Experiment 1 Frequency Spectrum and the Spectrum Analyzer

1.1 Objective
1- To demonstrate the frequency spectrum approach to signal analysis. 2- To introduce the spectrum analyzer as a test tool.

1.2 Basic Information

As signals and systems grow in complexity, their time domain representation becomes inadequate. The frequency domain representation forms a simple and straight forward alternative. Signals are decomposed into the sum of a number of sinusoidal signals. This makes analysis of such signals and systems much easier. The group of sinusoids representing any signal is called its spectral components. If the given signal g(t) is periodic with period To, the Fourier series expansion can be used to find the spectral components of the signal:

g (t ) a0 a1 cos 0 t b1 sin 0 t a2 cos 2 0 t b2 sin 2 0 t ...

where

g (t ) cos n T T
o
o

tdt , bn

g (t ) sin n T T
o
o

tdt , a0

g (t )dt T T
o
o

The representation of a periodic signal by its Fourier series expansion is equivalent to the resolution of the signal into its various harmonic components. The magnitude of the coefficients give the power in each component.

Figure 1.1 Square wave. For example the Fourier series expansion coefficients of the square wave shown in figure 1.1 is:

g (t )

4V

1 1 (sin 0 t sin 3 0 t sin 5 0 t ... 3 5

The dBm The most common power measurement unit in communications is the dBm. It is a reference measure compared to 1mW power:
dBm 10 log PO 0.001

The Spectrum Analyzer The traditional way of observing electrical signals in the time domain using an oscilloscope enables the recovery of relative timing and phase information needed to characterize circuit behavior. This is not sufficient for full characterization as it does not cover the frequency domain. One instrument used to display the frequency domain is the spectrum analyzer. It graphically displays power or voltage as a function of frequency on a CRT. The spectrum analyzer can be viewed as a radio receiver which shows the power of the signal it is tuned to. As the tuning is swept across many stations (frequencies) a graph showing power vs. frequency can be obtained. The most common spectrum analyzers sweep the spectrum of a signal through a fixed bandpass filter. Similar to the principle of superheterodyne receiver. A block diagram of a spectrum analyzer is shown in figure 1.2. The analyzer is basically a narrowband receiver, which is electronically tuned in frequency by applying a sawtooth voltage to a voltage-controlled oscillator (VCO). The same sawtooth is applied to the horizontal deflection plates of the CRT. The VCO frequency is mixed with the input to produce an intermediate frequency (IF).

Figure 1.2 Spectrum Analyzer block diagram. The frequency component of the input equal to the difference between VCO and the IF is shifted to the IF, filtered and amplified and passed on to the detector. This produces a voltage proportional to the power in that frequency component of the input, which causes a comparable vertical deflection of the CRT beam at the relative horizontal position. This gives a plot of power vs. frequency. It is important to note

that a peak always occurs at the zero frequency of the analyzer due to feedthrough from the VCO. This peak is called the zero frequency indicator and appears even when no signal is applied to the analyzer. The main controls of a spectrum analyzer are: The center frequency, which slides the display in the horizontal axis thereby changing the display position relative to the full spectrum. The span, which changes the horizontal scale allowing wider or narrower range of the display. The resolution of a spectrum analyzer is determined by its IF filter bandwidth. The marker, used to find the exact amplitude and frequency of a certain point in the trace on the CRT. The bandwidth, which selects one of several IF filters in the analyzer. Each filter has different width to allow higher resolution of adjacent signals. However, there is a limit to the narrowness of the filter bandwidth. This results in a pure sine wave to appear as a bell shaped trace instead of a sharp vertical line. The reference level, spectrum analyzer vertical scale starts from the top of the display. The reference level for the analyzers used in the lab is 27dBm. The vertical display is in dB, Where each vertical unit scale below the top represents 10dB step. The video filter is a process that displays the average of the display. This is useful when the signal is varying, such as noise. The display of the analyzer shows a spike at 0Hz. This is internally generated due to the VCO signal. This spike is useful in determining the 0Hz point on the display. Its amplitude is not important. Noise is a common pheneomena in all electronic systems. It is manifested by the shape of grass at the bottom of the spectrum analyzer display. The level of this grass is the noise power measured by the spectrum analyzer. It also puts a lower limit on the signals that can be measured by the spectrum analyzer.

Figure 1.3. Spectrum analyzer display.

The spectrum analyzer display in figure 1.3 shows a single tone signal. If the center frequency is 20MHz and the span is 10MHz, then the tone is at 18MHz. Given the reference level at 27dBm, then the tone power is about 43dBm. Since the spectrum analyzer has a 50 termination, the tone is 1.58mVrms.

1.3 Equipment
Spectrum Analyzer Dual-Trace Oscilloscope Digital Multimeter Frequency Counter

1.4 Procedure
Part A- Calibration 1. Set the function generator to supply a 3MHz sine wave of 30mVP-P on the oscilloscope. 2. Use a BNC T splitter to connect this signal to the spectrum analyzer and the oscilloscope. 3. Record the amplitude on the oscilloscope after connecting the spectrum analyzer. VPk

4. Explain why the amplitude dropped. 5. Set the analyzer to 20dB attenuation. 6. Adjust the signal on the oscilloscope to sine wave 28mVPk-Pk (-27dBm) at 5MHz. 7. Adjust the Y-position so that the peak of the pulse is at the level of the top of the 3rd square of the display of the analyzer (as shown in the figure).

Part B- Spectrum of a simple sinusoid. 1. Set the signal to 800kHz sine wave of 20mVPk-Pk Remove attenuation of the analyzer. 2. Set the center frequency of the spectrum analyzer to zero Hz, the span of the display to 2Mhz/div. Remove any attenuation. Set the filter BW to the maximum value, and the video filter to off.

3. Sketch the analyzers display.

4. Reduce the span to 0.5MHz/div. 5. Record the spectrum on the display, use the marker to identify the different frequencies and amplitudes.

6. Compare your result with theoretical calculations. Theoretical P(dBm) Part C- Spectrum of a square wave. 1. Set the spectrum analyzer input to 20dB attenuation. 2. Change the signal into the analyzer to a square wave 100mVP-P at 3MHz. 3. Record the display, identifying the fundamental frequency and the four following harmonics. 4. Compare these results with theoretical calculations. 0 Frequency (MHz) Theortical (dBm) Measured (dBm) 1 2 3 4 Measured

Part D- External signals. 1. Switch off the function generator and connect the antenna to the analyzer. 2. Find an FM station between 88 and 108MHz, and record its power. FM station frequency Power(dBm)

3. Expand the span and reduce the filter bandwidth to draw the shape of the signal.

4. Calculate the current through the analyzers terminating resistor due to this signal. Current (mA) 5. What is the approximate bandwidth of this station? Bandwidth (kHz) 6. Measure the power of the noise level. Use the video filter to get a clear measurement. Power (dBm)