Mechanical Filters

Foundations of Amateur Radio

The other day a whole new world opened up to me when I came across the idea that Voltage is the same as Force and Current is the same as Velocity. It all came about when I installed two mechanical filters into my radio. You heard that right, in my shiny solid-state radio, I added moving parts!

The purpose of this was to improve the way the radio ignores unwanted signals and as a result has an easier time hearing what it is you really care about. The radio already has filters built-in, but mechanical filters offer a cleaner output with less distortion across a wider range of temperatures. Another way is to say that - mechanical filters have a much higher Q.

Think of a pendulum swinging through oil, it's losing lots of energy for every swing and has a low Q. The same pendulum swinging through air retains most of its energy and has a high Q. The same is true for mechanical filters, less energy loss, better reproduction, better outcome for the things you want to keep and hear.

So how does this then work?

Turns out that our electrical theory with inductors, capacitors and resistors have mechanical equivalents, specifically mass, stiffness and damping. As I said when I started, Voltage is Force and Current is Velocity.

It turns out that all the maths we use to design electrical filters can also be used to design mechanical filters and 1946 Robert Adler from Zenith did exactly that. This worked so well that in 1952 the Collins Radio Company started manufacturing them and today we still use them in many different radios.

As an aside, you might be surprised to learn that the first filter that Robert Adler invented in 1946 was for a 455 kHz filter, which I could technically still use in my radio today, since the same Intermediate Frequency or IF is used.

The mechanical filter - vibrating bits of metal - resonate with specific frequencies, much like a tuning fork does, but your radio deals with electrons, not movement, so the electrical signal is first converted into movement by a piezoelectric transducer, a piece of material that distorts when you apply an electrical field and when you use it in reverse, distortion creates an electrical field.

So, you have a box with a wire at one end and a wire at the other and in between are two transducers and a bunch of mechanical resonators, much like a string of pearls on a necklace.

I mentioned earlier that mechanical filters have a much higher Q. An electrical Q might range between 100 and 500, the mechanical Q in 1946 using steel was several thousand and in today's filters using Nickel-Iron alloys, a Q of 10,000 to 25,000 can be achieved.

Without going into the maths, what is this Q really describing, other than the pendulum in oil and mechanical losses?

One way to explain Q is to say that it describes the "goodness" of a resonant circuit, the higher the Q, the better the circuit. In our case, "goodness" means that it resonates better where we want it to and not where we don't want it to.

Before you start wondering, why the letter "Q"? Turns out all the others were taken when K.S. Johnson was looking for a letter to describe the attributes of coils in 1920. Today we think of "Q" as Quality, but that's the cart before the horse.

Anyway, back to Amateur Radio. If you look at a theoretical filter, you'll see a lovely curve that lets through the bits you care about and ignores the bits you don't like, but when you then start looking at the real world where damping and resistance come into play, you'll soon learn that there are all manner of ugly spikes on this lovely curve.

A typical tuned electrical circuit will have artefacts, or distortions along the way, dipping down, instead of staying straight, or having an ugly peak when it should be smooth. This in turn results in something that you can hear, distorted audio where low frequencies are under or over represented and strange distortions occurring along the audio range.

It also means that adjacent signals, the ones you're trying to ignore get caught up in this same distortion and you'll hear some of those when you don't want them.

So when I say that a mechanical filter offers a higher "Q", it means a whole lot less distortion and a better representation of what's going on. As an aside, in transmit, a mechanical filter will also help contain the energy coming from your signal and transmit it where it's needed, rather than waste it where it's not helpful. As a QRP station, every little bit helps.

I just love this hobby, every turn is another surprise.

I'm Onno VK6FLAB