Development of the final project
The idea
I live in a small town in the east of Iceland, Neskaupstaður. The town is in a fjord called Norðfjörður. We are surrounded by tall mountains and the sea is so close. Sometimes whales, small and big, appear above the surface and I am so faccinated by the thought about everything that is hidden under the surface.
I find the Humpback whale very interesting. It is beautiful and playful animal. It can be found all around Iceland. Fortunately, this is not an endangered species. It is common for the Humpback whale to travel all the way to the Caribbean ocean and back to Iceland during their seasonal migration. You can read about a research on it here.
Map of Humpback's traveling routes. Mackay, 2015.
What will the final project look like?
In Week 02 I developed the idea and here you can see what it could possibly look like:
As you can see, the idea is to 3D print a model of a whale and make it hover over a map. To be able to let it float in the air I wanted to use a levitation module that uses magnetic force to let a float hover in the air. Then I would design the model in a way that the float would stay inside the model. I would rasterize the map on to a plexiglass plate and maybe mould and cast a small replica of Iceland, where I live. Then I would use LED that would light up one by one to show the traveling route of the whale.
I have worked as an art teacher in an elementary school for many years and I have also taught other subjects like nature science and math. As a part of the team of teachers in elementary schools in Iceland, we were often encouraged to work with STEM, which is an acronym and it combines the subjects of Science, Technology, Engineering and Math. When the A is added to STEAM, Arts are included. Since I love working with Art and my impression is that when you combine Art with other subjects, it can spark interest and creativity with students and magic can happen.
This project can be used in STEM or STEAM education to spark students interest in many fields. By what I have experienced when teaching, just by catching the students attention with something that inspires them, it can create an interest in them that can make them want to learn and experiment. I think it can be used when teaching elementary students and even older students in highschool.
It can also be for people who want to visualize, and in a certain way connect themselves to an animal or species. It should also be an reminder that we, the human kind, should respect and care for our nature and the living beings that share it with us.
Levitation
As I mentioned in Week 01, I saw a levitation model used in the final project called Cloudio in Fab Academy by Wim Lance and it made me want to use levitation.
My instructor, Svavar Konráðsson sent me a link to this video. Here it is explained how the levitation module works, where it can be bought (Bambu Lab) and how to make 3D printed models levitate. it also has links to other information on using a levitation module with different types of 3D models. In the video it is also shown how a 3D module can be balanced with steel balls that Bambu Lab suggests using when it is difficult to balance models.
This video gave me important information because in the beginning I was worried that it might be hard to find the right balancing point. I bought the levitation module before Fab Academy began and tested it. When trying it out I found that when I didn't find the right balance everything collapsed. Since I want to 3D print a whale, which is very irregular in shape, I might have problems with balancing it right. Maybe it might be a good idea to make room for steel balls to be able to balance it if necessary.
Here you can see information about the levitation model and how it works. I bought my levitation model Here.
Tracking of whales
I contacted the Marine and freshwater research institute in Iceland and they were so kind to send me GPS information on the travels of Humpback whale and a Mink whale. I was surprised to see that the paths didn't show their travels further south, but I was told that the batteries run out. After the batteries run out the whales can be identified by taking photos of their tail and over 1000 individual Humpback whales have been identified, as you can read about here. I find that very interesting.
On this Tracking map for whales - website you can see the tracking routes map that the the Marine and freshwater research institute sent me. It is the North Atlantic Marine Mammal Commision, or NAMMCO, that does these researches on the Humpback whale. You can read more about them here.
When I looked at information on Mink whales, I started to think about changing from Humpback whale to Mink whale, because I like the too, but I have to make a decision soon. Updated: I have decided to go on with the Humpback whale.
Travels of Humpback whales, Marine and freshwater research institute, 2025.
Travels of a Mink whale, Marine and freshwater research institute, 2025.
Videos of mini model of final project
In week 05 I made a mini model of the whale, the box and the plexiglass map (in blue and matt clear).
Video of the mini whale model that was 3D printed
Here you can see the 3D model that I printed in Bambu Lab X1 Carbon. With the model I was trying to visualize what the final model should look like. If the whale is supposed to hover in the air with the help of magnetic force, the levitation module will be placed in a base and the float has to be placed inside the 3D model. The float is round and flat, so the bottom part of the ocean in the 3D model, should contain the float. I think I made the sea too tall because if the model is too tall it might be hard to balance it. In the final model I need to create space outside and around the float so that I can add steel balls or something alike in the right places to balance it. P.s. In the background you can see the fjord where I live; Norðfjörður. We often see whales appear, not the Humpback whale but other species.
Mini whale model that I 3D printed
Video of the mini model
This video shows a mini version of the base, the whale and the plexiglass map.
Mini model of the final project
System integration
System integration
Read about the system integration in week 16.
Spiral development
Spiral development
In the student bootcamp Miriam Choi talked about spiral development and explained how the idea should be developed from a basic idea and then add details and further development one step at a time. This image from the slides explaines it visually. Rico Kanthatham created this image:
Spiral development - main tasks
I used spiral development when working on my final project. This is the whole story of how each task was solved.
