Using a Lecher Line To Measure High Frequency


How do you test the oscillator circuit you just made that runs between 200MHz and 380MHz if all you have is a 100MHz oscilloscope, a few multimeters and a DC power supply? One answer is to put away the oscilloscope and use the rest along with a length of wire instead. Form the wire into a Lecher line.

That’s just what I did when I wanted to test my oscillator circuit based around the Mini-Circuits POS-400+ voltage controlled oscillator chip (PDF). I wasn’t going for precision, just verification that the chip works and that my circuit can adjust the frequency. And as you’ll see below, I got a fairly linear graph relating the control voltages to different frequencies.

What follows is a bit about Lecher lines, how I did it, and the results.

What’s A Lecher Line?

Plain Lecher line and with test equipment in use
Plain Lecher line and with test equipment in use
The end with a loop
The end with a loop

A Lecher line consists of two parallel wires or rods that form a balanced transmission line. The technique results in a way to physically measure wavelengths. This method has been around for a long time. Its namesake is Ernst Lecher, a physicist from Austria who perfected the practice in 1888.

The length should be some multiple of the oscillator’s output wavelength. The oscillator’s waves are applied to one end of the Lecher line where the two wires are connected together, forming a loop in my setup. Also in my setup the other end of the line is open, the wires are not connected together there.

A metal bar, or a screwdriver in my case, is put across the width of the two parallel wires, shorting them. As I slide the bar along the wires, it influences the waves. When the bar reaches the wave’s nodes, positions along the wires that are at a half or full wavelength (the zero crossing locations), it can be detected in various ways. One of those ways is to have the bar be two terminals of a neon bulb. The bulb goes out at the nodes, where the voltage is zero. But I didn’t have a neon bulb so I’ll show another way below.

In the photos you can see a measuring tape running the length of the wires for measuring the distance from the end of the Lecher line to the nodes. By doing so, you’re measuring the wavelength of the waves and can use that to calculate the frequency.

The distance between the wires should be even and should be significantly smaller than the length of the wavelength being measured.

My Setup

Since my oscillator can produce between 200MHz and 380MHz, I needed a Lecher line that was long enough to accommodate that range. The formula for converting frequency to wavelength where electromagnetic waves are concerned is:

wavelength = speed_of_light / frequency

which gives:

300,000,000 m/s / 200,000,000 cycles/second = 1.5 m
300,000,000 m/s / 380,000,000 cycles/second = 0.79 m

I made mine 1.5 meters long.

To test the oscillator circuit, I formed a loop at the oscillator’s output that matched the loop at the end of the Lecher line. Part of that loop is a 6.8 Kohm resistor, there so that the oscillator circuit wouldn’t see a dead short. To apply the circuit’s waveform to the Lecher line, I simply put the oscillator’s loop very near to that of the Lecher line. i.e. The oscillator puts electromagnetic waves on the Lecher line using induction. In the first photo above, the oscillator is pulled back a bit to make things clearer.

The next step was to detect when the screwdriver was at a node on the Lecher line. For that I used a high frequency diode, an NTE583 silicon diode whose packaging says “Schottky Switching for High Level UHF/VHF Detection and Pulse Application”. I soldered wires to either end and formed a loop. As shown in the above photos, I added the diode loop to the collection of loops at the end of the Lecher line.

The diode’s two wires go to an analog meter set to the 1-volt scale. When the screwdriver is still less than half a meter from the oscillator end, and at the node for the first half-wavelength, the voltage across the diode is above 0.5 volts. When the screwdriver is between nodes, the voltage is less than 0.1 volts. But the further the nodes are from the oscillator, the lower the voltage is on the meter, until the meter’s needle barely deflects at all.

Making a measurement using the Lecher line
Making a measurement

Results

The node closest to the oscillator end of the Lecher line is a half wavelength. The next one further away is a full wavelength. When the screwdriver is at a location that causes a voltage peak on the analog meter, you’ve found a node and the distance from the oscillator end to that node is recorded.

The oscillator is a voltage controlled oscillator. My circuit includes a voltage divider and potentiometer for adjusting that voltage, which controls the oscillator’s output frequency. The yellow Fluke digital meter is there to show the control voltage. That’s recorded along with the distance measurement.

The oscillator’s control voltage is then turned up, increasing the frequency and decreasing the wavelength. The new distance and control voltage measurements are then recorded.

That’s repeated for the oscillator’s full 0 to 12 volt range of control voltages. The resulting data and graph are shown below.

The frequency is derived from the data using this formula:

frequency = 300,000,000 m/s / (2 * half-wavelength measurement)

The half-wavelength measurement (the 1/2 wavelength column in the table) is multiplied by 2 rather than using the full-wavelength measurement (the 1 wavelength column). That’s because, as was mentioned above, the further the screwdriver was from the oscillator’s end of the Lecher line, the lower the voltage was on the analog meter to the point where there was hardly any deflection at all.

Looking for more high frequency measuring projects here on Hackaday? There’s one using 74-logic for a DIY 100MHz frequency counter, an inexpensive timebase add-on for HP 53131 10MHz frequency counter that [Gerry] built instead of buying the stock $1000 one, and a project that started as a sub-project, the Nanocounter, built with an FPGA, STM32F072 and an Android front-end.

We’re always looking for unique mechanisms and measurements to dive into here on Hackaday. If you have a favorite in your collection of workbench tricks let us know below and we’ll add it to our list for upcoming articles!



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *