ECEN 489 - RF Circuits for Wireless Communication

Lab 5: Intermediate Frequency Amplifiers

In this lab, we will build the intermediate frequency stages that amplify the modulated signal after down conversion
from the front end. Front ends do not amplify the signal enough for it to be usable for baseband processing so the IF
stage serves to amplify the down converted modulated signal while performing additional filtration of RF and LO
feedthrough which for simplistic mixer designs can be significant. Due to various channels being received at the
front end having different signal strengths because of various distances between different transmitters, the constant
gain value of the IF stage can saturate the signal and possibly destroy the information contained within. We will
look into one such compensation method in this lab for AM IF stages known as automatic gain control (AGC).

Figure 5-1: First and Second Stage AM IF Amplifiers with AGC

Figure 5-2: First and Second Stage FM IF Amplifiers

The analysis for all stages is similar using first DC analysis to find the small signal parameters. The first stage is
analyzed using the small signal assumption. For the second stage, the input is considerably larger than the first stage
so we need to use large signal analysis since the nonlinear terms from the exponential characteristic of the base
emitter junction contribute significantly to the output voltage signal. We also need to consider the early effect of the
BJT because large bias currents will reduce the impedance at the collector node. Since bypass capacitors act as AC
shorts from the emitter to ground, we can treat the early effect as a large resistor 𝑟!" from collector to ground:
𝑉"#$%&
𝑟!" =
𝐼!
where 𝐼! is the DC collector current of the transistor. 𝑉"#$%& or 𝑉' is an intrinsic parameter of a BJT provided in the
datasheet. For the discrete BJT, 2N3904, the early voltage is 75V. Upon inspection, the small signal gain of the first
stage measured between collector and base can be found by relating the load impedance and the small signal
transconductance of the transistor:
'
𝐴'1 = ' ( = −𝑔( (𝑍),(#+ ||𝑟!" )
)
where 𝑍*,,#- is the input impedance looking into the tank at the resonant frequency as discussed in the previous
labs. For the large signal gain of the second stage, also measured between collector and base, we need to include the
modified Bessel function into our expression to account for the nonlinearity. The large signal gain of the second
stage can be found as:
' 2, (+) /'1
𝐴'2 = ' ( = −𝑔( +,1 (+) (𝑍),(#+ ||𝑟!" ) = −𝐺( (𝑥)(𝑍),(#+ ||𝑟!" ) 𝑥 = 0) 1
) 0
where 𝐼. (𝑥) is the modified Bessel function of the first kind and nth order, 𝑉1 is the input driving voltage, 𝑘/ is
Boltzmann’s constant, T is temperature in Kelvin, and q is the electron charge. 𝐺, (𝑥)/𝑔𝑚 can be found from a
standardized table or via manual evaluation of the Bessel functions. To check if the large signal gain expression is
reasonable, small inputs of x should reduce the ratio of Bessel functions to unity and approach the small signal
approximation.

Deliverables: Read over all parts before starting.

Part 1:
a) Design the biasing structure of the first AM IF stage assuming R38 = 100Ω, the diode current is 267 μA, the
emitter voltage is 10% of the supply voltage, and the base current is negligible.
b) Find the gain of the first stage at 455 kHz. You may use the simulated value of tank impedances.
c) The input signal to the second stage is on the order of hundreds of mV. Show that the small signal assumption is
inaccurate for the gain calculation of the second stage.
d) Design the biasing structure of the second AM IF stage assuming the current through R36 is 172 μA, β = 140
(base current is not negligible), and there’s 11% of the supply voltage at the emitter. Check your values against the
radio kit manual.
e) Assuming there is a 0.56V swing input (1.12 Vpp) into the second stage, find the gain of the second stage at 455
kHz.
f) In Multisim, construct both stages individually. Measure the DC biasing voltages/currents and verify the transistor
is active. Measure the gain from collector and base with transient waveforms using 0.5mV (1 mVpp) into the first
stage and 0.56V swing input (1.12 Vpp) into the second stage. Note that AC sweeps are only valid for linear
networks. Use the same RLC values for transformers in simulations from previous labs.
g) On the first stage, insert an AM source of 455 kHz carrier frequency and 2 kHz intelligence frequency between
the cathode of the diode D4 and ground. Increase the peak voltage from 1 to 10 V. Measure the voltage transiently
for at least 25 ms at the base and collector of Q8. What happens to the signal at the collector and why? Where does
this large source feedback come from and what causes it to be large?
h) Cascade the two stages (without the AM source at the diode) and obtain the transient waveform at the output.
Why might the waveforms be different in magnitude from the two stages individually? What effects might not be
captured in models of the individual stages?
g) Provide screenshots of all schematics, simulations, and calculations in your lab report.

Part 2:
a) Design the biasing structure of the first FM IF stage assuming R25 is 220Ω, the collector current is 400 μA,
current through R16 is 150 μA, the emitter voltage is 10% of the supply voltage, and the base current is negligible.
Note the supply in Figure 5-2 is simplified, be sure to include all resistors and capacitors in your calculations.
b) Find the gain of the first stage at 10.7 MHz with an input of 120 mV swing (240 mVpp). You have to use large
signal model for additional accuracy. Assume the transistor can operate at this frequency.
c) Design the biasing structure of the first FM IF stage assuming R25 is 220Ω, the collector current is 830 μA, the
current through R16 is 150 μA, the emitter voltage is 9% of the supply voltage, and the base current is negligible.
Note the supply in Figure 5-2 is simplified, be sure to include all resistors and capacitors in your calculations. Check
your values against the radio kit manual.
d) Find the gain of the second stage at 10.7 MHz with an input of 0.8V swing (1.6Vpp). Assume the transistor can
operate at this frequency.
e) In Multisim, build both stages individually and verify the DC bias points. Measure the gain from collector and
base with transient waveforms with the same inputs specified in part a) and c). Let the transient simulations run for
enough time for the signal to stabilize to perform your measurement. Use the same RLC values for transformers in
simulations from previous labs.
f) Provide screenshots of all schematics, simulations, and calculations in your lab report.

Part 3:
a) Read Sections 3, 4, 7, and 8 from your radio kit manual.
b) Following the instructions from the radio kit manual, construct all four IF amplifiers on your PCB and perform all
tests to ensure its functionality. Be careful not to burn yourself with the solder.
c) For the AM IF amplifiers, input a 1 mVpp 455 kHz signal at the first stage and measure the waveform at the
output of the second stage.
d) For the FM IF amplifiers, input a 120 mVpp 10.7 MHz signal at the first stage and measure the waveform at the
output of the second stage.
e) Measure the bandwidth of all four stages and find the Q factor for each stage.
f) Provide screenshots of your measurements. Be sure to comment on the differences.

Be sure to make a table of the values you obtained in this lab. Ask your TA for help if needed with simulations
and/or measurement equipment operation. Provide a conclusion.

Appendix:
Calculator for Modified Bessel Function of the First Kind and nth Order:
https://keisan.casio.com/exec/system/1180573475