the lowest frequency the LibreVNA is useful at? The purpose I have in mind right now is looking at PDN (power distribution network) and measuring combinations of decoupling capacitors. Measurements go from sub 1MHz to several hundred MHz (close to 1GHz). I have two VNAs; one goes from 10Hz to 300MHz and the other from 5MHz to 4GHz. There is overlap but for some measurements it sucks needing two instruments.

The LibreVNA is designed for 10MHz and up IIRC, but is usable down to 100kHz in my experience (and I do use it down that low), though the S21 dynamic range is only around 40-50 dB from 100k to 1M, not the 90dB dynamic range higher up in frequency.

[joeqsmith]

I would define what measurements that you need to make carefully before buying one. Write down use cases, examples and test criteria.

The existing DSA815-TG may be adequate if you add a directional coupler to it.

Agree.  You could end up wasting money that could have been put towards something that actually met your needs.   That's the advantage of the NanoVNA.  At $50, we are not talking about a big investment.  If you are just starting with VNAs, the small investment in your education would be a wise choice, IMO.   

I have a few of these low cost VNAs now and besides the lack of peripherals, low dynamic range and squarewave drive (pretty much what we would expect), the main problems I have run into have been firmware.   I have an H4 I wouldn't give away because the firmware was never stable enough to use it.   

- Software is not very polished (features are a bit hidden, may freeze up after a few hours of continuous use for some users), but has some advanced features found only in VNAs many times more expensive (TRL cal, time gating, 2x thru de-embedding etc.)

I'm a little disappointed to read TopQuarks post above about the LibreVNA.   It's been available for some time now and I would expect it to be very stable.  We had discussed the lack of documentation which would allow a third party to create software for the LibreVNA in another thread. 
https://www.eevblog.com/forum/rf-microwave/clone-85033e-vna-cal-kit-measured-against-pico-cal-kit-using-librevna/
It's been mentioned that you can write scripts to control their software but if that's not stable to begin with...     Then there's that whole heat sink deal. 

[tomud]

@joeqsmith

It must be remembered that libreVNA is still budget equipment with a good price / performance ratio. Haves some flaws but at this price they are forgivable. Let's say this, my set of cables and calibrators costs a lot more than libreVNA.

However, poor documentation is unfortunately a feature of probably all open source projects. From my practice, whenever I wanted to change something into open source, it required many hours of source code analysis...

I also bought NanoVNA once, but more out of curiosity. The offered price is ok. However, if someone wants to measure filters up to 1GHz, it's a pity to spend even $50. IMHO it is better to collect some money and buy something that will work better in these applications (I keep talking about equipment to buy at a low price compared to a professional VNA).

Personally, I mainly use nanoVNA in the field to tune HF antennas. Currently, if I had to buy my first VNA, I would  start with nanoVNA V2 or better.

[joeqsmith]

@joeqsmith
...
I also bought NanoVNA once, but more out of curiosity. The offered price is ok. However, if someone wants to measure filters up to 1GHz, it's a pity to spend even $50. IMHO it is better to collect some money and buy something that will work better in these applications (I keep talking about equipment to buy at a low price compared to a professional VNA).

...

Exactly my point.  If you know what you need, buy tools that fit those needs.   If you don't know what you need, start with the cheapest thing out there to at least learn something before you start making poor choices.   I'm not in a position where $50 poses a significant financial burden for my family.   In the video above, you can see some of my old test equipment for my hobby use.   Like you, the costs of what you see far exceed the total amount I have spent on these low cost VNAs.  That was hardly the point but rather the goal was to provide an example to a beginner of where these low cost VNAs may fall short.       

[tomud]

@joeqsmith

It's not even about money anymore, the price difference between nanoVNA is not that big. However, one gets a bit better dynamics and a wider frequency range.
I used to buy a lot of different equipment myself, but suddenly I ran out of space in my small workshop. I ended up giving these devices to my friends for free because I would have wasted more time selling on eBay etc than the hardware was worth. I think that if someone already has a spectrum analyzer with a tracking generator, they already have some knowledge of how to use such equipment so as not to damage it.

