The NanoVNA, Viewer Questions and Answers

In this video I cover several of viewer's technical questions. Several topics are covered including TDR, Beatty Standards, T-Check, full 2-port measurements, transfer relays and more ....
Rohde Schwarz published a paper on the T-Check:
scdn.rohde-sch...
Link to my eevblog posts:
www.eevblog.co...
Groups.io:
groups.io/g/na...
Source for the NanoVNA:
github.com/ttr...

[@DavidLopez-bz4rj]

You are the NanoVNA man! Forget those uneducated people who want all for nothing, keep doing what you enjoy, and lucky for us who learn new things. Thank you!

[@rodfarlee6157]

For those of us new to VNAs, this is insightful and informative in so many ways. Greatly appreciated!

[@ccwynn6657]

nanoVNA = $50. Instructions and Insight by Joe Smith = $50,000

[@cthoadmin7458]

Wow Joe you have a lot to answer for! Never realised Labview was so powerful or that vna theory and measurement was so interesting. Like many hams I bought the nanovna as a cheap SWR analyser, but as you have shown, it can do so much more. This has turned into a learning engine for me. Thanks for taking the time to produce these high quality and inspiring videos. Going to see if I can turn Labview into a spectrum analyser with one of those cheap eBay spectrum analyser boards. So many rabbit holes to go down, so little time...

[@johnwest7993]

This is a cornucopia of years of network analyzer experience distilled into a 51 minute video. I've watched it twice. Will probably watch parts of it several times.Thank you.

[@joesmith-je3tq]

Thanks. I made a few videos for these low cost VNAs. The software has changed a fair bit since this video was made.

[@James_Bowie]

Excellent presentations! I value all of the work that you put into this video.

[@fredlodden1538]

Exceptional. Very professional. Give something back? You gave heaps! Very informative video. To think I purchased this thing hoping to do some SWR measurements. And in fact it opens the door to a whole new universe. I haven't touched on some of these things since final year EE degree some 40 years ago. Awesome.

[@joesmith-je3tq]

Thanks for the kind words. I'm in the same boat. The math has been a good refresher.

[@barrykery1175]

Great video, very informative. I liked the idea of the SMA male to female adapters. Ordered same as soon as I saw your video. I actually ordered extra to hand out to my friends that have the nanoVNA. My nanoVNA is on order.....and now I wait ! Barry, KU3X

[@googacct]

I only just recently found out about the nanovna and have one on order. Thanks for posting the the very detailed videos about it. It has given me an idea of what I can expect out of it for performance. I am looking forward to experimenting with it.

[@AndreasSpiess]

Hi Joe, I am the Andreas you mentioned in your other video. You do a great job here and I learn a lot from you. I understand you do not want to go into PCB design. Nevertheless, it would be very valuable if you could give us some hints on how to create such PCBs with 50 Ohms, especially the dimensions. Or even a video about strip lines? Your design is very good for experimenting because it has ground on the same side as the signal. I found many calculators with only the signal on one side, which is not so handy for experimenting.

[@joesmith-je3tq]

Hello Mr Spiess. I'm sorry if I butchered your name. Your antenna tuning video was very easy for me to follow along and enjoyed watching it. I can see why this person asked to see how the Nano would compare with what you had shown. It made for a good example on what people can use this low cost analyzer for.

[@toastedcrumpets]

@@joesmith-je3tq Hi, I too would like to know if you can share your test pcb design (or at least the specifications)? I'm not sure how many iterations getting a nice 50ohm microstrip line out of JLPCB would take but any advice would be appreciated!

[@uploadJ]

Hey - TI GaAs Facility alumnus here ... that department was sold to TriQuint some time back. Worked for Dr. John Bealle under Tom Cordner at the North Building in those days ... did DC parametric test engineering on wafers before backside grind, etch and gold plating. Also did some Noise Figure software for the RF on-wafer test group.

