Electrical Engineer. Lab2. Dr. Lars Hansen

Electrical Engineer Lab2 Dr. Lars Hansen David Sanchez University of Texas at San Antonio May 5 th, 2009

Table of Contents Abstract... 3 1.0 Introduction and Product Description... 3 1.1 Problem Specifications... 3 1.1.1 Project #1(A) Specifications... 3 1.1.2 Project #1(B) Specifications... 4 1.1.3 Project #1(C) Specifications... 5 1.2 Problem Description... 5 1.2.1 Project 1 (A) Description... 5 1.2.2 Project 1 (B) Description... 5 1.2.3. Project 1 (C) Description... 6 2.0 Initial Design and PSpice Simulation... 6 2.0.1 Project 1 (A) Design and PSpice Simulation... 6 2.0.2 Project 1 (B) Design and PSpice Simulation... 9 2.0.3 Project 1 (C) Design and PSpice Simulation... 10 3.0 Protoboard Circuit... 11 4.0 Printed Circuit Board (PCB) Design... 13 5.0 Conclusion... 15 6.0 References... 16 David Sanchez Project 1 Page 2 of 15

Abstract For Lab II, project 1 was split up into three different stages; from this stages one will achieve a finished product on a PCB board. The first phase of the project was named 1(A), it consisted of building an active unity gain low pass filter. Furthermore, this filter had to be simulated on Pspice software and produce a prototype in a protoboard for demonstration purposes. Secondly, project 1(B) was developed to achieve a volume control, pre amplifier, and class AB power amplifier was developed. Similarly, this phase will have some of the aspects that took place in project 1(A), one will simulate it in Pspice software and adding the final result into a protoboard. Moreover, this phase will be connected directly after the low pass filter stage 1(A), which will be connected in the protoboard prototype also. Finally, project 1(C) included generating the required files to fabricate a PCB board, containing all three parts of the project. After the group received the protoboard they had to find any traces which were not well marked by the fabrications, the group used the VMM to measure this. Additionally, the group members had to start populating the PCB board and demonstrate it to illustrate the achievements of the project. 1.0 Introduction and Product Description EE 4113 will introduce the students to real implementation topics and procedures that will be helpful in the professional field. This course will work simultaneously with Senior Design class which will give the student a clear way of presenting their ideas to the public. This course will also guide the students in specific topics that are currently used in today s engineering society. 1.1 Problem Specifications 1.1.1 Project #1(A) Specifications 1. Use ±14V for DC power. 2. Use active-filter architecture based on the LF411-type (KF351). 3. The passband ripple should be less than or equal to 2dB. 4. High-frequency passband corner frequency f p,h = 14kHz. 5. High-frequency stopband corner frequency f S,H = 50kHz. 6. The circuit has a lowpass response and has gain within spec down to DC. David Sanchez Project 1 Page 3 of 15

7. The stopband must have an attenuation of at least 40dB below A p,max over the stopband frequency range f 50kHz. 8. Include 10μF electrolytic power-supply bypass capacitors where the two DC power lines enter the circuit. 9. Include smaller 0.1 μf bypass capacitors close to the positive and negative power supply pins of all op-amps in your circuit. 10. Use a 10kΩ load resistor for the output load of the filter. 1.1.2 Project #1(B) Specifications 1. Use ±14V for DC power. 2. Use an active-filter architecture based on the LF411-type (KF351) 3. The passband ripple should be less than or equal to 2dB. 4. High-frequency passband corner frequency f p,h = 14kHz. 5. High-frequency stopband corner frequency f S,H = 50kHz. 6. The circuit has a lowpass response and has gain within spec down to DC. 7. The stopband must have an attenuation of at least 40dB below A p,max over the stopband frequency range f 50kHz. 8. Include 10μF electrolytic power-supply bypass capacitors where the two DC power lines enter the circuit. 9. Include smaller 0.1μF bypass capacitors close to the positive and negative power supply pins of all op-amps in your circuit. 10. Include smaller 0.1μF bypass capacitors close to the drains of the two power MOSFETs. 11. Include even smaller 0.01μF bypass capacitors close to the drains of the two power MOSFETs. 12. The passband voltage gain must be adjustable over the range -10dB A p,max +20dB. 13. Use a class-ab push-pull final output stage constructed from an IRF510 and IRF9510 power MOSFET pair. Use appropriate biasing and possibly negative feedback with an op-amp in order to minimize cross-over distortion. 14. The circuit must be capable of generating an output voltage of at least 20V pp into a 20Ω load when driven by a 1000Hz input sine wave of amplitude 2.0 V pp. The output sine wave should have at Total Harmonic Distortion (THD) less than 1%. David Sanchez Project 1 Page 4 of 15

