Projects

Since I wrote the article about the Bahnhofssteuerung 2000, development has slowed down but never completely stopped. One aspect that in my opinion should always improve alongside a project are the tools that you use for development. Not only for better productivity, but also to stay excited about a project itself and its development progress. 

In this case, it is the hardware test setup that got a massive improvement. For this project, this is already the 3rd iteration on the hardware setup, and by a lot the most professional looking one. 

The first setup was simply put together on breadboards with jumper cables, mainly to verify that everything works as expected. This was fine to validate a proof of concept, but had issues with unreliable Dupont wires that often introduce hardware bugs that are difficult to identify, reproduce and resolve.

The second iteration was already much more reliable. It consisted of two perfboards that were connected using solder joints, with pin headers that allowed for swappable components. All the required connections were made on the back to give the board a clean look from the outside, together with a 3D printed base. I used this test setup for most of the development work and it performed flawlessly. However, I always wondered about the orientation of certain components, and I then had to look at the cable routing on the back to remember which line goes where. 

To stop this waste of time, and to finally look at something more beautiful, I threw myself into the world of PCB Design. I haven’t designed a PCB in years but KiCad gave me a blast. It was easy to use and I got the hang of it in less than a day’s work. I quickly came up with this layout.

Overall, this is a big step up from my previous setup. Everything is clean, understandable and perfectly reliable. For the price of a pizza, I got 5 PCBs delivered home that were only missing the surface mount components. Those however were easily installed and the first tests showed that I did not make any mistakes during the design phase.

Now with this beautiful setup, development will be even more fun in the future and I look forward to designing more boards for other applications, but also in the context of the Bahnhofssteuerung 2000. 

If you need a simple PCB for a project, feel free to contact me and I will be happy to help you with your custom PCB as well.

About a year ago, my dad asked me if I knew a way to control a bunch of tiny LEDs to light up his model railway buildings. At the time, I didn’t have an immediate solution, but I was up for the challenge. He had a straightforward wishlist: lots of LEDs, easy brightness adjustments, and some random events to add a bit of life to the mix. Last Christmas break, I finally started. Over the course of 3 months, this became my main side project and I put way more energy into this than I expected. But the outcome is totally worth it.

Here is a video showing the entire project in detail.

https://www.youtube.com/watch?v=cVVxhGMEgmw

And here a tldr, if you don’t want to watch through the entire video:

Initially, my plan was pretty basic – a small control box with an Arduino, a rotary encoder, and a small display to control the LED’s. But then, a friend needed help with his Arduino-based LED strip controller, which used a whole different approach, involving a WiFi-capable Arduino that doubled as a tiny web server. After helping him out, it struck me that this approach was perfect for my dad’s railway lighting.

Fast forward a few months, over Christmas at my parents’ place, I brought along a WiFi-enabled Arduino microcontroller and a PWM board. With my dad’s stash of components, we had everything needed for our first prototype, and in just three months, we went from concept to completion with Bahnhofs Steuerung 2000 version 1.0, exceeding my dad’s expectation in every way.

The current state of our project is an open-source, free-to-use platform capable of controlling up to 256 LEDs with this specific Arduino model. Everything is controllable via a user-friendly and internationalized Web App, supporting multiple languages. We’ve added features like LED grouping, customizable event frequencies, and channel names, all backed by an external EEPROM for persistent data storage, even while reprogramming.

Feel free to download the project from my Gitlab repo, and I’d love to see what you create with it! And if you’re curious about the building and its history, check out my father’s blog here – it’s in German, but the photos alone are worth a look. The main building is documented here and the side buildings are documented here. He also added an entire blogbost describing the light controlling, which can be accessed through this link.

What makes me proud is the fact that the Arduino team saw this project and decided to write an article for their blog about it.

Furthermore, I won the Electromaker of the Month award with this project!

https://youtu.be/Xqe1KfE-Y3k?t=667

If you want any custom modifications to this project or have a whole different project idea in mind, please contact me using this email address.

