“amplifiers oscillate, and oscillators amplify”

The statement describes the fact that, when attempting to design one, there are many pitfalls that cause you to have the other happen.  Usually when designing an amplifier oscillations are your enemy, and you tune them out with bypass capacitors and work to ensure the stability criteria is met.  When designing an oscillator, on the other hand, the circuit can readily swing into a stable state if you don’t have a very good design, and end up as just an amplifier.  Granted, with proper design and construction this is avoidable, but when you are building one for the first time or just quickly building one, this can happen.

I propose a similar saying for antenna design, although to make this work, i am going to use “radiator” instead of “antenna”:

“Radiators couple, and couplers radiate”

This only applies, really, to antennas with more than one port, and really can apply to any microwave circuit.  But this way the phrasing works out much better.  But basically says that usually when designing an antenna, you are battling this coupling between ports, trying to get efficient radiation.  Yet on the other hand, when designing a microwave circuit, one of your enemies becomes radiation, and you work to keep that to an absolute minimum, because unwanted radiation is unwanted loss.

Below is a function I used to plot smith chart results in matlab.  I used basic plotting code to generate the chart itself, and added a simply plot function to add the impedance locus and constant VSWR circles.  Hope this is helpeful!  Please let me know if anyone finds a bug.  I will note that there is no safeguard in the function for s-parameters that are greater than 1, which if you are plotting the active S-parameters of a multiport device, are possible.  This isn’t a big deal when the S-parameter goes slightly above 1, but if it swings well above 1 you end up with a tiny smith chart and these erratic line segments…it’s a mess.  For passive applications I haven’t come across a bug yet.

smithchart.m

If you are interested in taking a look at what I spent the last two years working on…you can find it  here.

The thesis was on a project aiming to design GPS antennas that fit into a volume of approximately 1.25X1.25X0.2″.

So I decided that, being able to compute all the data I need from S-parameters alone (as HFSS does), I exported the entire S-matrix to MATLAB and took a look at the impedance values I calculated.

HFSS gives the equation for finding the Z matrix as :

Z = sqrt(Zo*I)*inv(I-S)*(I+S)*sqrt(Zo*I)

where I=identity matrix, and S is the full S-parameter matrix

from this I plotted the exact same thing HFSS was giving for the impedances.  So I started plotting different chunks of the equation (e.g. inv(I-s) alone, (I+S) alone…) and found that the inv(I-S) term was spiking at the exact places my impedance was spiking, while the other terms were well-behaved.  Well, looks like that inversion is messing things up…indicating a singular, or more correctly a nearly-singular matrix (singular matrices have NO inverse defined).  Sure enough, using the function:

cond()

in MATLAB revealed huge condition numbers for the matrix (I-S) where the impedance spikes occurred, 2 orders of magnitude larger than the condition numbers calculated over the rest of the frequency sweep.  So I need to invert a matrix that is nearly-singular….now what??  I need single-value-decomposition (SVD) to take out those values that make the matrix singular and do a quasi-inversion, basically approximate the inversion.  MATLAB includes just such a function:

pinv() — where it’s use is  pinv(matrix, tolerance)

The tolerance can be set to an arbitrary value.  For my purposes, I can stand to lose some accuracy, so I chose a relatively large tolerance (around 0.02 or so) to take out those spikes but preserve the overal shape of the impedance waveforms.  Usually you will want to keep this very small to keep the inversion close to the true value.  But this solves my problem.

But it leads to another conclusion — HFSS, apparently, doesn’t test if its S-parameter matrices are nearly-singular (or at least I haven’t seen a way to have it do so).  Certainly it is using all sorts of inversion approximations in the solver the software uses (you can even do iterative matrix solutions in v11 of HFSS…), I don’t think it is likely that they have skimped on such checking on the data output front end.   Please post comments if you konw of any solutions in HFSS for this.

 http://www.falstad.com/antenna/

You can readily see the field pattern distortions that occur from things like making a dipole or loop longer than half a wavelength in size, the multiple lobes that form, and how the current distribution forms on the structure.  You can for from a triangular current distribution on the dipole, to 10 or 20 sine wave variations along the length.  Really easy way to see what happens to the pattern.  There are also some array examples of end-fire arrays and schelkunoff polynomial alrrays.

In celebration, watch this:

I found this xkcd comic strangely appropriate — hopefully I am not subconsciously doing something like this during my presentation:

I guess there aren’t imaginary antennas…

It was found that a coaxial resonantor could be formed by making a center conductor that is adjusted to tune the frequency. As the article states, many problems were overcome in order to make them as successful as they were, and the article estimates tens of thousands of radio transmitter sites around the world equipped with these bands of beer kegs transformed into resonators!

Maxwell taught emag to Jeans (at Cambridge),

Jeans taught emag to Smythe (at Princeton)

Smythe taught emag to Rusch (at CalTech)

Rusch taught emag to Strangeway (at University of Southern California)

Strangeway taught emag to me (at Milwaukee School of Engineering) – that’s 5th generation.

Hopefully someday I can add to the list:

Holland taught emag to _________ (at ___________) …

Now I need to find a big ground plane so I can measure it and see how it compares to my HFSS simulations.  Currently it is a 31X31X3mm susbtrate with a single patch on it that resonates at 1.5GHz, to provide a benchmark for my simulations.  I plan on building a number of these to get some experimental verification of some trends, and potentially piece together one of the stacked patch configurations that comes within 4mm of meeting my original design goals.  Assuming this goes well, I will be defending my thesis come early April…