Analog Effect Pedals EE333 Project 1 Francisco Alegria and Josh Rolles
Introduction For the first project, we ve chosen to design two analog guitar effect pedals. This report will discuss the schematic design, PCB design, and include project specifications and a complete bill of materials. Due to the nature of designing multiple effect pedals, each pedal will have their own section after this introductory portion of the report. Project Abstract The goal of this project is to create a completely analog dual effects pedal which includes a delay and a chorus effect. The pedal will interface seamlessly with an electric guitar to provide these effects. The design should work on a single supply, either 9V battery or 9V from a DC input jack. The design will take an input signal from a guitar and output the signal modulated by the effect out to the amplifier or audio interface. The final design will be built on a PCB allowing us to place the PCB within an enclosure and use the pedal in a live setting. Once placed in an enclosure, we will implement a stomp switch for durability to allow the pedal s on/off function to be operated by foot. Project Specifications The circuit will be powered by a 9V DC power supply or Battery High input impedance 1M ohm Low output impedance 1K ohm -20 dbu Input and Output signals - Guitar to Amplifier Delay time between 31 ms and 342 ms (for delay pedal) 1/4 input jack an1/4 output jack THD less than 0.13% at 31 ms Delay THD less than 1% at 342 ms Delay True Bypass Controls (delay): On/Off, Wet/Dry mix, Feedback, Repeats, Level Controls (chorus): On/Off, Depth, Rate, Wet/Dry, Level Schematics This section of the report will include schematics for both the chorus and delay pedals. A discussion will also take place on the details of each design. The performance and test results of each design will also be discussed here.
Chorus Pedal Schematic Figure 1: Complete Schematic for the Chorus Pedal Schematic Details The chorus pedal schematic consists of four parts: a power supply, a triangle lowfrequency-oscillator, input/output audio buffers, and the effect itself. Each of these will be talked about in the following section. Design Details In this section, we will discuss the design of each subsection of the pedal in further detail. Sub-circuit diagrams may be found here to see the implementation of each section of the circuit. 1. Power Supply The power supply takes a 9V DC voltage from the AC/DC wall adapter. The anode of a diode is connected to this supply and we use the voltage at the cathode as a source for powering our op amps in other parts of the circuit. Additionally, this voltage is fed into a 5V voltage regulator to output a 5V DC source to power our PT2399 integrated circuit. Finally, a voltage divider is created using two 100 KΩ resistors to provide a reference
voltage to be used in our comparator (implemented as a Schmitt Trigger) and integrator circuits. The first of the two SPDT switches in our footswitch is implemented in the power supply. Upon pressing the switch, each circuit will either be powered properly or be connected straight to ground. A diagram of our power supply may be seen below in Figure 2. Figure 2: Power Supply used in the Chorus Pedal 2. Triangle Low-Frequency-Oscillator (LFO) The triangle LFO is a two part circuit that includes a Schmitt Trigger and an integrator. Included in this part of the design are two potentiometers that control the frequency of the oscillator and the depth of the effect. The middle pin of the depth transistor shares a connection with a BC337 NPN transistor that will be discussed shortly. Lastly, the oscillation is visible on an LED in the circuit. The full triangle LFO circuit may be seen in Figure 3. Figure 3: Triangle LFO used in the Chorus Pedal 3. Input and Output Audio Buffers As with most audio circuits, buffers have been implemented to isolate the input and output data from the rest of the circuit. Also included in this part of the design is the
second half of the true bypass effect. True bypass means that when the musician wants to continue playing their music without the effect of the pedal, the signal will pass through without any degradation. This can be seen at the top of the circuit in Figure 4. If the footswitch is pressed, the circuit will switch between sending the input straight to the output and being processed through the effect. A 10 MΩ input impedance and a 220 Ω output impedance can also be seen in Figure 4. Figure 4: Input and Output Audio Buffers 4. Main Chorus Effect Aside from the LFO, the bulk of the effect produced by this pedal is produced in the remaining undiscussed circuitry. The effect is primarily generated using the PT2399 Echo Chip. For a chorus effect, we want a small delay out of the chip. The PT2399 has a minimum delay time of 25 ms, however it also has issues with latching up upon startup. The main delay circuit comes straight from the PT2399 data sheet as the data sheet provides several examples of implementation of the PT2399 chip. The amount of time in which the effect gets delayed is determined by pin 6 on the chip. Connected to pin 6 are two transistors. The bottom one is implemented as a current sink with its base connected to the depth potentiometer. The delay time is directly impacted by the current pulled out of pin 6, so the transistor allows us to vary the delay time. This is the pin that causes latch up issues as well. To fix the latch up issues, it was suggested that we implement a RC filter on the base of the first transistor. Pin 6 needs to see greater than 1KΩ to prevent latching up. When first powering the circuit, the NPN is off and the resistance on the pin is much higher that 1 KΩ. As the cap charges, the NPN turns on and allows us to control the current with the current sink below it. The final part of the design is the switch seen right before the output buffer. This switch acts as a wet/dry switch. When switched to dry, we get 50% of the effect and 50% of the dry signal. When we switch to wet, we get 100% of the effect, which ends up turning into a very strange pitch bend effect.
