ARS Explores the AIM4170 Antenna Analyzer,  Part 1

 

 

By Russ Carpenter, AA7QU

Co-founder, The Adventure Radio Society

 

Have you noticed that everybody and his brother is producing an "antenna analyzer?" For a long time, these devices seemed to be an MFJ monopoly. Now, there are a mess of them.

 

Take a look at this eHam page, in which 31 of these machines are reviewed: http://www.eham.net/reviews/products/31  Oddly, the most highly regarded box of all, the N2PK vector network analyzer isn't there. (See http://www.n2pk.com/) Maybe that's because it actually requires some home-brew surface mount construction, which appeals to about two and a half hams these days. But there are plenty of interesting alternatives listed on the eHam page. The off-the-shelf analyzer that consistently gets rave reviews is the AIM4170.

 

The AIM4170 has been thoroughly reviewed by QST, NCJ, and Rudy Severns, N7LF. These reviews can be accessed through the home page of Array Solutions (http://arraysolutions.com/Products/AIM4170.htm#top%20of%20page), and I won't repeat their substance here. Instead, our mission is to play with this box outside the antenna realm, especially in tuning a band pass filter and characterizing a crystal.

 

First, we should be clear that the AIM4170 is one half of a fine solution. It does a nice job of looking at reflections. Doing the whole job sometimes requires the ability to look at what comes out the other end of a device under test. But only those boxes or lashups with a "transmission" capability can do that. You'll see what I mean in a moment.

 

Within its limitations, the AIM4170 presents a joyful experience. It's a three-part thing: hardware, software, and humanware. The hardware is nicely packaged (although a built-in USB interface would be an improvement) and the software is perfectly serviceable (although not elegant). What really sets this machine aparts is good old human stuff. It is supported by an active, intelligent user group and, most importantly, by the earnest and timely participation of the inventor himself, W5BIG.

 

Let's get down to business. Here is what the front end of the box likes look.

 

 

 

 

We start with a double tuned bandpass filter. The purpose of this filter is to provide simple front-end filtering for the 14 MHz band. Good explanatory material appears on page 3.10 of Experimental Methods of RF Design ("EMRFD").

 

 

 

 

 

There are lots of ways to tune up one of these animals. The age-old method is to build the rest of the receiver, connect an antenna, and twiddle the trimmer capacitors for maximum noise. This sort of works, but it requires you to finish the darn receiver project. In addition, max noise may result in a filter whose band pass characteristics are, in part, outside the 14 MHz band, or are too narrow or wide.

 

Another way is to terminate one end of the filter in 50 ohms and connect an old timey antenna analyzer to the other end. Then you tune for minimum something-or-other at the center of the band (something-or-other could include SWR, reflection coefficient, or return loss). You can also set the analyzer to the edges of the band to see what's going on there. But this is very tedious and real people, being impatient, tend to maximize performance at the center and call it done. This method almost guarantees the same errors you get with the max noise approach.

 

The problem with the old fashioned analyzers is they look at one frequency at a time; they don't scan. However, more downtown scanners, like the AIM4170, do. Scanning allows you to really nail the adustment of the filter. Here is our test filter adjusted for a pretty slick in-band reflection coefficient. The red line is the reflection coefficient and the purple line is the "theta," or phase angle. In this test, the red line is the main thing.

 

 

 

Pretty nice, eh? You could move on to the rest of your receiver at this moment. But, there is actually more to learn about this filter. Trouble is, you can't learn it with a device limited to reflections. What the world needs is a box that's just as reasonably priced and well implemented as the AIM4170, able to handle both reflections and transmissions. One of these days, it will exist. With the possible exception of the TAPR/Tectec VNA, it doesn't now. More on this in Part 2 of this article.

 

To whet your appetite, here is an image from my HP spectrum analyzer acting as a scalar network analyzer. Here, of course, we are monitoring what's coming out of the far end of our famous filter. It shows that the AIM4170 has allowed us to do a brilliant job aligning the filter. And it also shows us the famously asymmetrical shape of a double tuned circuit response. With a little more work, we can also document the insertion loss of the filter.

 

 

 

 

Now we move on to a capability of the AIM4170 that could change the world (well, almost). First, consider this propositiion:

 

    1. There are two main reasons why home-brewing in amateur radio has dropped to such embarassing levels:

 

       A. invisible parts, and

 

       B. crystal filters.

 

As to A, we can use visible parts, even if they are technically obsolete. As to B, the AIM4170 rides to the rescue.

 

I suppose that B needs a word or two of explanation. Assuming that you want to play with single signal receivers (unless you are a DSP genius, or an analog phasing devotee), you have to cope with crystal filters. That's a tall order, because characterizing your basic 75 cent computer crystal is prickly. (Once crystals have been characterized, it's actually easy to design a filter.)

 

Here is a test thingamabob I built to find the "motional parameters" and Q of crystals, based on the material in pp. 3.19 and 7.38 of EMRFD. It works, but requires a lot of fire power (like a frequency counter, signal generator, and spectrum analyzer).

 

 

 

 

There has to be a better way. And there is! The AIM4170 makes it oh so easy. To show you what I mean, I've chosen these three cheapo crystals. Two are ECS: 9.216 Mhz (HC-49U) and 9 Mhz (HC-49US). The third is the Epson 9.83 MHz in the teeny CA-301 package. They are all sold by Digi-Key. The Epson is wonderful; high Q and narrow frequency tolerance. The only way to fly.

 

 

To save space, I present only one of the AIM4170 scans here: it's the Epson, and you're looking at the series resonance.

 

 

After you've run this scan, you just make another menu choice, and this is what you get:

 

 

 

 

 

Brilliant! I can't resist showing you the Qs for the three crystals we tested:

 

Epson 9.83 MHz   278, 734

ECS 9.216 MHz       96,551

ECS 9 MHz              78,108

 

The crystal parameters are all you need to make a crystal design program happy. We'll do that, as well as some other gripping things, in Part 2 of this article. Stand by.

 

*   *   *   *   *   *   *   *   *   *

 

Russ Carpenter, AA7QU, is a co-founder of the Adventure Radio Society living on the McKenzie River in Oregon.

 

 

 Create your own free wiki!Support CommunityPublic, Silver wiki Privacy Policy RSS