[@joesmith-je3tq]

Looks like you found the video. I dug out my old transfer relay that I mentioned during the video but have not yet tried it out with the Nano to see what sort of gains I get.

[@uploadJ]

Yup. Found it! Sorry for all the comments I make, I like to 'talk' (also a ham. Doing 160 meters successfully from a small city lot the last few years. Doing 160 m WSPR right now as a matter of fact. WSPRnet.org click on Map, choose 160m band etc) These are the records we get back from reporting stations - Record elements are: Timestamp Call MHz SNR Drift Grid Pwr Reporter RGrid km az Data (just in): 2019-09-14 22:34 AA5CT 1.838021 -17 0 EM13qc 20 K9AN EN50wc 1086 42

[@NelsonSilvaFilho]

Hello Joe Very nice videos. My NANO VNA arrived yesterday and I am learning a lot with you. Thank you! BTW What is the software (interface to the NANO VNA) you are using in this video? It looks great... Any place to download it?

[@jspencerg]

Resolved to rewatch videos now that I have nanovna to do some of the experiments. -->help avoid ignorant questions.

[@joesmith-je3tq]

Looking forward to it.

[@jspencerg]

@@joesmith-je3tq Haha

[@pigrew]

Nice video, I like the calibration coupon. According to Keysight, SMA should be 5 in*lb. 8 in*lb is for 3.5mm. (I have damaged connectors before, pulling the nut off of the male connector. However I'm this case, I'm more worried about cracking the PCB tracks.) Another note is that the applied torque depends on where you hold the wrench. Often there is a line about 1cm from the end of the handle marking where to apply force. Picking a few more nits: when using a single cable, put the cable on port two. The reflected signal is attenuated round trip (twice) in the port 1 cable, but only once in the port 2 cable, so it makes sense to optimize port 1.

[@joesmith-je3tq]

Amphenol's basic specs may be found here: www.amphenolrf.com/connectors/sma.html Looking at the error of the torque wrench shown, using a knife blade and going from one end of the handle to the other, the variance is about 1" lb. Still well within spec no matter where you hold it. If you really choked up on the wrench (using the chromed area) you may get into trouble. I take it yours changes a fair amount more. I don't have another click or break over type to compare it with. I have a CDI fixed torque screwdriver that I use an opened sidewall deep well socket with. This has a few advantages. You can get into very tight areas with it and hand placement is basically a non issue. Just a matter of using the right tool. Personally, I wouldn't use a VNA without both cables attached for the reasons I mentioned. I doubt with the Nano you would see much of a measurable difference using a cable on one side or the other. You would need to run some experiments and prove it to me.

[@DM-fz3ly]

Would you please consider doing a video explaining the pros/cons and especially the PITFALLS for the typical amateur wanting to use the nanovna to test various parts of an antenna system. How the Nano might compare to other types of amateur test equipment. How to properly calibrate and how to eliminate error and how the various lengths of coax might need to be considered and calibrated out or taken into consideration when trying to test and measure the effectiveness of an antenna system. Thanks again for sharing

[@joesmith-je3tq]

I doubt I will make another video about antennas beyond the first one.

[@DM-fz3ly]

@@joesmith-je3tq No problem Joe, Thanks for sharing. Quick Q if you don't mind here (or I can take it over to the eevblog forum, just let me know) I did a quick check of the relative output power of the NanoVNA, MFJ 269 and YouKits FG-01 antenna analyzers. At 14MHz the NanoVNA puts out -13 dBm, MFJ +6 dBm and the YouKits +15 dBm as measured by my Rigol spectrum analyzer. Is the low output power of the NanoVNA enough to accurately measure say an antenna at then end of 100' of LMR 400.