Figure 5: Main Chorus Effect portion of system Design Performance The chorus pedal was built onto a breadboard and followed by a perfboard. The perfboard model may be seen below in Figure 6.
Figure 6: Perfboard model of Chorus Pedal The following tests were performed for the chorus pedal: frequency response, minimum/maximum rate, minimum/maximum depth, and a sound test. 1. Frequency Response The frequency response of the pedal was measured using NI Signal Express. The normal audio range of 20 Hz to 20 KHz was used to test. As seen below in Figure 7, the lower frequencies that aren t typically produced by an electric guitar don t respond as well as the other more common guitar frequencies. The input used to generate this plot was a perfect sine wave coming from the function generator. 0.25 Frequency Response 0.2 0.15 mv 0.1 0.05 0 20 200 2000 20000 Frequency (Hz) Figure 7: Frequency Response of Chorus Pedal
2. Minimum/Maximum Rate The rate of which the LFO could operate was measured using a sine wave input and the oscilloscope to view the oscillating output. The rate potentiometer was varied from all the way counter-clockwise to all the way clockwise allowing us to see the minimum and maximum potential rates of the LFO. We found a minimum rate of 51 milliseconds and a maximum rate of 2.96 seconds. These results may be seen in Figures 8 and 9, respectively. Figure 8: Minimum Rate Figure 9: Maximum Rate 3. Minimum and Maximum Depth The amplitude of the effect was measured to using a sine wave input and varying the depth potentiometer from all the way counter-clockwise to all the way clockwise. We found a minimum depth amplitude to be 35 mv and the maximum depth amplitude to be 81.8 mv. These results may be seen in Figures 10 and 11, respectively.
Figure 10: Minimum Depth Figure 11: Maximum Depth 4. Sound Test Finally, a sound test was performed to ensure that we weren t observing unwanted noise and the pedal was producing the desired effect. Upon using a basic synthesizer app on an ipad, we found the pedal to be working well. After verifying it on the ipad, we used a guitar and an amplifier and verified the circuit was performing as expected. Final Schematic Discussions for Chorus Pedal The chorus pedal was fun to make and design. Even without a PCB, the perfboard model is easy to implement into an existing pedal lineup. The use of the PT2399 chip was helpful due to the many example circuits in the data sheet. In the future, the use of more precise components may lead to a slightly cleaner effect, but the current model is still a very cool thing on its own.
Delay Pedal Schematic Figure 12: Complete Schematic for the Delay Pedal Schematic Details The delay pedal is very reminiscent of a circuit in the data sheet for the PT2399 echo chip, aside from some added features to push the effect to its maximum feedback and/or trim the effect to an exact value. This trimming ability features an additional two switches to switch to the programmed effect. The schematic contains a power supply, input/output buffers, the delay effect, and the added trim potentiometer section for setting exact sounds desired. Design Details In this section, we will discuss the design of each subsection of the pedal in further detail. Sub-circuit diagrams may be found here to see the implementation of each section of the circuit. 1. Power Supply The power supply for the delay pedal matches the power supply for the chorus pedal. Please refer to Figure 2 on this report for a circuit schematic and the power supply section of the chorus design details for a summary of the circuit.
