EMRFD Message Archive 5833

Message Date From Subject
5833 2011-02-09 22:44:16 Norman Dennett SA vs VNA
Last October the subject of building a SA was discussed at some length on this site and on the Rsgbtech group.
As a result I decided to wait until the W7ZOI PCB and parts were available again from KangaUS. Unfortunatly the kit has not materialised and once again I'm trying to decide what to do.
Quite a few people advised me to go for a VNA. I get the impression that the VNA has superceded the SA and that many people are now using the VNA in preference/instead of, a SA.
I must confess I'm not sure that my basic theory knowledge would stand the strain! The AC theory lectures, some 50 years ago, must have all been on sunny afternoons and although my short term memory got me through the exams it hasn't stood the test of time. The various VNA groups offer considerable support and hopefully my knowledge would "grow" with need and use, my problem is trying to make up my mind in the first place!
There is a tremendous amount of information available about the VNA but I'm afraid that most of it I find very confusing. Could somebody point me in the direction of a straight forward explanation of the VNA and the differences and pros and cons compared with a SA.
I'm sure my comments have shown just how ignorant I am so any information and advice would be most welcome.

5834 2011-02-09 22:47:36 ka7exm Re: SA vs VNA

Perhaps the first question one would want to pose is this: What kind of measurements do you intend to take with either instrument?


The Spectrum Analyzer can be thought of as a power meter which sweeps in frequency. It is architected as a receiver. The display reveals signal power level (the y axis) vs. frequency (the x axis). A tracking generator is a signal source which sweeps the x axis perfectly. With this, you can measure the insertion loss of a filter, for example. With an external coupler, the spectrum analyzer & tracking generator will allow you to see the return loss of an antenna. The most sophisticated display on the spectrum analyzer is an X-Y graph.

The VNA adds phase detection to the spectrum analyzer. Not only can you see the magnitude of the excited power, but you can capture the phase delay of the signal as well.

It is common to measure transmitted power on a VNA. This is by exciting a device from one port and detecting the power on another. The scattering parameter "S21" ("ess-two-one") would be the transmitted power measured at port two when excited from port 1. With an RF coupler, you can measure reflected power on the same port of excitation. "S11," for example. Add some switches and you can measure all 4 combinations if necessary. Amateur VNA's begin with two independent ports. You can always add the couplers and a switch matrix for convenience (if necessary).

Because phase information is now available on the network analyzer, it is common to see results plotted on the screen in many forms. Magnitude and phase may be plotted on X-Y plots, but they may also be plotted in polar form, or directly on the Smith chart. Although somewhat daunting at the start, one can quickly build up a ton of intuition by watching scattering parameters right on the Smith chart.

One additional complication of the network analyzer is that of calibration. It is necessary to remove the loss and phase effects of the cabling to get to your device to be tested. This is usually set up within the analyzer's software, such that you can run the calibration any time you change part of the apparatus, or any time the sensitivity of the instrument drifts enough to warrant re-calibrating. The calibration step sets up measurement reference planes. The final measured values will reveal phase information with respect to that plane and nothing else. If you want to measure a capacitor, you can see the reactance of the capacitor only, and not the phase delay of the 4' cable between the analyzer and the cap.

My main measurement instrument at work is a VNA. This is mainly due to the nature of the type of measurements I take. For home applications, however, I can see how most-everything I would want to do on the bench could be covered with a spectrum analyzer. At least for now.

If you are experimenting with transmitters and filters, I'd suggest the spectrum analyzer. It will reveal a wealth of information for you to absorb. In addition, the spectrum analyzer by itself is an outstanding project. Even if you move on to a VNA in the future, it is always useful to have the utility of a spectrum analyzer on your bench.

Keep checking with Kanga on another SA kitting.

Good luck with your decision and the constructi
5835 2011-02-09 23:31:31 R Wall Re: SA vs VNA
Hi Norman,

Maybe you can have a SA, VNA and TDR in the one box.
I notice that Dave Roberts myVNA software that I use to drive my N2PK VNA, can also be used to drive Scotty’s SA. I guess that also means that Scotty’s SA is now also a VNA.

Tom’s (DG8SAQ) VNWA VNA software also has a SA function. I don’t know how good the VNWA SA function is as I don’t have a VNWA VNA.

Tom’s VNWA software now also has a Time Domain (TDR) function. Note, the VNWA software can also be used to drive the N2PK VNA as a TDR.

Scotty’s SA:

Tom’s VNWA:

I bet you are now even more confused,

Roderick Wall, vk3yc.

5836 2011-02-10 06:48:20 Michael McShan Re: SA vs VNA
How about a SA that fits in an Altoid tin? Take a look at VK2ZAY's latest project:


Not to be confused with a full featured SA, but pretty cool.

Mike N5JKY