Laser Harp

This page covers my laser harp project (a work in progress), the design and construction of an electronic musical instrument using laser beams for strings.

Some design decisions were made at the beginning which had a major impact on how the project would unfold:

• I wanted to create a design that doesn't use any high-performance or modern micro-controllers. I know it would be very easy to use a modern ESP32 or other micro which had abundant processing power and ample GPIO pins. The design using those parts would be very simple and boring, with one micro-controller doing everything with ease. I really wanted to work with smaller, simpler parts, and have a design that needed some thought to get around the limitations of using simple controllers like the 8-pin ATtiny85. With this design restriction in place, the door opened to the use of creative solutions like the use of the LM324 op-amps in the beam detection circuit, the MUX to reduce the need for GPIO pins, and the two-controller design with one ATtiny85 serving as the beam detection slave and one ATtiny85 acting as the master.
• I also decided at the outset that I wanted to have some kind of auto-play mode supported. I can't play any instrument at all, so in order for the laser harp to be something that I could actually play, I would need it to help me out. To this end, I decided to have the harp support two modes: 1) a manual play mode where musically-talented people could play it like it was any other stringed instrument, and 2) a mode where the user playing would not need to pluck the correct string, just any string. When in auto-play mode, the user controls when a note is played, but not which note is played. The design includes a 24LC512 EEPROM with an I2C interface, allowing songs to be loaded which users with no musical talent are able to play.
• The chip selected to generate the actual audio signals is the VS1053, which is available on a breakout board from Adafruit: VS1053 Breakout Board. The chip is booted into real-time MIDI mode, after which its sole role is to accept MIDI commands over a serial interface from the master ATtiny85 controller. The final version of my custom PCB will have the VS1053 directly on it, but the selection of this chip made development work easy, as a pre-assembled breakout board was available for testing. This chip sounds great considering the price, and gives the Laser Harp access to lots of different instruments and a number of audio effects as well.

Once the project is complete, I'll share all the KiCad files with the full schematic and PCB designs as well as the complete firmware.

First iteration (v1.0) of the Laser Harp development board:

The development board design was done so that I could have one small board with all the parts and all the needed functionality so that I could finish all the firmware development before having to worry about the physical layout of the final design. The final design will eventually have a body for the instrument and provide much more space for the laser beams, have a physical form that is easy to hold and play, etc., but I wanted to have something simple and self contained to work with so that I could finish all the circuit and firmware design without having to think about all those final details.

That decision turned out to be a really good idea, as the initial design had some issues around the voltage levels switched on by the TEPT4400 photo-transistors serving as the beam detectors. The initial design was using 5V for this, sending ~5V to the LM324 op-amps to compare to a reference voltage and generate the logic-level signal that eventually finds its way to an ATtiny85 through a MUX. The issue is that the LM324 doesn't support having an input that close to the positive rail, so I needed to introduce a circuit to bring down the voltage level that's fed to the TEPT4400 photo-transistor switches. In this first image, one can see the green PCB attached to the side of the board, which takes the 5V input and uses a TL431A to bring that down to just a smidge under 3V. Unfortunately this fix required cutting traces on the PCB, etc., but was easy to incorporate in the second version of the board shown below.

The first iteration also did not have the rotary encoder used as a control knob integrated into the board. I did plan ahead a little bit, and connected the free pins from the master ATtiny85 to a pin header so that I could play around with some ideas for what to do with them. The ATtiny85's used here are 8-pin parts, with two used for power and ground, leaving 4 pins available for GPIO. However, one of the pins is reserved for for chip reset, unless a special HV (12V high voltage) programming process is used to re-purpose the pin. This design needed all four pins for IO, so I did need to use HV programming to allow use of that final pin, which is used as an address bit to the MUX allowing support for 8 beams; without the final pin, it would only support 4 beams. In the end the rotary encoder turned out be a great addition, and testing was easy to do with the pictured bread board. The first version of the dev board PCB was perfect for finalizing development of the firmware and figuring out any open questions on the circuit design.

Second iteration (v1.1) of the Laser Harp development board:

This second version of the board fixed all the issues with the voltage level used in the beam detection circuit with the LM324 op-amps, incorporated the rotary encoder, and also did a better job of accomodating the red laser diodes as well. There is a small slot cut out of the PCB below each of the laser diode modules, making it super easy to place them in the correct place with good alignment. The power leads from the laser diodes also go through holes in the PCB and are connected to the back of the board with some surface-mount pads, which also gets the wires out of the front and really cleans up the appearance.

This second iteration worked perfectly, and all that is left is to finalize the design to be used for the production version of the laser harp. Overall, I'm quite happy that I only needed to spin one more revision of the dev board to get everything working perfectly. Another nice thing about the dev board is that this is a fully-working version of the laser harp with all the features, the only thing lacking is it being a small board instead of a physical form factor like a real instrument that is easy to play. -- I have four more of these boards, so if anyone wants a free Laser Harp development board (you would need to assemble yourself), let me know, and I'll ship one to you free of charge.