[@joesmith-je3tq]

@@DM-fz3ly Not being any sort of radio enthusiast, I would need to start by looking at the datasheets for the coax you mention. Still though, it seems like if a person was wanting to measure antennas, they would have a chamber, turntable, receiver, calibrated antennas....... not a $50 pocket VNA. If you just wanted to measure the resonance, maybe. If you were wanting to measure the SWR of just the antenna, you would need to take into account all the coax. Really would depend what you were trying to do. That said, I am still wondering about using the second channel to measure the reflection. I have not tried this yet but hope to set up an experiment in the next few weeks.

[@d3xdrive]

Some additional advice to learn from (bitter) personal experience, it can also be important to have a frequency sweep that starts near enough DC. While the frequency range of nanoVNA is 50k - 900m; I have an Agilent E7495A which is 325m - 2.9g. This lower bound of 325 MHz is incredibly limiting! (more...) If the cable is of low enough quality or it has a low pass characteristic, then all of this energy will be lost anyway. In the security video world, you might see baluns on the ends of twisted pair. Or just really crappy fake coax-type cables. In both cases, my entire frequency span (325 to 2.9 GHz) was useless. All if it was far too high in frequency to make it down and back. I've ordered a nanoVNA because of your videos, thanks.

[@joesmith-je3tq]

For the case of the TDR measurement, the low frequency is very important for deriving the DC part. For normal work, I have an old 3589A that can perform S11 and S21 with a lower end of 10Hz. A very useful unit as I do a fair amount of work below 100MHz.

[@lens1736]

I see you are at 9.99k subscribers, hopefully with my subscription today, you reached the 10k milestone. Oh ya, the other reason i subscribed is you are very knowledgeable and a good clear teacher and i just purchased a NanoVNA

[@joesmith-je3tq]

Thanks for the kind words.

[@BarefootBeekeeper]

Really useful, thanks.

[@messageobliquespe100]

Thank you - another 25mins of house chores completed.

[@joesmith-je3tq]

lol. You're very welcome.

[@JimPBarber]

Where can I get that software??? I have seen that software on several videos, but cannot find it online.

[@uploadJ]

BACK IN THE DAY ... out at the Renner Antenna Range (Texas Instruments days) we had a few hundred foot length of RG214 used as 'delay' to extend the reference plane out to a 'fixture' located on a pedestal on one of the small ranges out behind the main building ... this would have been HP 8410 class gear in 1975 ... The HP 8410 being HP's first full-featured uWave VNA.

[@joesmith-je3tq]

If you watched that first video for the Nano, you saw my vintage VNA. The manuals show 1978 for that one.

[@uploadJ]

We were doing MLS (Microwave Landing System) work at that time, the antenna was a 64-feedhorm array 'scanned' 4 feedhorns at a time back and forth at a 5 GHz frequency. It was a pretty good sized assembly ... about 25 ft wide and 15 ft tall. Testing involved an HP 9830 programmable 'computer" that talked HPIB at the time to test instrumentation ... fun days ...

[@W8ASA]

Thanks for posting this. Many people are purchasing this VNA only to use as an HF or VHF antenna analyzer. There is so much more that this can do. Luckily, I have some new high-quality transfer switches so I'm going to try that out. You mentioned that your home-made transfer switch didn't have much isolation. What was the component you used spec'ed at for isolation? Ken W8ASA

[@joesmith-je3tq]

Hello Ken, Its an Omron G6K RF.

[@avejst]

Fantastic walkthrou Thanks for sharing :-)

[@joesmith-je3tq]

No problem. I'm glad you enjoyed it.

[@BrendaEM]

Good video. I saw a photo of a Lecroy Wavepulser, which seemed to have some relays on the front end, which are likely the transfer relays you mentioned. Thanks. You might want to chuck out that foam. As it breaks down, it might offer chemicals that can corrode the connectors.

[@joesmith-je3tq]

Yeah, that foam is in rough shape but it's not sticky goo like I've seen. The transfer relays I mentioned can be seen here: www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg2701350/#msg2701350 You may be able to find something NOS. The two that I made are just not good enough but that one shown in the link would work nice.