The enclosure
Spiral 1
Spiral 2
Spiral 3
The base
Step file for PCB used to make everything fit
When I had exported the step file from Kicad and imported it into the Fusion file, I could see what fitted and what needed adjustment. The PCB for the Raspberry Pi Pico was too big for the design and I saw that I needed to change the brackets. I was also planning on using the holes in the PCB as a reference when creating the holes for the screws.
More adjustments needed
There were a few things that needed to be changed in this model. After testing the model in the System integration week, I saw that to keep a short distance between the magnetic levitation module and the magnetic float, the columns under the module were far too high and they could not be smaller unless I changed the height of the enclosure.
Svavar Konráðsson also 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.
The almost final model in Fusion
I thought this would be the final model but then I saw that I had not given the USB cable enough space.
The final model in Fusion
I moved the brackets for the PCB with the microcontroller and added walls to keep the power cables in place.
Many mistakes
Failures
After 3D printing the enclosure I saw many failures. The walls that were meant to keep the power cables in place were not fitting. The holes that I created in the brackets and planned to put screws in were not good, because there was not enough space to add nuts. Also, there was not enough space for the Neopixels and the PCB on the lower plexiglass plate, so that the upper plexiglass did not sit on the shelf.
I had designed brackets for the PCBs to slide in, but it was very hard to slide all three in their place. I had to find another way to keep the PCB with the Rapsberry Pi Pico in place.
The space under the brackets was little and it was hard to remove support from there. This space was not useful, so I wanted to remove it. I kept on changing the enclosure and you can see the model here:
Changes in enclosure
Changes led to warnings
I tried to make the model smaller by reducing the height of it and get rid of the small, empty space under the brackets. When I did that, almost everything turned yellow or red with warnings, as you can see in the image below. I tried to make changes and go backwards but that did not go well.
Another way to make changes
After quite some time of trying, I decided to use another method to make changes. I clicked on Inspect and then Section analyzis. Then I chose which plane I wanted to start on and used the arrow to drag where I wanted the cross section to be. Then I measured how much I would have to extrude the brackets to fill the empty space.
More space for Neopixel PCB/plexiglass
I Extruded the top brim and then I extruded the shelf to create more space for the plexiglass plate with the Neopixel PCB.
More space for wires
I used the Extrusion and Cut option to shorten the lower brackets where I needed more space for wires. This is the corner where the wires from the Raspberry Pi Pico connect to the voltage regulator.
Fasteners for power cord
I created a new sketch on the bottom and drew two rectangles. Then I extruded them. The next step was to create a new sketch on the side of these cubes and draw a rectangle and a circle.
Midpoint used to place circle
I drew a line in the middle of the cube and used the Midpoint option under Constraints to place the circle in the middle.
Cut
Then I used the Extrusion and Cut option to create a hole. Now I had created two fasteners for the power cord. The cord should be able to slide in and stay in place.
Move
Then I saw that I had to move one of the fasteners. I right-clicked on the screen and chose Move/copy.
Soft corners
Finally, I added fillet to the corners of the fasteners.
Inner walls
System integration and packaging
I wanted to make the system integration and packaging better......
Top closed
I made a new sketch on the side of the wall and drew a rectangle on it. Then I extruded it until it closed the gap above.
Cover over voltage regulator
I added a cover over the voltage regulator by extruding a rectangle from the wall.
Sockets for heat inserts
Sockets
I was worried that I might ruin the enclosure if the heat inserts would not melt into the poles the right way. Svavar Konráðsson suggested that I could use sockets and I thought that it would be a sensible way to solve this.
Then I made the original poles on the bottom of the enclosure shorter and created a new sketch to extrude hollow poles around them. Then I added a new sketch on top of them and extruded massive poles.
small holes for heat inserts
The heat inserts have to melt into the poles but I was worried that it would not be easy. I decided to create small holes, hoping that they would guide the heat inserts the right way and still have enough material for the heat inserts to melt the pla so that they would be well fastened.
Another positive thing about using poles with sockets for fastening the magnetic levitation module is that if I have to remove the module it would be easy to lift them off.
Enclosure around magnetic float
Directions
I followed this video called How I Designed A Simple Threaded Container With Autodesk Fusion 360 to design an enclosure around the magnetic float.
I also foun directions with video called How To Create Threads in Autodesk Fusion - 2024 Update.
Circle with offset
I began by drawing a circle. Note that in this image the circle is 62mm but I ended up by making it 85mm. Then I clicked on Modify and Offset. The offset was set to -4.
Extrusion
I extruded the bottom (inner circle) for 5mm and then the outer circle for 20mm.
Thread
I clicked on Create and Thread. Here you can see the settings I used. I had to use offset to place the thread at the top. I was not using the same size as in the video so I was not sure if I would do this the right way, but I hoped so.
Fillet
I added a fillet to the bottom.
?
A
A
?
A
?
A
?
A
?
A
Final enclosure
Final enclosure
Here you can see the final enclosure. I added walls inside to hide wires and cover the voltage regulator. I also added fasteners for the power cord. The inner walls are slided down over the lower bracket and can be lifted easily off. They also stop the PCBs from moving around.
Enclosure 3D printed - material and settings
Plexiglass map and plexiglass PCB
Plexiglass for Neopixels and plexiglass at the top
I used this map when I made the plexiglass map at the top of the enclosure.
The whale model
Spiral 1
In spiral 1 I tried to create the basic form of the whale. I experimented with this both in Blender and by modelling it out of clay and then 3D scanning it.
Designing a whale in Blender
Blender
When designing the whale in Blender in week 02, I have been using three images as a reference:
-
This image from ABC News supplied by ORRCA member, Wayne Reynolds.
-
This image taken by photographer David Edgar and appeared on his Instagram account.
-
This image taken by photographer David Edgar and appeared at Daily Mail.
-
This image by Jón Baldur Hlíðberg. Jón Baldur Hlíðberg is an amazing icelandic artist who has done many artworks from icelandic nature, both animals and plants. I searched for his phone number, contacted him and asked if I was allowed to use his artwork as a reference and he approved.
Modelling a whale out of clay and 3D scan it
Whale modelled out of clay and then 3D scanned.
When I modeled the mini whale out of clay in week 05 I did a Google image search and used the overall results with many images when I modelled the whale. Here is a link to the Google search results.
Second attempt at drawing a whale in Blender
Drawing in Blender
I decided to give Blender a second chance after awful results in drawing a whale in week 02. This was done in week 15, but document it all here, so there is no link to it.
I used this video to begin with. It shows how to draw a great white shark, but I like the way he explains things. This is what I did when following the instructions.
In the video the default cube was deleted, so I began by doing that. Then I split the screen into two screens by moving my mouse up to the right corner of the screen until it showed a cross cursor. Then I dragged the splitting line over to the middle and then I had two seperate screens. After that I chose which view I wanted in each screen and this image from the video shows which views are chosen:
Here you can see what my screen looked like and how I chose the right view. I used the front view in the left window and the right view in the right window. Then I added one more as seen from the bottom. My numpads were not working as in the video, so this is how I chose the view:
To add a reference image I pressed on **Shift** and **A**. Then I chose **Image** and then **Reference**. Then I could add an image to use as a reference in Blender.
Name the reference and use opacity
I named the reference image as Front.
I clicked on the image symbol (Data) down on the right and checked in the box marked Opacity.
I added three images in total as references.
Shortcuts in Blender
A few shortcuts in Blender that I learned about in week 02:
Shortcuts in Blender
Here is a list of shortcuts in Blender. If you have an active keyboard:
By hitting the number 1 on your keyboard you get the front orthographic view. Nr. 3 takes you to the right orthographic view or side view.
SHIFT+D = Duplicate SHIFT+A = Opens up list of actions, f.ex. if you want to select an image and then click on Reference.
If you want commands to happen on a specific axis, x-, y- or z-axis, then you can hit the button with the first letter in the command and then write x, y or z after that. This way you can f.ex. hit the letter S (scaling) and then right after that, the letter Y. Then you can scale the object along the y-axis.
Hitting G and then x means that you can move it on the x-axis. Choosing E and then Y is for extruding on the y-axis. Use the R-key to rotate a line, plane or point.
I used the G and then x, y or z to move the images so that they would align at the same beginning point.
The whales in the images were not of the same size so I adjusted the sizes until they looked similar. Here below you can see how I was adjusting the sizes of two images.
I moved the images so that they look as if they are walls and a floor with room for the model between them.
Then I used Shift+A and chose Mesh and then Cube.
By clicking on S I could scale the cube down.
Edit mode or object mode
You have to be in Edit mode to edit the model and object mode to move the model. By clicking on TAB you go into edit mode or you switch between Object mode and Edit mode** at the top, as you can see in the image here.
I moved the cube where I could see it in the front view. Then I clicked on CTRL+R for creating a loop cut right in the center of the cube. It splits the cube in two parts. Then I dragged the mouse over the left half of it and chose Delete vertices. As you can see, the vertices that are chosen are orange in colour.
Then I dragged the mouse over the half of the cube that wasn't deleted, clicked on the Modifiers tool and clicked on the plus for Add modifier.
After clicking on Add modifier I chose Generate and then Mirror.
Object mode or edit mode
You have to choose between Object mode to see and move a model or Edit mode to be able to change the model.
In FreeCad and Fusion you can use the mouse to select a point, line or face, but in Blender you first have to choose which of these you want to select and work with. These tools are at the top left on the screen.
By clicking on the small Tool logo on the right of the screen you can choose a Modifier. I chose the "Mirror". Then I clicked on clipping.
By hitting Z you can click on Wireframe and then you can see through the object and it's split in half by the mirroring axis. If you want to remove the face where the two mirroring objects touch, you have to delete it. You click on Face select on the upper right, click on the face and delete it. To delete you can click on x and then choose Delete
Swithcing betweeen Wireframe and Solid view
To go back from Wireframe and to Solid view you just click on Z and then choose Solid.
Should I continue?
Maybe this is not my cup of tea. I am not sure if I should continue in Blender....
Final whale modelled out of clay
Modelled out of clay and 3D scanned
I bought clay and sculpted the whale. I used leftover materials that I found at my workplace as a support in the model; a piece of wood and a strip of metal that could be easily bent.
After sculpting the model out of clay, I used the Kiri engine app that I used in week 05 when I scanned a mini model. I placed the model on top of a chair when scanning it and the app scanned most of it, but not all.