1.1.3 Project #1(C) Specifications 1. The maximum allowable board size is 4.5" x 3.25" = 4500 x 3250 mils. 2. The BOTTOM layer should be the routing layer. This facilitates soldering since parts with radial leads are difficult to solder when the traces are on the same side as the part. It also makes for an easier connection to the BNC connectors whose center pins exit on the BOTTOM side of the board. 3. The TOP layer and any unused area on the BOTTOM layer should be used for ground plane. Be sure and place dynamic copper over these areas and make sure to attach to net 0. 4. DC ±14V power and ground will come into the board via wires and be soldered to the board. 5. Input and output signals will enter and leave the board via BNC connectors. 1.2 Problem Description 1.2.1 Project 1 (A) Description The first stage of the entire project was to build an active unity gain low pass filter that will meet the specifications listed above. To achieve this low pass filter design the group used a program developed by Texas Instruments called Filter Pro. The group decided to develop a Butterworth filter to achieve the given specifications. Furthermore, a low noise junction gate field-effect transistor dual amplifier was used. Moreover, the DIP utilized in this phase of the project was ST s TL072 which has small harmonic distortion. The main specifications were that the high frequency passband corner frequency had to be equal to 14 khz. Additionally, the stopband corner frequency had to be equal to 50 khz. Along with these requirements the project also required a passband ripple had to be less than or equal to 2dB. Finally, the stopband should have an attenuation of at least 40db/decade below Ap, max over the stopband frequency range greater or equal to 50 khz. After all this specifications where met, the group started simulating in Pspice for the required waveforms. Lastly, the schematic was built in the protoboard for real value testing. 1.2.2 Project 1 (B) Description This stage of the project consisted on designing and developing a volume control, pre amplifier, and a class AB amplifier. The specifications for this phase where: The volume control had to attenuate between the passband voltage gain which was -10dB and +20dB, it also had to achieve a voltage output of 20 volts peak to peak with no visible Total Harmonic David Sanchez Project 1 Page 5 of 15

Distortion (THD). Furthermore, this output voltage had to be driven by an input sine wave with amplitude 2 volts peak to peak and with frequency of 1 khz. After simulating this circuit in Pspice and looking at its final results it was implemented to the protoboard (Remember that the first phase is already on the protoboard, the second phase comes directly after the first phase). 1.2.3. Project 1 (C) Description The third stage for this project was to develop the files needed to fabricate a printed circuit board (PCB). Prior to developing these files the students had to read and fulfill all the examples that were given in the PCB Tutorial provided by Dr. Hansen. This tutorial will be a clear guide on how to develop this specific PCB. Moreover, the PCB will have the available space to populate the board with the specific components discussed in parts A and B. After the PCB is fabricated the group started testing the board for any shorts. Finally, the group began populating process on the board. The PCB board then was tested for previous specifications. 2.0 Initial Design and PSpice Simulation Initial design for this project requires basic understanding of the components that where going to be utilize in this project. Afterwards, the components where simulated in Pspice to have a clear understanding of how the system will work. This will help the group know how each phase of the circuit will react; it will also help achieve the specifications of each stage. It also will benefit the group to quickly simulate any changes in the real circuit and simulate in Pspice to see if there was any improvement on the system. 2.0.1 Project 1 (A) Design and PSpice Simulation For the first stage of the project the group had to design an active unity gain low pass filter. Furthermore, the group members started their research on how the lowpass filters worked and their main components. Afterwards, the group decided to design a Butterworth low pass filter to achieve the specifications for the given project. The group started simulating on the Filter Pro simulation from Texas Instruments (Texas Instruments), this will give the required frequency and right attenuation. The filter will consist of two LF 411 operational amplifiers, four capacitors, five resistors (including the load resistor which will be removed to unify project 1 (B) to this phase). David Sanchez Project 1 Page 6 of 15