[tautech]

Any serious antenna work rules these cheap devices out.
Sure in a lot of the world 315 MHz is available for telemetry but in many other places it's reserved for other needs and 433 MHz used instead which is getting past where even the proponents of these devices ascertain is beyond their accuracy capabilities without even venturing into the popular 800 -900 MHz bands.

But then of course there are the LF Ham bands which do fall within cheap instrument capability.

[joeqsmith]

Any serious antenna work rules these cheap devices out.
Sure in a lot of the world 315 MHz is available for telemetry but in many other places it's reserved for other needs and 433 MHz used instead which is getting past where even the proponents of these devices ascertain is beyond their accuracy capabilities without even venturing into the popular 800 -900 MHz bands.

But then of course there are the LF Ham bands which do fall within cheap instrument capability.

I think we paid about $130 for the LiteVNA which I consider cheap.   I soldered up a couple of pork and bean cans and swept the antenna with the Lite.     

https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg4175104/#msg4175104

I think this was tuned around 2.6G or so.   Data was collected using this software:
https://github.com/joeqsmith/Solver32-LiteVNA-V2Plus4

[joeqsmith]

@joeqsmith

It's not even about money anymore, the price difference between nanoVNA is not that big. However, one gets a bit better dynamics and a wider frequency range.
I used to buy a lot of different equipment myself, but suddenly I ran out of space in my small workshop. I ended up giving these devices to my friends for free because I would have wasted more time selling on eBay etc than the hardware was worth. I think that if someone already has a spectrum analyzer with a tracking generator, they already have some knowledge of how to use such equipment so as not to damage it.

I also gave up selling on eBay years ago.  It goes to the recycle bins or I give it away.   My home labs becoming a bit of a functional museum. 

[bd139]

700 items sold here including at least two NanoVNAs. Gotta eat 

[rf-loop]

I have both a nanovna-f and .....

However the nanovna-f works far better at lower frequencies.  The price is much better too.

Imho, that model (genuine Deepelec, not a copy ) is a pretty good machine (in its price range). Well done. When I lived in China, I bought two of them directly from Deepelec official store, version 3.1. I brought them with me from there. Of course I also left some devices in China so I don't have to get it again later. Sure, it's much better than the old Nano V2. The Nano V2plus4 is a bit different and better in terms of dynamics,  although in terms of manufacturing quality it is almost at the poor level, when they don't even know or care how to attach the RF connectors correctly.  But it worked out with a little rework. On the other hand, Deepelc's F is manufactured as a completely decent well finished product, unlike many cheap devices in China, which are often made like with bare feet on the garage floor.

[spacehen]

I have both a nanovna-f and a sva1015x.

The sva1015x works well from 10MHz to 1.5 GHz.  If i am working in that range it is the one I use.  I especially like the log frequency sweep feature and slightly greater dynamic range.  The interface is much nicer on the Siglent.  It is also faster.

However the nanovna-f works far better at lower frequencies.  The price is much better too.

I've watched some of your youtube videos on the  sva1015x Chris, some great information. My one concern with the sva1015x is the fixed 0dbm output power for the VNA. Is this really something to be concerned about? I assume I can just add an attenuator to the input of the DUT? Just a bit annoying I guess when you want to measure the 1db compression point for an amplifier.



LibreVNA also looks like a neat product

[joeqsmith]

The cost of good step attenuators, bias-Ts, blocks, cables, adapters .....  are all something to consider.   

I did see a post about an improved version of the Lite being available now.  I have no idea how well it actually performs but may be worth looking into.   
https://groups.io/g/liteVNA/topics

[tautech]

Spacehen
Both TG power level and Port 1 Stimulus level are user adjustable from -20 to 0dB.
This info is in the datasheet.

[nctnico]

LibreVNA also looks like a neat product

Why not try that one first? From what I've read so far in this thread and other information, it seems to be the best option in your budget range.

[joeqsmith]

LibreVNA also looks like a neat product

Why not try that one first? From what I've read so far in this thread and other information, it seems to be the best option in your budget range.

Cost is about $600.  Another member had posted about the software not being stable and hanging.  I assume you own one or you would not be recommending it.  Are you seeing similar problems with yours hanging and could maybe provide some additional insight?