Spiral 2
Spiral 2
In spiral 2 I thought about how to connect the model to a base with the float that will make the model hover in mid air. So the next thing to add to the model was a cylinder. But first it was necessary to delete the chair and the support that had been scanned with the whale model. I had a lot of trouble when trying to delete the lower part of the scan....
Final whale model on SketchFab
The PCBs
Spiral 1
Spiral 1
In spiral 1 I designed and produced a PCB board with a Raspberry Pi Pico microcontroller and two Neopixels. The board was designed in week 02 and produced in week 08. The plan was to have the Raspberry Pi Pico and the Neopixels on one PCB and fasten it with screws to a plexiglass plate. Then I planned on having a button on a seperate PCB.
Spiral 2
Spiral 2
In spiral 2 I decided to create three seperate boards for the Raspberry Pi Pico microcontroller, the Neopixels and the button. This was just a decision and was not performed.
Spiral 3
Spiral 3
In spiral 3 I decided to add a Doppler radar to the final project. I tested the Doppler radar in week 11 and managed to use it to communicate with Neopixels, so that the Neopixels turned on the light when the Doppler radar sensed movement. When I thought about how to organize the PCBs, I decided that it would be best to place the Doppler radar on one board. It think it's not necessary to mill traces, because the Doppler radar is ready on a board, but I needed to fasten it on a bigger plate. That would make it easier to fasten it to brackets.
Since the previous plan was to have the other components on three PCBs, the final number of PCBs was up to four.
PCB with Raspberry Pi Pico W
Mistakes
Second attempt to design the PCB with RPPW
Second attempt at designing the PCB with the RPPW
Below is the schematic design for the PCB with the Raspberry Pi Pico W. Button is connected to GPIO 4 Doppler radar is connected to GPIO 26 Neopixels are connected to GPIO 10
PCB editor with the RPPW 2
This is a screenshot of the design in the PCB editor.
PCB with Neopixels
Schematic design
Schematic
This is the schematic design for the PCB with the Neopixels. I used it for different designs in the PCB editor, both for a PCB board and soft copper foil traces when I was experimenting to find the best solution.
Which pins to use for Neopixels
Connect Neopixels to these pins
According to Adafruit Neopixels only work if they are connected to GPIO10, GPIO12, GPIO18 or GPIO21. The standard pin is GPIO18.
I connected the Neopixels to GPIO 10.
PCB board for Neopixels
I found a video that explained how you could import images into Kicad. I did a quick check and saw that I would have to zoom in on the world map for the Neopixels to fit on the map. Then I realized that it would probably be best to cut the traces in the vinyl cutter instead of using a PCB board. Below is the quick check that I made in Kicad and I explain the steps later.
I took a screenshot of this map and imported it into Inkscape (see image below). I chose Object and then Trace bitmap to create a vector drawing of the image. Then I deleted the original image. After that I adjusted the size of the vector image. Finally, I saved the image as a .dxf file.
Importing a .dxf drawing for reference into Kicad
Importing a .dxf drawing into Kicad
I imported the .dxf file into Kicad by choosing Import and then Graphics.
Choosing which layer you want the drawing to be in
I chose the User.drawings layer.
Settings for track width and more
Settings for traces in Kicad
I clicked on File and then Board setup. There I clicked on Netclasses. I chose the settings that you can see in the image below:
Arranging components with a drawing as a reference
Drawing used as a reference
I placed the Neopixels where I wanted them to be on the map. I placed the capacitors so that each of them was close to a Neopixel and by placing them above the Neopixels I could use them as a bridge over traces from other Neopixels. I placed the resistor on the Data track. Since the pins were supposed to be under the plate, The traces were placed in a straight line up so that they would fold easily around the edge of the plate and they are long enough to go over and under.
The copper foil, that the traces would be cut out of, is very thin. After measuring it the size of it was drawn in Kicad, just to make sure that the traces would fit onto it. This was limiting because this meant that the pins had to be above and not on the left side, close to the microcontroller, but that was not a problem. It only meant that the wires had to be longer.
When everything was wired together and the Design Rules Checker confirmed that the design was okay, I turned of the layer with the map and the plan was to export the design as a .svg. This image below shows the design in the PCB editor.
Issue with exporting a .svg from Kicad
Exporting a .svg from Kicad
I wanted to export the design as .svg from the PCB editor so that i could create a cutting file in Inkscape to cut the traces out of copper foil. I searched and just couldn't find out how.
Svavar told me that Bjartur had used File, Fabrication output. Then the Plot option and chosen .svg as an output. I checked and it was possible by changing the Plot format (see the upper left corner). Then I chose the file destination by clicking on the small folder symbol at the upper right and then i clicked on Plot.
Creating a cutting file for traces in Inkscape
Traces prepared in Inkscape
I opened the .svg file and deleted everything except the traces. I also had to combine all forms by clicking on Path and then Union because each straight line appeared to be a single form and the traces would all be cut apart if this was not done. In the image below you can see this process halfway; some lines are still single and in other parts of the traces the forms have been united. When everything was combined I turned off Fill, turned Stroke paint on and tuned the red colour up to 255. Then I set the Stroke style to 0.02. Then I clicked on File and Document properties to Resize to content**. Finally, I saved a .pdf file.
Copper foil tests
When I attended the Bootcamp in Húsavík, Svavar helped me do the first test in cutting a copper foil. The first test did not go well and then I ran out of time, so I had to continue on my own in Neskaupstaður.