[@BrendaEM]

@@joesmith-je3tq , I hope that at some point, someone makes a NanoVNA just a little different so that it can do TDR on circuit-board traces. The processing might have to be done on a host computer. Perhaps a NanoVna board could be made with your transfer relays integrated.

[@joesmith-je3tq]

@@BrendaEM I suspect I am not understanding what you are suggesting. Even the original Nano can provide some level of TDR. I've demonstrated it in these videos and posted some results in the forums. The V2+4 should do an even better job of it. The problem isn't so much of seeing discontinuities, its getting a test fixture good enough that minimizes them. lol. If we had a transfer relay, we may also want bias T's. Then we will want step attenuators. Then we will want 4 receivers. ...... Of course we can get everything today, but not for the $120 the V2+4 is selling for. Then again, I am amazed of what they are offering in this price range. I'm very impressed with that original Nano. Looking forward to running experiments with the latest hardware. Software first....

[@NiHaoMike64]

It would be nice if there's a version that covers the 2.4GHz and 5GHz ISM bands.

[@joesmith-je3tq]

I would like to replace my vintage HP at some point with something newer, smaller, faster, more dynamic range, more data points..... I just don't use it enough to justify the cost. Sure, a pocket 5GHz VNA for $100, I would be all over that deal!!! lol

[@TheRetiredtech]

Very impressive but without access to the software not all that useful. The video I mean. I understand your hesitation to release the software but without it most of this instruction is useless. I have ordered this unit and also have the EU1KY unit. But am lacking your expertise. I did use HP analyzers back when I worked. I still enjoy seeing your knowledge being demonstrated.

[@uploadJ]

Maybe even release just a VI or two - that would make slapping together a LabVIEW app a little easier than starting from scratch (for those of us who CAN program in LabVIEW).

[@pentachronic]

Would this trial version of LabView for 45days be compatible ? store.digilentinc.com/labview-home-bundle/

[@wadehsu2347]

One thing to be aware of with the torque on the SMA connector. Brass SMA connector are rated at lower torque vs the stainless steel one. If it’s gold plated, most likely it’s brass which only requires 3-5 in-lb instead of 7-10 in-lb for stainless connector. So with that Pasternack wrench you are probably stressing the connector.

[@joesmith-je3tq]

Gold plated stainless are fairly common but I appreciate your concern for my $50 Nano. For others looking for details on torque, I suggest: www.microwaves101.com/encyclopedias/connector-torque

[@NodeEntry]

Wow. I learned a lot. Thanks, Joe. How long did it take you to write your own Windows application for this VNA? When did you start writing your own applications? How long did it take you to get proficient? What resources did you use to learn?

[@timothytapio4252]

Joe, I have both SAA-2N and V2_Plus 4, can load your software for each on same machine (not to run at same time)? Thanks for your time, I have not finished reading all 178 pages of your manual, working on it. K4SHF

[@joesmith-je3tq]

I have ran two instances of my software. There are a few tricks in general to do this depending on your OS. Google search running multiple instances and start reading.

[@timothytapio4252]

@@joesmith-je3tq Thanks!

[@MeineVideokasetten]

G R E A T video. 👍🏻

[@DM-fz3ly]

Thank you for sharing your knowledge! It is a treat to watch and learn from your technique and how to think through various tests and results. A ham radio antenna question. If I do a proper calibrate with standards attached directly to the nanovna and then want to measure an antenna connected to a length of, say 50-100 feet of LMR 400, will I be able to evaluate the antenna or do I need to take the calibration standards to the end of the coax and "zero" out the coax?? Thanks for your time in sharing your knowledge 73 N7KBC

[@joesmith-je3tq]

I guess that will depend what your goals are for your particular test. As I stated in that first video on antennas, there is a lot to testing an antenna. I would start by doing some research. That book I referenced would be a good start.

