10/31/2019 at 03:30 •
Vasospaz has been working hard on building an upright laser harp with the design files I shared on Hackaday and Instructables. I am so excited that he is getting close to completing the build and that there is another upright laser harp out there. In our messages, Vasospaz brings up some important points for building the system and some helpful tips.
Here are photos of the system he built:
10/01/2019 at 07:02 •
I completed the design and layout for the upright laser harp shield, and now I am working on the shield's CAD model. Here you can see how the ULH shield will stack on top of the Arduino Mega and Music Maker shield.
The following connectors are on the shield:
- Laser and photoresistor signals for corner one (7-pin JST)
- Laser and photoresistor signals for corner two (7-pin JST)
- Laser and photoresistor signals for corner three (7-pin JST)
- Laser and photoresistor signals for corner four (7-pin JST)
- Potentiometer (3-pin JST)
- Rotary encoder (5-pin JST)
- Photointerrupter (4-pin JST)
- Motor driver board (5-pin header)
Instead of the hundreds of cable connections required for prototype one, the second version will only have a few cables running through the box, making the electronics much cleaner and easier to assemble. The white boxes in the diagram below are JST connectors.
The first version (pictured below) is a lot of wire management! This challenge will go away in version 2.
09/30/2019 at 00:18 •
The instrument wheel turns using a stepper motor and displays the current instrument sound of the laser harp. When stepper motors are used for many projects, you need to know the position of the motor. For example, stepper motors with lead screws in 3D printers need to be calibrated so that the extruder doesn't run into the end of the rails. Some motors have outputs to signal their relative positions, but most motors just keep stepping.
In the first version of the laser harp, I used a Hall Effect sensor to "home" the instrument wheel. The motor would turn one step and the Arduino would read the output of the Hall Effect sensor. When the sensor was over the magnet, the signal would change. With this procedure, one knows the position of the instrument wheel and can sync the MIDI instrument being played with the panel displaying the instrument.
The Hall Sensor worked fine for one system, but it isn't the best setup for building more laser harps and making kits. I am worried about dealing with small magnets and glueing them onto the wheel in the correct position. The challenge is even greater for bipolar Hall Sensors because the build requires getting the magnet orientation correct. Error in the positioning of the magnet could be offset in the code, but I want better repeatability in the hardware.
Therefore, I switched to a photointerrupter for the instrument wheel home sensor. A photointerrupter has an IR emitter and phototransistor oriented towards each other. When something gets in the way of them, the phototransistor signal changes. The object getting in the way is called a flag and is positioned at a location for homing the motor.
I redesigned the instrument wheel for this new type of home sensor. One of the panels on the wheel is extended. As the motor rotates, the photointerrupt is read by the Arduino. The path between the IR emitter and phototransistor is blocked only when the one spoke passes through.
09/29/2019 at 06:09 •
I finalized the design for the kinematic laser mounts, and now I am designing the top layers of version 2. Aside from these mounts, there were a few things I wanted to improve:
- Hide the wires down the side of the device
- Hide the bolts for adjusting the kinematic mount
- Decrease the curvature of the plastic diffuser for the photoresistors because it is difficult to fit them into the current mount
Here is how these components look in the first version:
The following screenshots show how version 2 overcomes these shortcomings.
With the cover on:
And with the cover off. You can see the wires running down through the hole.
Version 2 CAD design for the top of the system:
09/27/2019 at 03:13 •
I have been taking a short break from upright laser harp V2 to summarize the performance and cost of the first prototype. I made a shorter video about the device, motivation, and future plans. In the video, I describe my goals for the project, specifically making it into a unique kit at the intersection of art and engineering. I want the kit to promote interdiscliplinary learning and motivate musicians and engineers to learn coding, electronics, and how to play music.
I also made a spec sheet and bill of materials for the first prototype. These two documents give all the essential information for the device and are uploaded on Hackaday. The current cost of goods for one unit is around $180, which doesn't include energy costs for running the laser cutter and labor for assembly. I think the price would go down if components were ordered in larger numbers. Additionally, if the device was made into a kit, then the labor would only go into putting all the parts together in a box.