Changing offset, speed and force in the Roland GX-24 vinyl cutter
Cutting settings
I found this information about settings for the Roland GX-24 vinylcutter. I was not sure which settings were ideal for copper foil, so I decided to do some testing. By setting the force to 150 the results were good.
I used a 60° blade and according to these information, I had to change the offset because the measurement from the edge of the blade to the middle of it differs between blades of different angles.
Offset, force and speed settings
I knew how to change the settings for force but I had no idea on how to change offset or speed. I found this video that explains how to change settings in the Roland vinyl cutter. It is not obvious so I will describe each step later, but first you can see a screenshot from the video that shows different offset settings for blades with different angles.
Offset, force and speed settings - step by step
To change the settings for offset, speed or force you first have to put the material (any material) in place and click on the Down arrow and then choose between Piece, roll or edge. Then click Enter. When the plotter has moved to it's place you click on Menu.
Click on Menu two times
You have to click on Menu two times to get the Unsetup text on the screen.
Click on the Down arrow
When you see the word Unsetup you click on the Down arrow until you see the word Condition.
Condition
When you see the word Condition then you click on the Right arrow. Then you see the word Force.
Force
If you want to change the force, then you click on the Right arrow and adjust the force with the up and down arrows. When you have made your changes you can hit Enter and then Menu to get quickly to be ready to cut a design. You can also back out of the settings with the left arrow.
If you were not planning on changing the Force and want to find other settings, such as for the Speed or the Offset, then instead of using the Right arrow you click on the Down or Up arrows.
Offset
When you see the Condition text, you click on the right arrow. Then you use the Up or Down arrows to find settings for the Offset. If you want to change the offset, then you click on the Right arrow and adjust the offset with the up and down arrows. When you have made your changes you can hit Enter and then Menu to get quickly to be ready to cut a design. You can also back out of the settings with the left arrow.
Speed
When you see the Condition text, you click on the right arrow. Then you use the Up or Down arrows to find settings for the Speed. If you want to change the speed, then you click on the Right arrow and adjust the speed with the up and down arrows. When you have made your changes you can hit Enter and then Menu to get quickly to be ready to cut a design. You can also back out of the settings with the left arrow. After hitting the Menu button you can see the main settings on the screen (see image below).
The copper foil tests
Tests
These are the four attempts at cutting the traces out of copper foil. None of them is good and the 60°knife (second from bottom) is not better than the other ones. The best one (lowest one on the picture) was done with Force set to 150, speed at 10cm/s and by using the 45°knife (with offset .25). I was not happy with the results.
New PCB for Neopixels since vinylcutting did not work
New design using the same schematic
I decided to do a new design for a PCB board but use the same schematic. It was supposed to be fastened onto a black plexiglass plate so I drew holes on it. The plate was barely big enough, 15.2cm x 15.2cm, so I had to place the holes unevenly.
Errors and warnings
I had to move silkscreen markings and to add a connection through the resistor for all to be without errors and warnings.
Generating Gerbers and using Gerber2PNG website
Gerber2PNG
In week 08 I explain how to generate Gerber files and also how to use the Gerber2PNG website that Fab Lab Kerala made to genetate .png.
Placing holes in PCB design match with holes in Inkscape
Holes in the PCB plate had to fit with holes in plexiglass
I exported the PCB design as a .svg by choosing File, Fabrication output, Plot and then choose .svg as an output by changing the Plot format. Then I chose the file destination by clicking on the small folder symbol at the upper right and then i clicked on Plot. Then I opened the file with Inkscape and added the outlines for the plexiglass plates and the map. This way I could place the traces where I wanted them and also use the lines for holes (see the grey circles).
Holes in the PCB plate had to fit with holes in plexiglass
Then I deleted everything that I didn't need until I had only the cutting lines for outlines and holes. I also added a small rectangle where I wanted the wires to be held in place and go into the enclosure.
So many mistakes
Too many lines and holes too big
When I cut the plexiglass I realized that I had made two mistakes. I forgot to delete the outline of the PCB that was supposed to be surface mounted on top of this plexiglass. I could also see that instead of making the diameter of the holes 3mm I set the radius to 3mm and the holes were too big. Fortunately the Roland Monofab SRM-20 was still just milling the traces so I could change the edge cuts for the PCB. Unfortunately I had to make another plexiglass.
The top line in the edge cut was not done in the MonoFab SRM-20
When milling and cutting the board in the MonoFab SRM-20, the top line was not cut. It looks as if the line went out of the working area of the machine, so the next time I will have to place the plate differently.
There was so little that needed cutting that I decided to use a small saw, that I have used for working with silver, to cut the small parts.
Soldering did not go well
When I soldered the board it did not go well. I used a soldering station and the Neopixels melted so easily. I ruined some of them and I think the reason is the tip of the solder iron, which was not small. Then I accidently broke the pins off, even though I was trying to be careful.
I decided to do an experiment and solder wires directly onto the board, instead of using pins. This way I would need less space between the upper plexiglass and the plexiglass that this PCB would be surface mounted on.
Fab Academy on Fishermen's day in Iceland
Fishermen's day
In the beginning of june the Fishermen's day is celebrated in Iceland. In my hometown ships and small boats sail together on this day. Here I was milling the PCB and watching my family and friends sailing.
