EE 420L Engineering Electronics II - Lab 4

 

Authored by David Flores

Email: flored6@unlv.nevada.edu

Due: February 27, 2019

 

Lab Description

For this lab we are going to solve for bandwidth for inverted and non-inverted op-amps with a set unity frequency gain and a gain at 1,5, and 10. We also be looking into the slew-rate and how it works with both a square wave input and a sinusoidal input.

 

Pre-lab

 

 

Lab Instructions

Again, this lab will utilize the LM324 op-amp (LM324.pdf).

For the following questions and experiments assume VCC+ = +5V and VCC- = 0V.

 

 

 

 

Ensure that your html lab report includes your name, the date, and your email address at the beginning of the report (the top of the webpage).
When finished backup your work.

 

Experiment 1: Non-Inverting Op-Amp

Estimate, using the datasheet, the bandwidths for non-inverting op-amp topologies having gains of 1, 5, and 10.

Experimentally verify these estimates assuming a common-mode voltage of 2.5 V.

 

 

 

From the datasheet we can see that the Gain Bandwidth Product is equal to about 1.3MHz. This is also known as the unity gain frequency (fun).

 

With a gain of 1 the we would have  which means that the Bandwidth is equal to the unity gain frequency which is 1.3MHz

                   

For a gain of 5 we would have   bandwidth is equal to 260kHz.

 

For a gain of 10 we would have   bandwidth is equal to 130kHz.

 

Gain of 1:

 

Schematic 1:

 

Oscilloscope Measurement at a Gain of 1

Output is our blue waveform and it is at a magnitude of about 0.707V~67.2mVthis value would put it at 3dB which is where we measure our bandwidth but for this specific experiment we do not have a roll-off like normal since it’s a gain of 1 so the bandwidth is the unity gain frequency.

 

Gain of 5:

Schematic 2

 

Oscilloscope Measurement at a Gain of 5

Here the results of the oscilloscope show that the roll-off point is at about 3dB or 190kHz. For this experiment we had a bit of error this is due to the fact that we were not using an ideal op-amp. The hand calculations were made assuming we had an ideal op-amp

 

Gain of 10:

 

Schematic 3:

 

Oscilloscope Measurement at a Gain of 10

Output is our blue waveform the magnitude is about 0.707V at a frequency of 96kHz. The estimated calculations are not that close to the results of the waveform again the ideal op-amp can affect the output waveforms.

 

 

Experiment 2: Inverting Op-Amp

 

 

For this experiment we will be doing the same exact thing as experiment 1 except the op-amp topology is inverted. This is the same experiment except its inverted so only results will be provided.

 

 so

 

Gain of -1:  

                   

Gain of 5:

 

Gain of 10:

 

 

Gain of -1:

 

Schematic 4:

 

Oscilloscope Measurement at a Gain of -1

Bandwidth = 800kHz

 

 

Gain of -5:

 

Schematic 5:

 

Oscilloscope Measurement at a Gain of -5

Bandwidth = 150kHz

 

 

Gain of -10:

 

Schematic 6:

 

Oscilloscope Measurement at a Gain of -10

Bandwidth = 83kHz

 

 

Experiment 3: Slew Rate

Design two circuits for measuring the slew-rate of the LM324. One circuit should use a pulse input while the other should use a sinewave input.

 

 

In this experiment we built two circuits a unity follower circuit for both, but one has a square input and the other has a sinusoidal input.

 

Non-inverted with AC input and 2.5V DC offset                Oscilloscope Measurement square wave input

                                   

                                                                                                We have a rise of 1.01V and a run of 3.36us =0.3V/us

 

Square Wave Slew Rate Hand Calculations:

 

Here we can see that we designed a circuit that can show the slew-rate the “rising” voltage is too much for the “running” time to keep up so instead of showing a square wave like it should. Instead it shows a triangle wave. We noticed that as the input voltage increases we get a triangle wave and then after that it starts turning into a sinusoidal which means that the peak voltages are getting cut off because there is not enough time to keep up with the voltage. The data sheet says that we have a slew rate of 0.4V/us which is close to our experimental value of 0.3V/us.

 

Oscilloscope Measurement Sinusoidal input

 

 

 

 

 

Return to Index

Return to EE 420L Students

Return to EE 420L Homepage