Here is a look at the Technical specifications:
09/21/2019 at 06:22 •
The design for the Upright Laser Harp shield is completed! This shield will be placed on top of the Music Maker Shield and Arduino Mega. The following components will be connected directly to the shield:
- Laser control signals
- Photoresistor analog output
- Volume potentiometer
- Instrument wheel rotary encoder
- Hall effect sensor
- Buck converter (12V input to 5V output)
The most difficult part in the upright laser harp build is all the wire wrapping and building up the prototype board. Now all the these components will just connect via JST into the board. With this design improvement, there will be no tracking wires to the right Arduino pins required and no wire wrapping.
The entire electrical setup with be compact and fit within the footprint of the Arduino Mega, which is much smaller than the current setup up with prototype board. In the first prototype, all the signals from the lasers and photoresistors connected to the prototype board and then had to all be wired again to the Arduino. Now there will only be one connector to the Upright Laser Harp shield. Here are some screenshots of the PCB design in Autodesk Fusion.
I know there is a lot going on in these schematics, so I made a high-level overview of how everything connects to the shield. The white rectangles are JST connectors. In the screenshots above, the laser and photoresistor signals from the four corners of the device are labelled as LAPH0, LAPH1, LAPH2, and LAPH3 (LAPH is a abbreviation for LAser and PHotoresistor signals).
After a lot of brainstorming and designing, I finally arrived at the second iteration of the electronics for the upright laser harp. I ordered some boards for building up version 2.
09/19/2019 at 05:30 •
I completely changed my mind for the electronics layout for V2 of the upright laser harp. Initially, I was a little disappointed, because I put a lot of time into brainstorming and designing PCBs for each layer of the laser harp. It was only when I reached the end that I realized something was fundamentally wrong with the design.
I started counting up all the connectors and PCBs and realized that my new "improvement" for version 2 was actually inefficient! I would need a PCB for each laser and photodetector, totally 12 PCBs. I would also need a connector cable between each PCB. In the electronics box, I would still have to connect the front panel components to the Arduino. It turns out that my first laser harp prototype was not so bad: one prototype board connecting all the photoresistor outputs, laser control signals, and front panel components. So the initial plan for V2 electronics is officially axed.
The new plan is to optimize the original format with a single PCB in the form of an Arduino mega shield. The upright laser harp shield will connect directly to an Arduino Mega and on top of this shield will be the Music Maker Shield.
Although this change back to the original electronics design scheme feels like a setback, I know it is a better plan that will make the second version easier to build. This is the goal anyway! There is always something to gain from being open to starting over or reverting back to an original idea. You just never know at what stage of a design (or in this case redesign) you will realize that you are going down the wrong direction!
Here is the current state of the new schematic: the Upright Laser Harp Arduino Shield:
An explanation and full schematic will be coming soon!
09/18/2019 at 04:45 •
Each corner of the laser harp has three pairs of photoresistors and lasers. Therefore I need to run down the following signals to the electronics box:
- Laser module control signal 1 (LO1)
- Laser module control signal 2 (LO2)
- Laser module control signal 3 (LO3)
- Photoresistor output 1 (PO1)
- Photoresistor output 2 (PO2)
- Photoresistor output 3 (PO3)
In the first prototype, the leads for the photoresistors and lasers were pulled down through each layer and connected to the control circuit in the electronics box. So the laser and photoresistors at the top required longer cables. This design makes it difficult to differentiate the control/output signals and manage the cables.
Instead of this strategy, I want to have a small PCB on each row for version 2. The PCB will include the electronics for generating the control signal and analog output that can connect directly into the Arduino. The output of each PCB would connect to the input of the next PCB on the way down to the electronics box.
I wanted to use the same PCB for each layer, so I needed a way to connect the control/output signals for each laser and photoresistor to the right pin. The solution was to add solder pads at the output of the laser/photoresistor and then three separate solder pads connected to the appropriate control pins. On each board, the laser output solder pad would then get shorted to the appropriate control signal pin. So for layer 1, the LMS node (control signal for the laser module) would get connected to LO1 (the first laser control signal). I designed the board in Autodesk Eagle with the help of my favorite tutorial by randofo on Instructables.