View from Fab Lab Austurland on Fishermen's day
New PCB milled and soldered
New PCB milled and soldered
Since I was not happy with my work on the PCB I decided to mill another one. The one I milled had the cutting line at the top missing and also a part of the GND marking. This time everything was milled and cut. This tells me that I have to be accurate when placing PCBs in the Roland MonoFab SRM-20 because the working area seems to be limited. Se the older one and the new one below:
Mistakes
Troubleshooting
When I connected the board, only six of ten Neopixels lighted up and I could not find out what was wrong. I asked Svavar Konráðsson and Þórarinn Bjartur Breiðfjörð for help and they recommended checking if there was a short circuit somewhere and also check if the connections were good enough. I had already done that and could not see the problem. I had also removed and placed new Neopixels for nr. 6 and 7.
Þórarinn Bjartur Breiðfjörð also mentioned the oscilloscope and I used it to see where the signal was appearing and where it was not. This helped because I found that the signal was not coming out of the DOUT (Data out) on Neopixel nr. 6. This way I knew where the problem was.
More troubleshooting
I got worried that I might destroy the PCB by removing a Neopixel once more, because the traces can be destroyed and burned. Neopixels are fastened in four places and that is what makes removing them difficult. I asked a co-teacher, Viðar Guðmundsson, to help me and he used one soldering iron to heat up one side of the Neopixel, two pins at a time, and I heated up the other side. Then I could lift it off with tweezers. He told me to measure also the empty space between pins and different pins. He also told me that I could compare Neopixels, so that you measure between two pins (all pins, not just power and ground or only data pins) on one Neopixel and then the same pins on all the other Neopixels and this way you might see a difference and find a short circuit. This way I found out that there were tiny bridges under Neopixel nr. 6 that led to the empty space and connected pins that should not be connected. After cleaning the area with Quick braid and soldering a new Neopixel, the board with the Neopixel worked!!!
Videos showing Neopixels working
Neopixels showing how the humpback whale travels
These videos show how the Neopixels work; how the humpback whale travels from Iceland to the Caribbean ocean.
Neopixels blinking on the PCB
Neopixels blinking on map
PCB with button
Buttons - small and big
I used the same processes as described before when I made the PCB with the SMD button. A bigger silicon button, that I made in moulding and casting week, will be placed in front of this small button. Here you can see the small and big button in the enclosure design in Fusion.
Schematic and PCB design for button
Schematic of button PCB
This image below shows the schematic design for the PCB board with button. There are only two pins, one for ground and one for connection to GPIO 4 on the microcontroller.
PCB design of button PCB
This is a screenshot from PCB editor of the board.
Button PCB board milled and soldered
This is the PCB board with the button milled and soldered. I decided to add markings for power and ground.
The code
Spiral 1
Spiral 1
In spiral 1 I learned how to use Neopixels and made two Neopixels change colours. This was done in week 08.This is the code I used and it was Svavar Konráðsson who helped me write it:
import neopixel
from machine import Pin
import time
pixPin = 0 # Numper of pin
pixSize = 2 # How many LEDs
pix = neopixel.NeoPixel(Pin(pixPin), pixSize) # Initialization
yellow = (255,255,0)
red = (255,0,0)
while True:
pix[0] = yellow
pix[1] = red
pix.write() # Tells the LED to perform the command
time.sleep(1)
pix[0] = red
pix[1] = yellow
pix.write() # Tells the LED to perform the command
time.sleep(1)
Spiral 2
Spiral 2
In spiral 2 I had decided to add a Doppler radar to the program as an input. In the program the Neopixels start blinking when the Doppler radar senses movement. In this code I mixed the code from spiral 1 (here above) and a code that I used for Doppler radar in week 11.
I noticed that my Neopixels did not turn off in my previous experiments, so I have to tell them to turn off. I found here that I could possibly set the Neopixel to (0,0,0) to turn them off. I decided to set it up as off = (0,0,0) and it worked!
Here you can see the code after I mixed them together:
from machine import Pin, I2C
import network
import time
import neopixel
sensor_pin = 26
sensor = Pin(sensor_pin, Pin.IN)
pixPin = 0 # Numper of pin
pixSize = 2 # How many LEDs
pix = neopixel.NeoPixel(Pin(pixPin), pixSize)
blue = (0,0,255)
white = (255,255,255)
off = (0,0,0)
state = 0
val = 0
while True:
val = sensor.value() # read sensor value
if val == 1: # check if the sensor is HIGH
pix[0] = blue
pix[1] = white
pix.write() # Tells the LED to perform the command
if state == 0:
print("Motion detected!")
state = 1 # update variable state to HIGH
else:
if state == 1:
print("Motion stopped!")
state = 0 # update variable state to LOW
pix[0] = off
pix[1] = off
pix.write()
time.sleep(0.1)
Video from Spiral 2 - programming
Here you can see a video that shows how the Neopixels are activated when the Doppler radar senses movement.
Turning the globe
Spiral 3
The final code in progress
In spiral 3 I added Neopixels to the code....(work in progress)
I want the path to light up all the way from Iceland and to the Dominican Republic and then I want it to go backwards, because the whales travel back and forth. Each time a new Neopixel lights up I want the previous one to turn off, so I begin by setting the previous one to off, which I defined as (0,0,0). Then I tell the next Neopixel to turn on the blue light by writing blue, which is defined as (0,0,255).