The implementation stage started after the team knew which components where going to be taken into consideration for the design. The group started simulating the lowpass filter with resistor values with tolerances of 5% and capacitor values of 10%. After this values where inserted into Pspice, the group decided to simulate it using the Monte Carlo method which will give the user the various outputs one can achieve with the given inputs. Monte Carlo methods are a class of computational algorithms that rely on repeated random sampling to compute their results. Monte Carlo methods are often used when simulating physical and mathematical systems. Because of their reliance on repeated computation and random or pseudo-random numbers, Monte Carlo methods are most suited to calculation by a computer (Monte Carlo Method). Monte Carlo plot verifies that passband ripple < 2dB also shows all possibilities using resistor and cap values with tolerance. This plot will be illustrated in figure 1.0. Figure 1 Monte Carlo Simulation After modeling the schematic in Pspice with the Monte Carlo method the group was ready to implement the real values into the design. The parts where tested with the Digital measurement equipment available in Lab 1. Figure 2 shows the PSpice schematic of the lowpass filter and all the values of the components and the power supplied to the circuit. David Sanchez Project 1 Page 7 of 15

Figure 2 Real Values used for the Lowpass Filter After inserting the schematic into Pspice the group decided to run the simulation to verify that the circuit will meet the specifications listed for the project. By looking at Figure 3 one can see that the specifications for this project still meet, less than 2dB ripple and at 50 khz less than 40 db. Figure 3 Simulation for Real Values in Lowpass Filter David Sanchez Project 1 Page 8 of 15

2.0.2 Project 1 (B) Design and PSpice Simulation Stage of two of this project consisted of creating the volume control, pre amplifier, and a class AB amplifier. The basic schematic for this stage was provided by Dr. Hansen, but the group had to figure out the values for each of the components. The components that where implemented into this stage are: one LF411 operational amplifier, seven resistors, one potentiometer, and two power MOSFETS. First the group should be aware that the more efficient the amplifier becomes, the more power consumption it requires. The group used the recommended potentiometers to set the DC bias on the MOSFET; they had to adjust the pots so that the gate to source voltage of the two MOSFETS have no voltage. This stage should be added to part 1(A) (remember the resistor load from stage 1(A) should be removed), the schematic illustrating both stages added is shown in Figure 4. Figure 4 Schematic for Stages 1(A) and 1(B) David Sanchez Project 1 Page 9 of 15

Just as stage 1(A) this phase should also be simulated to make sure that this stage meets all the specifications. Figure 5 will illustrate the output for this stage which will already integrate the lowpass filter from phase 1(A). Figure 5 Output stage from part 1(B) added to stage 1(A) 2.0.3 Project 1 (C) Design and PSpice Simulation In phase three the group will create the files to manufacture the PCB which will be populated with the components that where design in stages 1(A) and 1(B). As mentioned before the group had to understand how the PCB creator works and to learn this the team had to go through a tutorial provided by Dr. Hansen. Before these files are created the users have to change the footprints for each of the components on the Pspice file. Moreover, the group also added some connections to be used as the power supplies. Additionally, the team had to add some BNC s to represent the incoming input signal. The BNC s will also be utilized to output the signal to the oscilloscope. Bypass capacitors where added to the schematic to be used as bypass capacitors. Electrolytic capacitors where also added to the design to reduce the noise incoming from the power rails in the system. After all this components where properly added and modified to the schematic the final result will be illustrated in Figure 6. David Sanchez Project 1 Page 10 of 15

Figure 6 Final Schematic to be send into the PCB creator 3.0 Protoboard Circuit In stage one; the lowpass filter was prototype with no extreme changes in the components. Different from stage one the rest of the design had several changes to make the specifications meet. Prototyping for phases one and two was done after the design was created and tested on Pspice. Phase two was really challenging for the group, since no one in the group understood how the potentiometers work and how they are measured with the VMM. These potentiometers had to be adjusted to have zero voltage through them to protect the MOSFETS. As power consumption increase, current starts flowing through the circuit and the MOSFETS might burn out. The group encountered many challenges in this stage, but solved it the only way electrical engineers can, by starting from scratch. David Sanchez Project 1 Page 11 of 15

After resolving all the problems that were presented in stage 1(B) the group started to take screenshots of the output for the final output from the circuit. Figure 6 will illustrate the desired output voltage of 20 volts peak to peak. Figure 7 Output for stage 1(B) Since the group already had the assemble part 1(B) into the breadboard it was only logical to take the Oscilloscope screenshot of the FFT of the 1000 Hz 10 Vpp sine wave across the 20 Ohm load. Figure 8 will illustrate this output with a frequency input of 3 khz. Figure 9 Figure 8 THD of output for stage 1(B) @ 3 khz David Sanchez Project 1 Page 12 of 15