I'm happy with this board for the upright laser harp V2. Now I am onto importing the PCB into Fusion and integrating it with my new kinematic mount (which also needs a bit more work).
In the mechanical model, you can see the two 8-pin JST mounts for the input and output of the PCB (refer to the top of the log for the pin description). The board also has a 2-pin JST jack for the laser module. The only downside to this design is that I now need to accommodate larger 8-pin jacks running through each layer of the laser harp.
09/16/2019 at 00:29 •
I designed and began testing the next version of the kinematic laser mounts. I took a few components from the previous design and started brainstorming modifications. I decided to attach the flexible mount to the rear component holding the screws instead of requiring a front component. This greatly reduces the size of the module.
Initially I designed the flexible part that could be 3d printing, which I think would be easier to get working. I completed the part and soon after getting the 3d printer ready, I started to reconsider. The whole goal was to make the second prototype easier to manufacture and 3D printing 12 flexible parts for each harp was going to take a lot more time than the first prototype. I went back to the drawing board and came up with a design to test.
3D printed design idea
Lasercutting design idea:
After finalizing the design, I laser cut the parts and put together a small test assembly. I then tested the alignment range (i.e. the angle the laser can be steered out of the mount) by placing graph paper 15cm from the laser mount.
By calculating the displacement of the beam on the graph paper, I can calculate the steer angle. The steer angle helps me to figure out if the mount is working well enough because I know roughly what is required for the laser harp. Here is a diagram showing the calculation.
I am able to adjust the beam around +/-4degrees in one direction and +/-1.5degrees in the other direction. I am getting close with this mount, but I need to make a few more tweaks before making a bigger test. Once I am happy with the mount, I will build up a layer and test it.
09/15/2019 at 03:15 •
The upright laser harp has been a successful first prototype, but there are a few parts of the design that make it difficult and time consuming to construct. My ultimate goal is to make the upright laser harp into a robust kit. I have the design plans for the first version freely available, but the current build is challenging with some tedious construction stages. Here I will outline my plans for version 2 of the upright laser harp that will make the whole thing easier to construct so it can reach a larger audience. I am also considering selling assembled versions of the upright laser harp.
1. Improved kinematic mounts
I spent the most time developing the kinematic mounts for laser beam alignment, but I am still not 100% satisfied with the results. In my first prototype, the lasers are glued into place and there is a "course" adjustment required. In other words, there is variation in how the lasers are positioned in the mount and if the laser is not put in the mount correctly, the fine tuning with the kinematic mount isn't enough to align the beams.
For the next iteration in the laser mount design, I want the lasers to fit firmly in place and make the kinematic mounts flexible enough that they can be used for the full alignment procedure. I also want the kinematic mounts to be more compact.
2. PCB for the photoresistors and laser diodes
In first prototype (V1), the wires from the photoresistors and lasers are fed down through the laser harp. This makes cable management messy and frustrating. My goal for V2 of the upright laser harp is to design PCBs to be positioned at each layer with cable outputs:
- Photodiode output
- Laser output
These outputs can be directly connected to the Arduino Mega without the need for a prototype board. With these PCBs, the cables can easily be connected between layers of the laser harp. I also want to use photodiodes instead of photoresistors because of faster response and power consumption.
3. PCB for combining Arduino and Music Maker Shield
The Arduino Mega and Music Maker Shield are expensive and I don't need all the pins. In the next stage of prototyping, I want to design a single PCB that combines the ATmega2560 and VS1053 ICs. This custom PCB will be specifically designed for the connectors from the photodiode/laser module PCBs and will cost less than buying the two boards separately.
4. Optimize mechanical design
Finally, I want to improve the overall mechanical design be reducing the requirements for glue, hiding the cabling down through the device, and optimizing the tolerancing. One major issue with the current design is the time required for jamming parts together and glueing because of tight or loose fitting components.