2. Input and Output Audio Buffers As with most audio circuits, buffers have been implemented to isolate the input and output data from the rest of the circuit. These input and output buffers are very similar to the chorus pedal. The output impedance for the delay pedal is 100 Ω instead of 220Ω like the chorus pedal had. Figure 13: Input and Output Audio Buffers 3. Main Delay Effect A 100 nf coupling capacitor is added before the delay line to make the signal less muddy. Within the PT2399 chip, there are two op amps to be used by the user. In this case, the first op amp is implemented as a second order low pass filter with a cut off frequency of 2 KHz and gain of 3dB. After the signal has been filtered, it passes through the delay line on the chip and is sent out to pin 12. The signal then goes through the second op amp implemented as an active low pass filter and is also in series with a passive low pass filter, giving a cut off frequency of about 1 KHz. There is a connection between the active filter and the passive filter that will be fed back into the delay line with the repeat potentiometer. We use these filters to remove all of the unwanted noise that happens throughout the D/A/D conversion. After passing through all the filters, the signal then is sent to the mix potentiometer. This is similar to the depth switch on the chorus pedal. The JFET in the circuit acts as a switch. When the JFET is on, the audio passes through the delay line. When it is off, audio doesn t pass through the line, but the remaining echoes of the last sound can still be heard decaying. Similarly to the chorus pedal, the rate of the delay is controlled by pin 6 on the PT2399. We vary this current using the delay potentiometer in the circuit. As mentioned before, we also added trim potentiometers to allow us to reach an exact effect that is predetermined. We are able to switch between this programmed effect and the main effect using additional switches. By adding extra potentiometers, we were also able to increase the maximum rated delay time by quite a bit. The results will be seen in the following section. All of these details can be seen in Figure 14.
Figure 14: Main Chorus Effect portion of system Design Performance The delay pedal was built onto a breadboard and followed by a perfboard. The perfboard model may be seen below in Figure 15. Figure 15: Perfboard model of Delay Pedal
The following tests were performed for the chorus pedal: DFT of various input types, minimum/maximum delay, maximum feedback, and a sound test. 1. Discrete Fourier Transform for Various Inputs For the first test, we used the following parameters: Input type: Pulse burst Frequency: 440 Hz N Cycles: 44 Length 0.1 second The input signal versus the output DFT can be seen below in Figure 16. Figure 16: Pulse burst input and it DFT output The next test, we took the DFT of the output signal while the pedal was being bypassed. This can be seen in Figure 17 below.
Figure 17: Bypassed signal and its DFT For the next two tests, we turned the pedal on and used its minimum and maximum built in delay settings (without the use of the added potentiometers). These results can be seen in Figures 18 and 19, respectively. Figure 18: Minimum delay time (22 ms) and its DFT output
Figure 19: Maximum delay time (454 ms) and its DFT output 2. Minimum/Maximum Delay Time These tests were performed to see what the minimum and maximum delay times were with the added potentiometers. We found the minimum time to be 22 ms as stated above, and were able to take the chip s maximum delay time from 454 ms to 1.506 ms. Figure 20: Maximum Delay Time
3. Maximum Feedback This test was performed to verify that we were generating the maximum amount of feedback possible with the delayed signals. The result may be seen below in Figure 21. Figure 21: Maximum Feedback 4. Sound Test Finally, a sound test was performed to ensure that we weren t observing unwanted noise and the pedal was producing the desired effect. Upon using a basic synthesizer app on an ipad, we found the pedal to be working well. After verifying it on the ipad, we used a guitar and an amplifier and verified the circuit was performing as expected. Final Schematic Discussions for Delay Pedal The delay pedal design went very smoothly due to the circuit being very close the circuit in the datasheet. The addition of the maximum feedback and adjustable quick settings made the circuit unique and unlike any other delay pedal available in mass quantities. Implementing this in series with the chorus pedal we developed provides some very unique sounds.
