Skip to content

16. System Integration

Assignments

Here you can find a recording of the lecture from the 7th of May.

This week's assignments and learning outcomes, see here:

Individual assignment:

  • Design and document the system integration for your final project

Questions to be answered/from Nueval:

Have you answered these questions?

  • implemented methods of packaging?
  • designed your final project to look like a finished product?
  • documented system integration of your final project?
  • linked to your system integration documentation from your final project page?

System diagram of final project

System diagram of final project

Here you can see a system diagram that explains the main functions of the final project.

System integration

Travels of a humpback whale

The main idea behind the final project is to show on a map how many humpback whales travel all the way from Iceland to the Caribbean ocean. My imaginary whale travels from the east of Iceland to the Dominican Republic and back. I made drawings to explain the system integration.

A 3D model of a whale will hover in mid air. A magnetic float will be placed inside the 3D printed enclosure. Around the float there are holes where steel balls can be placed if it is necessary to adjust the balance of the model. This idea came from Bambu lab and it is explained in this video.

To create the model of the whale, I will mould it out of clay and then 3D scan it. Then I will add a cylinder at the bottom. This cylinder will fit into a socket on top of the enclosure around the magnetic float. I want to use this method to have the possibility to change the model and tell a story of another animal in the future.

The backside and the inside

The magnetic levitation module will be placed in a 3D printed enclosure. It came with fasteners and screws, so by having four columns extruded from the bottom, heat inserts can be placed in them and the levitation model can be fastened there. The colums would also would also allow airflow under the module, which is good if the module emits heat.

Extrusions from the bottom will hold the power cord (for the magnetic levitation module) and the USB cord (for the Raspberry Pi Pico W) in place.

There will be three pcb boards to control all functions. One of them is for the On/Off button and another one is for the Doppler radar that will be used to sense motion with microwaves. The third one is for the Raspberry Pi Pico W microcontroller that will control everything. To fasten the PCB boards there will be L-shaped brackets that will grab the top and the bottom of the PCBs. The PCBs will be slided into the slot from the side. Then they will be fastened with screws, bolts and washers through holes that will be on the PCB boards and on the brackets.

At the back of the enclosure I planned to have three holes, as you can see in the drawing below. One of them for the On/Off button and the other two holes would be used for power input to the levitation model and the microcontroller. Then Svavar Konráðsson told me that by using a 5V, 3A Step-Up/Step-Down Voltage Regulator S13V30F5 I could use one cord to power both the magnetic levitation model and all the electronics with the microcontroller. This means that I only need one hole for power input, not two holes like in the drawing.

The front and the inside

At the top of the enclosure is a shelf with brim around it. A plexiglass plate will be placed on this shelf. It will be made of blue plexiglass. On another shelf, a few millimeters under the plexiglass map, there will be another plexiglass plate. On it, individual Neopixel LEDs will be placed in an irregular line. They will light up one at a time and will shine through the blue plexiglass map at the top. When they light up, viewers can visualize a whale travelling over long distances on a map.

At the front there will be a hole for the Doppler radar to sense movement. When it senses movement, it will turn on the Neopixel LEDs. Then the Neopixels will light up one at a time.

The tree PCBs will be placed on the inside of three walls and they will be wired together. The lower brackets, that were mentioned before, will cover most of three sides and they will hold all wires in place.

The pins on the PCB with the Raspberry Pi Pico W microcontroller will have pins pointing in three directions. The pins will point to the Doppler radar and to the On/Off button. Then there will be pins pointing upwards, connecting the microcontroller to the Neopixels on the plate above it. A small area will be cut out of the plate for the wires to go up from the Raspberry Pi Pico W and connect to the Neopixels.

The On/Off button

I made a mould and casted a silicone button to use as an On/Off button. When it is pushed, it will turn on the Doppler radar. A small surface-mounted button will be soldered on a PCB board and placed thightly behind the silicone button. A rectangular frame will be extruded from the wall of the enclosure, to hold the silicone button in place. On the backside there will be a circular hole for the button. The PCB plate will be fastened to brackets with screws, bolts and washers. There will be pins ont the PCB plate, so that the button can be connected to the Raspberry Pi Pico W, that will be placed on another wall.

The PCBs

This image explains how the PCB with the microcontroller will be fastened. It also shows where the pins will be placed and which pins are connected to which components. Below is an image of the PCB in the 3D viewer in Kicad. You can rotate the model and if you choose the Left view on the cube, you can see the front side of the PCB.

Plexiglass as a PCB with Neopixels

This image explains how I plan to use a plexiglass plate as a PCB for the Neopixels. This will be the lower plexiglass plate in the model. Above it there will be another plexiglass plate with a map rasterized on it.

The lower plexiglass will have the Neopixels on it, as I mentioned before, and the traces will be cut out of copper foil in a vinylcutter.

Svavar Konráðsson mentioned that if I would make the traces go around the plate in 90°, the wires could possibly be hidden inside the 3D printed enclosure. I thought about it and decided to let them go around in 180° and fasten the pins on the underside of the plexiglass. The pins will be soldered on to a small PCB board that will be fastened to the plexiglass plate with screws and bolts. The plexiglass plate will be black and the screws will also be black. Only the top of the screws will be visible at the top of the plexiglass.

The copper foil traces will follow the edges from the Neopixels at the top and all the way up to the small PCB board with the pins on the bottom.

The magnetic levitation module

Here you can see the magnetic levitation module, the magnetic float and the Raspberry Pi Pico W microcontroller. You can also see the Doppler radar and the silicon On/Off button. The images show how these elements will be arranged in the middle and on three walls.

Note that the enclosure/base will be 3D printed and to strengthen sensitive extrusions, a fillet will be used to soften corners.

Power

The magnetic levitation module

The magnetic levitation module came ready to plug in the next socket. The only thing I needed was an adapter to be able to plug it into an icelandic power outlet. You can find information on the adapter here.

The Raspberry Pi Pico W needs between 1.8V and 5V to operate, according to the information here. The RPP will be powered with a 5V charger with Micro-USB and it will supply the Neopixels and the Doppler radar for power.

According to the information here, each Neopixel needs 60 mA. 0.06x10 = 0.6A. The Doppler radar can operate at 4-28V, according to the information here.

Testing

The magnetic levitation module

As the model will be stored indoors in safe conditions, I don't have to worry about external effects like humidity or heat. These are the tests that need to be done:

  • Turn the magnetic levitation module and all electrical components on and off, to see if doing this repeatedly will have an effect.

  • Leave the magnetic levitation module and all electrical components on for some time, to see if it produces much heat and also to check out the endurance.

  • Check whether the microcontroller, the Doppler radar and Neopixels work as they should so close to a magnetic field.

  • Check whether the magnetic float will float in air when there are two plexiglass plate and a thin plate of PLA between it and the magnetic levitation module.

Test results

In the video below you can see the test being performed. The magnetic float is in a box made of PLA and there are two plexiglass plates on the top of the module. The Raspberry Pi Pico with Neopixels and the Doppler radar are close to the magnetic module and they functioned as they should.

The magnetic levitation module can be turned on and off repeatedly without problems but the module does emit some heat, but not that much. The module will sit on supports made of PLA.

There is one more problem: The magnetic float hovers only a few centimeters above the magnetic levitation module. I knew from the beginning that this could be a problem and I think it will work, but I will have to make the distance as little as possible.

Testing

A model of enclosure in process

A live model

Here below you can see the model. I might make some changes to it but this is what I plan for it to look like. If you click on the Explode icon, you can take a closer look at all components. You can rotate the model by left-clicking on your mouse and rotate it.