will illustrate this same output with a 5kHz frequency input. Figure 9 THD of output for stage 1(B) @ 5 khz 4.0 Printed Circuit Board (PCB) Design In this stage the group gathered all the files necessary to fabricate their final PCB which was then populated with all the components design in previous sections. The group became familiar with the PCB editor to layout all of the components required. Furthermore, the components had to be edited on Pspice with their respective footprints. While the group was grouping the components in the PCB editor they had to make sure that the wires would not cross each other this wiring crossing is commonly known as ratsnest, Ratsnest command cleans up the un routed wires. A ratsnet reports the number of air wires remaining (Build your own circuit board). This nest would become more complicated as time went one and it took several hours to reduce it to where no wire would cross. The PCB editor had the option of auto routing the components, this is highly recommended since it will achieve the cohesive wiring for the components. Moreover, the group had to make sure that long wiring loops where not created to decrease any induction in the David Sanchez Project 1 Page 13 of 15

circuit. The group had many difficulties with the auto routing option since they were unfamiliar with PCB editor program. Once again the group had to start from scratch the design. After the routing took place with the correct distance that was required for all the components sizes group started creating the files for manufacturing. Furthermore, the group received the final PCB. The group was then required to check for any shorts in the board. If all the traces where check then the population process would begin. The population process required concentration and steady hands for the soldering to begin. The group started soldering the main components, such as the power amplifier. Afterwards, these amplifiers where tested to verify that they worked correctly. Since the ground plane was on both the top and bottom; the components that required both sides grounded where soldered on both sides. After all the testing for the amplifiers was verified by the team the pre amplifier and lowpass filter where soldered into the PCB board. Similarly to previous stages the team had some trouble, try to find different solutions to the problem but did not know what was wrong with the board. After Dr. Hansen took a look into the board we figure out that the board was correctly assembled and manufactured, the group then notices that one of the probes used for testing the board was not in the right setting. After the PCB was populated with all the components and was tested to meet specifications it was presented to the instructor to see if it met specifications. The results achieved by the group where: The passband voltage gain must be adjustable over the range - 10dB to 20dB, the group had 380mv to 20 Vpp. The screenshot for this result will be shown in Figure 10. The passband ripple should be less than or equal to 2dB, the circuit has a lowpass response and has a gain that will meet with the specifications, the group achieve Ap,max = 11.69 and (.79)*(Ap,max)=9.23. The high frequency stopband corner should be 50kHz, the stopband attenuation had to be at least 40dB, the team obtain (Ap,max)*(.01) = 117mV. Finally the team had to find if there were any visible THD, Figure 11 will demonstrate that no visible THD was in the circuit. All of this specifications where achieved by the group; they were proud of how their design concluded. David Sanchez Project 1 Page 14 of 15

Figure 10 PCB output with input 1KHz 20Vpp across 20ohm load Figure 11 Screenshot of FFT, PCB output with input of 1KHz 5.0 Conclusion After achieving a successful project I was satisfied with all the knowledge achieved from all the different stages in the project. It was really a gratifying experience to build a PCB, since this was my first time designing and working with an actual one. I learned a lot of how the David Sanchez Project 1 Page 15 of 15

design process for these boards are brought into real life components. This project will also be helpful for our Senior Design course. It was good to encounter some of the problems we as a team arrived, since this will help foresee any problems we will have in future projects. Furthermore, testing each stage worked concurrently with the given specifications was also challenging but all verified to achieve a great final result. When populating the PCB one as a student wants to do it all at once, but this is not the correct way since each stage was designed to meet specifications. Overall, I think this was a great experience on how PCBs are design and manufacture. I personally enjoyed this project and will do it all over again. 6.0 References "Low-pass Filters." All About Circuits. Feb. 2006. 15 Mar. 2009. "Capacitor Types and Colors." Elecraft. 20 Feb. 2009 <http://www.elecraft.com/apps/caps.htm>. Hansen, Lars. Electrical Engineer Lab 2. University of Texas At San Antonio, San Antonio. Spring 2008. Horowitz, Paul, and Hill Winfield. The Art of Electronics. New York: Cambridge UP, 1989. Monte Carlo Method Wikipedia, The Free Encyclopedia. 15 Apr 2009, 14:27 UTC. 1 Mar. 2009. <http//en.wikipedia.org/w/index.php?title=monte_carlo_method%oldid=287056722> Texas Instruments Filter Pro Simulation <http://www-s.ti.com/sc/techzip/slvc003.zip> Williams, Al. Build Your Own Circuit Board. New York: New York, 2004 David Sanchez Project 1 Page 16 of 15