Bill of Materials (Chorus) team digi-key part description quantity price each Falegria & JRolles Lab Kit 220 Ω Resistor 3 n/a Falegria & JRolles Lab Kit 1 MΩ Resistor 2 n/a Falegria & JRolles Lab Kit 10 KΩ Resistor 7 n/a Falegria & JRolles Lab Kit 15 KΩ Resistor 2 n/a Falegria & JRolles Lab Kit 100 KΩ Resistor 4 n/a Falegria & JRolles Lab Kit 47 KΩ Resistor 1 n/a Falegria & JRolles Lab Kit 4.7 KΩ Resistor 4 n/a Falegria & JRolles Lab Kit 22 KΩ Resistor 1 n/a Falegria & JRolles CF14JT82K0CT-ND 82 kω Resistor 1 0.1 Falegria & JRolles Lab Kit 100 nf Capacitor 7 n/a Falegria & JRolles Lab Kit 1 uf Capacitor 3 n/a Falegria & JRolles Lab Kit 10 nf Capacitor 2 n/a Falegria & JRolles Lab Kit 1 nf Capacitor 1 n/a Falegria & JRolles Lab Kit 2.2 nf Capacitor 1 n/a Falegria & JRolles Lab Kit 10 uf Capacitor 2 n/a Falegria & JRolles 478-1839-ND 10 uf tantalum capacitor 1 0.85 Falegria & JRolles Lab Kit 47 uf Capacitor 3 n/a Falegria & JRolles 445-173132-1-ND 10 uf Non polarized capacitor 1 0.61 Falegria & JRolles 293-14997-5-ND TL072 Op amp 2 0.69 Falegria & JRolles Not available on digi-key PT2399 Delay Chip 1 2 Falegria & JRolles LM78L05ACZFS-ND 5V Voltage Regulator 1 0.4 Falegria & JRolles BS170-D27ZCT-ND BS170 MOSFET 1 0.55 Falegria & JRolles BC33740TACT-ND BC337 NPN 2 0.34 Falegria & JRolles 1N4148FS_ND 1N4148 Diode 3 0.1 Falegria & JRolles L20165-ND Green LED 1 1.01 Falegria & JRolles SC1085-ND 1/4" Female Audio Connector 2 1.86 Falegria & JRolles EJ502A-ND 2.1mm DC Input Jack 1 2.7 Falegria & JRolles 987-1649-ND 10 KΩ Linear Potentiometer 1 1.44 Falegria & JRolles PDA241-SRT00-254A2-ND 250 KΩ Log Potentiometer 1 3.84 Falegria & JRolles Lab kit SPDT Toggle Switch 1 n/a Falegria & JRolles EG5497-ND Latching DPDT Stomp Switch 1 6.5 Falegria & JRolles 1470-2773-ND 9V AC/DC Wall Adapter 1 6.5
Bill of Materials (Delay) team digi-key part description quantity price each Falegria & JRolles lab Kit 10 MΩ Resistor 1 n/a Falegria & JRolles Lab Kit 100 Ω Resistor 2 n/a Falegria & JRolles Lab Kit 1 MΩ Resistor 1 n/a Falegria & JRolles Lab Kit 15 KΩ Resistor 1 n/a Falegria & JRolles Lab Kit 100 KΩ Resistor 1 n/a Falegria & JRolles Lab Kit 10 KΩ Resistor 11 n/a Falegria & JRolles Lab Kit 4.7 KΩ Resistor 1 n/a Falegria & JRolles Lab Kit 1 KΩ Resistor 4 n/a Falegria & JRolles Lab Kit 100 nf Capacitor 9 n/a Falegria & JRolles Lab Kit 10 nf Capacitor 2 n/a Falegria & JRolles Lab Kit 1 nf Capacitor 1 n/a Falegria & JRolles Lab Kit 2.2 nf Capacitor 2 n/a Falegria & JRolles Lab Kit 10 uf Capacitor 5 n/a Falegria & JRolles 478-1839-ND 10 uf tantalum capacitor 1 0.85 Falegria & JRolles Lab Kit 47 uf Capacitor 1 n/a Falegria & JRolles Lab Kit 100 uf Capacitor 1 n/a Falegria & JRolles Lab Kit 47 nf Capacitor 2 n/a Falegria & JRolles 293-14997-5-ND TL072 Op amp 1 0.69 Falegria & JRolles Not available on digi-key PT2399 Delay Chip 1 2 Falegria & JRolles LM78L05ACZFS-ND 5V Voltage Regulator 1 0.4 Falegria & JRolles J112FS-ND J112 JFET 1 0.57 Falegria & JRolles Lab Kit 1N4001 Diode 1 n/a Falegria & JRolles 1N4148FS_ND 1N4148 Diode 1 0.1 Falegria & JRolles L20165-ND Green LED 1 1.01 Falegria & JRolles SC1085-ND 1/4" Female Audio Connector 2 1.86 Falegria & JRolles EJ502A-ND 2.1mm DC Input Jack 1 2.7 Falegria & JRolles 987-1649-ND 10 KΩ Linear Potentiometer 2 1.44 Falegria & JRolles Lab kit SPDT Toggle Switch 1 n/a Falegria & JRolles EG5493-ND Stomp Switch 3 6.01 Falegria & JRolles 1470-2773-ND 9V AC/DC Wall Adapter 1 6.5 Falegria & JRolles lab Kit Trim Potentiometer 10k 1 n/a Falegria & JRolles lab Kit Trim Potentiometer 100k 2 n/a Falegria & JRolles 36-232-ND 9V Battery Connector 1 0.6 Falegria & JRolles PDB182-K420K-203B- ND 20KΩ Dual Potentiometer 1 1.91