[@Gridl6]

Very impressive video. I am worried when testing 160m antennas will local AM BCST station over power the input. If I have this hooked to the coax in my shack that goes out to my 160m 3/8 wave length Inverted L, how susceptible do you think this might be to local AM BCST stations throwing off the results. For example say running say 5000 watts on 540 Khz 10 miles from me.

[@joesmith-je3tq]

I don't know what tests you are running or how you are conducting them but from this video, it seems like if you stay below 0dBm, you should be fine. That's my swag anyway based on the tests I ran showing the 1dB compression point.

[@uploadJ]

Looks like you're to the point where you need a firmware re-write in the STM microprocessor to simply act as interface to your LabVIEW code ... you're probably way ahead of me on this too idea-wise ... Someone will do this too, in the not too distance future.

[@joesmith-je3tq]

I would be happy to just run the unit headless but I imagine there are people who want it for a standalone portable unit.

[@itdevops]

hi, & ty for your video. i want say which cable ref. you use for your cables ? if i use a SO-239 adaptator you think that's take effect ? on result calibration ect... ? ty

[@aimeiz7637]

Great video Thank You. Where could I download this wonderful VNA software from?

[@joesmith-je3tq]

Have you looked at the free open source software available? If not, I suggest you have a look. I suspect it is much nicer than mine.

[@kellypainter7625]

He used Labview which would require you to have your own license to use it. Labview is very expensive and not all that great IMHO - I avoid it like the plague. He is using Labview quite effectively and has good info in this video but not so hot in the SW dept. due to license issues. I use this: github.com/mihtjel/nanovna-saver It isn't too bad to set up (100% python other than using Qt and some free libraries) and it does implement the segmented sweep feature. It is actively being developed so check it out. edit: I forgot that they have binaries available so you wouldn't have to set anything up. I didn't use them though.

[@joesmith-je3tq]

@@kellypainter7625 I just noticed your post had been blocked because of the link. I often wonder why people feel the need to post on subjects they are woefully ignorant about. Yes, the project was created in LabView. That does not mean that it requires a license to use it as you suggest. The software would have an installer and run like any other software for a PC. NI is planning to release LabView for free in the future for hobby/educational use. As I understand it, these tools will not be crippled and will include the application builder. There have been water marked versions of LabView available for several years that will cost about $50. The reason I use these tools is for the time I save in the development cycle.

[@ramjet4025]

Joe, Which model of the Relay, G6K-2F by OMRON, ?RF-S-DC3 RF relay DPDT 1A 3V ROHS?

[@joesmith-je3tq]

G6KU2FRFTR03DC5

[@ramjet4025]

Joe, Your calibration boards are essential tools. I have searched and cannot find them. Have you a link to anyone who sells these? If I build this, Can you please advise the type of PCB and the size of the tracks? I can make it if I know what sort of PCB it is and the size of the tracks. Is it double sided PCB? Can it be any sort of Double sided PCB?

[@joesmith-je3tq]

kzbin.info/www/bejne/fXuxfqGnbcp3Zqc

[@willernst8376]

From what I have read about these units, they have a si5351 chip that can only produce frequencies up to 200mhz . So how is it that it can go up to 900mhz ? Are they mixing or doubling or what? It is bothering me that i can't figure out how it's done. I'll just have to buy one

[@joesmith-je3tq]

Fundamental to 300, using harmonics for higher frequencies. I've shown some data with the 1.5GHz firmware installed. You can find that on the EEVBLOG forum. At $50 it's a no brainer for a learning tool.

[@willernst8376]

@@joesmith-je3tq thanks for your reply, I just found your channel and it is great. I figured they were using some kind of upper harmonic, but I thought to get any use out of that you had to heavily filter so as to not get garbage out of the mixers. I'm relatively new to this so I'll keep studying. Thanks for the reply

[@thomaskallmyr]

😅interesting😮

[@mrtablet4662]

Whats the name/version of the software please?