I have to find out the best way to create a loop and I found these information on different ways to create repetition or loop. (work in progress - have to test it)
Small mistake in code
Too many Neopixels
This code has too many Neopixels. Since it begins counting at zero, it should only go up to nine if using ten Neopixels.
from machine import Pin, I2C
import network
import time
import neopixel
sensor_pin = 26
sensor = Pin(sensor_pin, Pin.IN)
pixPin = 10 # Numper of pin
pixSize = 10 # How many LEDs
pix = neopixel.NeoPixel(Pin(pixPin), pixSize) # Initialization
blue = (0,0,255)
off = (0,0,0)
# Initialize state variables
state = 0
val = 0
while True:
val = sensor.value() # read sensor value
if val == 1: # check if the sensor is HIGH
pix[0] = blue
pix.write()
time.sleep(1)
pix[0] = off
pix[1] = blue
pix.write()
time.sleep(1)
pix[1] = off
pix[2] = blue
pix.write()
time.sleep(1)
pix[2] = off
pix[3] = blue
pix.write()
time.sleep(1)
pix[3] = off
pix[4] = blue
pix.write()
time.sleep(1)
pix[4] = off
pix[5] = blue
pix.write()
time.sleep(1)
pix[5] = off
pix[6] = blue
pix.write()
time.sleep(1)
pix[6] = off
pix[7] = blue
pix.write()
time.sleep(1)
pix[7] = off
pix[8] = blue
pix.write()
time.sleep(1)
pix[8] = off
pix[9] = blue
pix.write()
time.sleep(1)
pix[9] = off
pix[10] = blue
pix.write()
time.sleep(1)
pix[10] = off
pix[9] = blue
pix.write()
time.sleep(1)
pix[9] = off
pix[8] = blue
pix.write()
time.sleep(1)
pix[8] = off
pix[7] = blue
pix.write()
time.sleep(1)
pix[7] = off
pix[6] = blue
pix.write()
time.sleep(1)
pix[6] = off
pix[5] = blue
pix.write()
time.sleep(1)
pix[5] = off
pix[4] = blue
pix.write()
time.sleep(1)
pix[4] = off
pix[3] = blue
pix.write()
time.sleep(1)
pix[3] = off
pix[2] = blue
pix.write()
time.sleep(1)
pix[2] = off
pix[1] = blue
pix.write()
time.sleep(1)
pix[1] = off
pix[0] = blue
pix.write()
time.sleep(1)
if state == 0:
print("Motion detected!")
state = 1 # update variable state to HIGH
else:
if state == 1:
print("Motion stopped!")
state = 0 # update variable state to LOW
pix[0] = off
pix[1] = off
pix[2] = off
pix[3] = off
pix[4] = off
pix[5] = off
pix[6] = off
pix[7] = off
pix[8] = off
pix[9] = off
pix[10] = off
pix.write()
time.sleep(0.1)
Add button
The next step was to add a code for the button. I found directions here and it is mentioned that some microcontrollers do not have a pull up resistor on all pins, so I Googled and found information here that explained how the Raspberry Pi Pico has internal pull up resistor.
The code for the button comes from here and looks like this:
I noticed that this code is very similar to the code from here and used to control a motor in week 10. The difference is that the motor code uses from machine import Pin and defines motor instead of button. The IN in machine.Pin.INdefines the button as input.
I will use from machine import pin and use GPIO 4.
from machine import Pin, I2C
import time
import neopixel
button = machine.Pin(4, machine.Pin.IN, machine.Pin.PULL_UP)
sensor_pin = 26
sensor = Pin(sensor_pin, Pin.IN)
pixPin = 10 # Numper of pin
pixSize = 10 # How many LEDs
pix = neopixel.NeoPixel(Pin(pixPin), pixSize) # Initialization
blue = (0,0,255)
off = (0,0,0)
# Initialize state variables
state = 0
val = 0
while True:
if button.value():
time.sleep(0.1)
if not button.value():
val = sensor.value() # read sensor value
if val == 1: # check if the sensor is HIGH
pix[0] = blue
pix.write()
time.sleep(1)
pix[0] = off
pix[1] = blue
pix.write()
time.sleep(1)
pix[1] = off
pix[2] = blue
pix.write()
time.sleep(1)
pix[2] = off
pix[3] = blue
pix.write()
time.sleep(1)
pix[3] = off
pix[4] = blue
pix.write()
time.sleep(1)
pix[4] = off
pix[5] = blue
pix.write()
time.sleep(1)
pix[5] = off
pix[6] = blue
pix.write()
time.sleep(1)
pix[6] = off
pix[7] = blue
pix.write()
time.sleep(1)
pix[7] = off
pix[8] = blue
pix.write()
time.sleep(1)
pix[8] = off
pix[9] = blue
pix.write()
time.sleep(1)
pix[9] = off
pix[10] = blue
pix.write()
time.sleep(1)
pix[10] = off
pix[9] = blue
pix.write()
time.sleep(1)
pix[9] = off
pix[8] = blue
pix.write()
time.sleep(1)
pix[8] = off
pix[7] = blue
pix.write()
time.sleep(1)
pix[7] = off
pix[6] = blue
pix.write()
time.sleep(1)
pix[6] = off
pix[5] = blue
pix.write()
time.sleep(1)
pix[5] = off
pix[4] = blue
pix.write()
time.sleep(1)
pix[4] = off
pix[3] = blue
pix.write()
time.sleep(1)
pix[3] = off
pix[2] = blue
pix.write()
time.sleep(1)
pix[2] = off
pix[1] = blue
pix.write()
time.sleep(1)
pix[1] = off
pix[0] = blue
pix.write()
time.sleep(1)
if state == 0:
print("Motion detected!")