Printed Circuit Boards PCB Layout Both the chorus and delay PCBs may be found below in Figures 22 and 23, respectively. Figure 22: Chorus Pedal PCB Figure 23: Delay Pedal PCB PCB Overview Each PCB was designed using a 2 layer board. The chorus PCB ended up being a little larger than the delay PCB due to the addition of an extra op amp and the resistor labels being placed outside of each resistor rather than in the white box in the silk screen. The dimensions for the chorus pedal were: 3 x 3.63. For the delay pedal, the dimensions were 2 x 3. PCB Design Details Both PCBs were designed by importing a MultiSim schematic into Ultiboard and utilizing the tools within Ultiboard for the layout. The components on each board were organized by hand to allow for easier routing of the traces. The auto routing function was helpful in these designs due to the large quantity of wires, but some handmade traces were still made. The PCBs were sent to Oshpark for fabrication, and we have received them back. PCB Performance Due to a lack of parts, both PCBs weren t 100% constructed by the end of the semester. There is no reason to believe the boards won t work once completely assembled, but without
having them put together we don t have a concrete knowledge of their performance. Both PCBs were modeled based off of the perfboard models we created, so we are confident that once assembled they will meet our performance expectations. Final PCB Discussion We learned that Iowa State may have disabled the two layer auto-routing function within Ultiboard while attempting to complete the design. The auto-routing function would not converge due to it attempting to make every trace on a single layer. We also learned that the auto-router can make some very unintelligent traces, so it s best to double check the wiring job it did. Lastly, we learned it is very important to label the cathode and anode on electrolytic capacitors since they re polarized. The chorus PCB does not have labels, so we had to look at the traces and figure out which end went where. Thankfully this was easy to do since it is a two layer board, but in the future we should pay closer attention to these small details. Conclusions Overall, this first project has taught us a lot about how to work through a system design. For larger systems such as these pedals (in comparison to the first soldering lab we did), perfboard models can be a really big pain. The amount of soldering and trace making required takes hours when a breadboard model can be constructed in a much shorter time. This was very good practice on making a more complex PCB as it showed us how important it was to spend time laying out your components in an intelligent way. The PT2399 chip made our lives a lot easier designing these pedals as there were many example circuits available to us in the data sheet.