[@Boediprasetya]

Master nanoVNA...what software do you use 20:47

[@joesmith-je3tq]

It's all custom written.

[@gbennett58]

I can't find a similar torque wrench. Link to supplier?

[@gbennett58]

Nevermind, found it at Fairview Microwave. Only $233.78. Type N, SC Fixed Torque Wrench in Break-Over Type 13/16 inch Bit that is Pre-set to 8 in-lbs

[@felixcat4346]

If the effective length to check a cable is approx 20 feet, I don't think its really going to cut the mustard.

[@joesmith-je3tq]

Sorry, but as usual your comment was lost on me. What is effective length and where is the 20 feet coming from and why do you feel it won't cut the mustard?

[@joesmith-je3tq]

Maybe you didn't understand that the resolution was 5" and the length is based on the samples, which I can program to pretty much anything within reason. The problem is the unit is just so slow to pull the data out. Maybe this helps you understand. That or I still have no clue what you're going on about.

[@ChatGPT1111]

joe smith he must be talking about classic French’s or perhaps the high viscosity Grey Poupon lol.

[@scharkalvin]

The soldering on some of these Chinese clones might not be up to good practice. So, you could snap the connectors off the board, even with proper use of that torque wrench, if the connector was being held on by cold solder joints with not enough solder. Worse, you might also rip some of the trace off the PCB in the process.

[@joesmith-je3tq]

A lot of things could happen indeed. Obviously they have not and I continue to torque the connectors. If I destroy the unit, its of no great loss.

[@scharkalvin]

@@joesmith-je3tq Well, you could flow some more solder onto the connectors and make sure it's REALLY attached! (And while you're in the neighborhood make sure the shields are really grounded).

[@joesmith-je3tq]

@@scharkalvin Don't worry, it's fine.

[@uploadJ]

re: "So, you could snap the connectors off the board, even with proper use of that torque wrench," >> At 50 bucks a copy (and going forward where these things get cheaper) gear of this type or caliber is a "consumable".

[@Stover159]

may i ask what you do for a living? are you a ham operator?

[@joesmith-je3tq]

I had an amateur license when I was a kid but let it lapse as I wasn't doing thing with it.

[@nlo114]

If it's only $50, buy 2 then couple them together for S21 and S22 capability.

[@joesmith-je3tq]

I would imagine there are people out there who actually believe this.

[@muppetpaster]

10:30 It does not seem really fair to compare the HP to the Nano.........

[@joesmith-je3tq]

And why not? The basics are the same which is all we are talking about here.

[@kb2cxj]

nice video very informative thank you I downloaded and installed labview no clue hoe to use it do you have a project file or something I can open in labview and work ? the software from the distributer is a joke don't even work right

[@joesmith-je3tq]

The best advice I can offer you is to go through the getting started manual. Work the examples. Like anything, it will take time to learn.

[@kb2cxj]

@@joesmith-je3tq labview is going to take longer to learn than the VNA lol thanks Joe

[@joesmith-je3tq]

@@kb2cxj That may depend on how much you want to learn about VNAs and your level of education. There's a little bit of math involved if you want to dive in.

[@felixcat4346]

Are you trying to hypnotize with your jumpy hands?

[@JxH]

So many people give advice about protecting the instrument, but they're seemingly oblivious that IT'S $50 !!!! Crikey, relax folks. Yes, it's a VNA, but it's not $50,000. I really laugh when they give the inevitable lecture about not spinning the connector when screwing it on. Yeah sure, here's the one-piece SMA Calibration Short - go ahead, show me.

[@joesmith-je3tq]

The idea is to develop good habits before using a $500,000 VNA. BTW, all the shorts I received with these low cost VNAs are two part. There's just no way to lock the center in place when torquing the nut.

[@user-iu3rt4zt3u]

Нихрена не понял, но интересно.