Experiment 1 Introduction to the Spectrum Analyzer

Prepared by: Arman Isayan Eric Goss Date: 9/9/2010

Introduction: The purpose of this experiment is to introduce the student to the use of the spectrum analyzer. The spectrum analyzer is a piece of test equipment that is used to measure the frequency spectrum of a signal. A spectrum analyzer displays signals in the frequency domain in the same way that an oscilloscope displays a signal in the time domain. Due to the frequency basis of this course, a spectrum analyzer will be used frequently.

Equipment List: Resistors: Capacitors: Equipment: 1k 10nF Spectrum Analyzer Function Generator Oscilloscope Software: PSPICE

Theory: In this experiment, a number of signals will be displayed on the spectrum analyzer. The student will verify that the spectrum that they observe is close to that which is predicted by Fourier analysis of the signal. In this comparison it should be noted that in general the voltages measured will not correspond in magnitude to the Fourier series coefficients. However, their frequencies will be accurate and their relative amplitudes will be accurate. For example, say that a signal:

is input to the spectrum analyzer. We know that the Fourier transform of this signal is:

1 The spectrum analyzer will display:

While the magnitudes may not be 5 nor 2.5, but the ratio of the two magnitudes will be 2. Likewise the components at 2000Hz. 1000Hz will have equal amplitudes as will components at

Preliminary Calculations: Attached

2 Procedure: Part 1: Set-up Procedure Work through self-guided demo for the spectrum analyzer.

Results: The spectrum analyzer is truly an amazing piece of testing equipment. It allows the user to visualize signals in frequency domain, collect all important data, and save all the work onto a floppy disk.

6 Part 2: Sine Wave a) Adjust the function generator to produce a 10 kHz, 1V p-p.

Results:

The image cannot be displayed. Your computer may not have enough memory to open the image, or the image may have been corrupted. Restart your computer, and then open the file again. If the red x still appears, you may have to delete the image and then insert it again.

7 b) Set the center frequency of the spectrum analyzer to 0Hz, and adjust the frequency span to 5 kHz/div.

Results:

c) Observe and describe the spectrum analyzers display as you vary the frequency of the function generator from 5 kHz to 20 kHz.

Results:

It is clear that as the frequency of the sinusoidal signal increases, the conjugate impulses get pushed out and away from the DC Gain position (zero frequency)

9 Part 3: Square Wave a) Adjust the function generator to produce a 10 kHz, 1V p-p square wave. Observe this wave form on the spectrum analyzer.

Results:

b) What frequency are these components at? Why?

Results: The frequencies at which impulses occur are integer multiples of fundamental frequency n 0. However, the reason that even multiples of
0

can not be detected is their zero magnitudes.

And the reason is periodic functions as such produce harmonics with decaying amplitudes.

10 c) Find the ratio of each of the observed components to the first component. Compare this to the theoretical ratio of the Fourier series components.

Results: Measured Values Theoretical Value

n=-1,-3

n=-1,-5

11 d) Change the vertical scale to 10 dB log. How many dB down is the 2nd non-zero component from the first? Compare these results to those of part c. Now compare the 3rd non-zero component to 1st.

Results:

Measured Values (dB)

Measured Values (mV)



12 Part 4: Triangular Wave a) Adjust the function generator to produce a 10 kHz, 1V p-p triangular wave. Observe this wave form on the spectrum analyzer.

Results:

b) What frequency are these components at? Why?

Results: The frequencies at which impulses occur are integer multiples of fundamental frequency n 0. Moreover, all impulses located at multiples of
0

have non-zero amplitude. And the reason

is periodic functions as such produce harmonics with decaying amplitudes.

13 c) Find the ratio of each of the observed components to the first component. Compare this to the theoretical ratio of the Fourier series components.

Results:

Measured Values n=-1,-2

Theoretical Value

n=-1,-3

14 d) Change the vertical scale to 10 dB log. How many dB down is the 2nd non-zero component from the first? Compare these results to those of part c. Now compare the 3rd non-zero component to 1st.

Results:

Measured Values (dB)

Measured Values (mV)



15 Part 5: Low Pass Filter a) Apply a 1 V peak sine wave to the input of the low pass filter below. Vary the input from 1 kHz to 100 kHz, making sure that the amplitude remains constant. Using the oscilloscope, measure the output voltage (peak value) over this frequency range.

Results:

16

17

b) Plot the measured values of frequency?

versus frequency. What is the cutoff

Results: Given  nF

17

18

Part 6: High Pass Filter a) Apply a 1 V peak sine wave to the input of the high pass filter below. Vary the input from 1 kHz to 100 kHz, making sure that the amplitude remains constant. Using the oscilloscope, measure the output voltage (peak value) over this frequency range.

Results:

19

  

19

b) Plot the measured values of frequency?

versus frequency. What is the cutoff

Results: Given  nF

Conclusion: Spectrum analyzer is a powerfull tool in communications since almost every discussion takes place around frequency domain. A number of different signals were fed into the spectrum analyzer and crusial data was displayed. The only problem that exists is the high pass filter tranfer function in log scale, which does not match with prelab predictions. 20