state = 1 # update variable state to HIGH
else:
if state == 1:
print("Motion stopped!")
state = 0 # update variable state to LOW
pix[0] = off
pix[1] = off
pix[2] = off
pix[3] = off
pix[4] = off
pix[5] = off
pix[6] = off
pix[7] = off
pix[8] = off
pix[9] = off
pix[10] = off
pix.write()
time.sleep(0.1)
Powering the project
Powering the magnetic levitation module and the Raspberry Pi Pico W
Svavar Konráðsson showed me how to use a 5V step-up/step-down voltage regulator to connect the Raspberry Pi Pico and the Magnetic levitation module with one power input. This video, Power for the Raspberry Pi Pico - Guide to using VBUS, VSYS and 3V3 for external power circuits, provides some basic information on how you can power the Raspberry Pi Pico. Here are some points from that video:
Through the VSYS the Raspberry Pi Pico can be powered at mininum 1.8V and maximum 5.5V but through the VBUS it should be powered with 5V (plus or minus 10%). The current can be between 18mA to 95mA.
-
3.3V can give a maximum 300mA output.
-
The USB is connected directly to the VBUS.
-
If the plan is to connect external power to VBUS it is recommended to use the USB rather than pin 40.
-
You can connect external power to the VSYS pin, which is number 39.
-
A protection diode is placed between the VSYS and the VBUS. It prevents backfeeding power to the USB and protects against incorrect polarity.
-
Pin 36 can be used for external circuits at maximum 300mA.
-
A schottky diode can be used between VBUS and VSYS to prevent external power supply from backfeeding the USB power supply. It also prevents the external circuitry from using power from the USB supply.
-
If you use the schottky diode like this you can use an external power supply and connect the Raspberry Pi Pico through the USB without disconnecting the external power supply.
Then I connected the magnetic levitation module using this as a reference, see screenshot:
Code to check if the levitation model and the microcontroller work
Svavar Konráðsson wrote a simple code, based on Neopixel code from week 08 with all eight Neopixels turning red. We used this code just to see if everything worked as it should after connecting the Magnetic levitation module and the Raspberry Pi Pico W to the same voltage regulator, see week 08. This is the code:
import neopixel
from machine import Pin
import time
pixPin = 0 # Numper of pin
pixSize = 2 # How many LEDs
pix = neopixel.NeoPixel(Pin(pixPin), pixSize) # Initialization
red = (255,0,0)
while True:
pix[0] = red
pix[1] = red
pix[2] = red
pix[3] = red
pix[4] = red
pix[5] = red
pix[6] = red
pix[7] = red
pix.write()
time.sleep(1)
Presentation slide and video
final presentation title slide
Files for final project
File for tracking of whales
File for webpage with the tracking of whales
This webpage with information about the tracking of different species of whales comes from the North Atlantic Marine Mammal Commision.
File for mini 3D model of whale
File for mini 3D model of whale
This model was made in week 05. It was first sculpted out of clay and then scanned with the Kiri Engine app. Next it was resized and painted in Bambu Studio and finally 3D printed in Bambu Lab X1 Carbon.
This is a link to the .stl model on Sketchfab.
File very large
I downloaded the .3mf file from Bambu Studio after painting it and it was very large, or 10.2MB. I could neither upload the file to SketchFab nor my Repository. For that reason I uploaded it to my personal drive and here is a link to it. It is a read only file:
File for mini plexiglass map
File for mini plexiglass map
These mini plexiglass maps were made in week 07 (not as a part of the schedule). This small version of the final project helped visualizing how the final project could be done. I used a world map that I found on this site.
PDF file for mini plexiglass map with rounded corners_1
svg file for mini plexiglass map with rounded corners_1
File for mini 3D printed base
File for mini 3D model of base
I made this model in week 07. It was not a part of the schedule, but I wanted to see how it would turn out.
Box file 2
File for mini 3D model of base
This is a file of the first version of the base for final project.
Files - whale model
Modelled out of clay and 3D scanned
File for final whale model
This model was too big for my repository so here it is as a model on Sketchfab.
Electronics files
PCB with Raspberry Pi Pico W - first attempt (mistake)
PCB with Raspberry Pi Pico W - second attempt (final)
Second PCB with RPPW - .gbrjob
Second PCB with RPPW Edge_Cuts - .gbr
Second PCB with RPPW F_Cu - .gbr
PCB with button
Files for PCB with Neopixels
Traces for vinylcutting copper foil- .pdf
Traces for Neopixel PCB - .pcb
Traces for Neopixel PCB - .prl
Traces for Neopixel PCB - .pro
No electronics, only a PCB plate for Doppler radar