Some time ago, I got myself new lamps for my workbench. Unfortunately and due to a mishap shortly after installing the lamps, one of the joints broke in half. Since the lamp was basically new, I just could not put the lamp in the trash so I left it broken for a few months as a future project.

This week, when I was organizing my workspace, I stumbled over the still broken lamp and I decided that it was finally time to fix her.

First I thought about my design requirements for this project:

• must be durable
• must not alter the lamp any further
• should be easy to manufacture/install
• should be easily adjustable

When I started to model the joint in CAD I had the Idea that in fact a great solution could be fixing the joint by creating an entirely new joint, with a different axis of rotation. And to make it adjustable the joint could be equipped with interlocking teeth and a big screw running through the center doubling as the new axis of rotation.

Here is what I came up with

The fix consists of 6 different parts that can be 3d printed with no support structures. Only 8 threaded inserts and 8 M4x15 screws are required for the assembly to be complete. The two halves clamp one tube each, and can then locked against each other with the large screw. This design matches all my requirements and is aesthetically pleasing as well.

The printing was trivial and only minor work needed to be done afterwards, such as inserting the threaded inserts (The 8 brass colored parts).

The assembly was straight forward and can be done in minutes

Now the lamp is usable again, with the broken part not hidden away but visible to highlight the importance of fixing appliances and not immediately throwing things out if thy get damaged.

I published all the files for this project under public domain. They are available at printables.com

If you also need to repair something with a custom part, feel free to leave me a note using this email address.

Some time ago, I saw a challenge online from the Luxembourgish newspaper RTL. They released a crossword every week, for years. Now their new challenge was to get all the results from the crosswords published to this date to win a nice price. The thing is, there were almost 170 Crosswords, and when I tried to solve one by hand it took me roughly 20 minutes. So I figured there is not a lot of competition if I can gather all the results myself.

Since I am not a linguist or a crossword lover of some sort, but a nerd, I knew I wanted to come up with a way to break the game. So I played around with it to find its weaknesses. One big weakness was that whenever a letter was typed into a case, the game would automatically tell you if that letter was wrong or right. I am not sure if that was intentional but that allowed for a very straightforward brute force attack on the crosswords.

So I went ahead and I wrote a program that took screencaptures, checked the screen for the crossword, figured out which cell to brute force and started to type a letter. The browser then automatically colored the border of the box red if the letter was in an incorrect spot. The program then takes another screenshot and checks if the letter is solved and moves on.

This was quite slow so to save me some time, I started to look for optimizations. One way to get more progress in less time was to not use the alphabetical order of characters to try, but the statistical distribution. This saved me lots of processing time and the list of crosswords solved was growing quickly.

During the development I also found that you could just brute force the cases that compose the solution. But for my own satisfaction, I nonetheless wanted to solve the entire crossword puzzles.

After two evenings of fun “solving” crosswords, I had all the results I needed and handed in my results. I did not win. But it was a lot of fun coming up with my own creative way of solving this challenge.

I did not publish the results, but the code is open source and you can have a look at it if you like using this link to my github repository.

If you also have a repetitive task that you want to automate, feel free to contact me using this email address!

Usually my projects are machines, functional parts such as tools and/or nice to have gimmicks, but only rarely for aesthetic reasons. Moreover, I then rarely find these projects interesting enough to write something about them, so most of them never get documented. But this one is different. Mainly, because it is the single most complicated part that I ever designed. Why? Because I just didn’t want to be left defeated by a … pine cone. Well yes. I spend way more time than I thought designing something that trees produce in masses.

The why is easy enough. On and off, I am working on my wall plotter project and this time around, I was fed up with the counter weights that I put on the belt tighteners. For more than two years now, I have had two plastic bottles filled with water hanging in my workroom and from day one I did not like the look of it. I always thought I would replace them sooner than later with something nice. But time went by and I just lived with the temporary solution. Then I moved, so I packed everything up (Yes, also those water bottles) and started putting everything back together again. And I really hated to put the bottles back up again. So I opened ebay Kleinanzeigen and I started looking for clock weights. I always wanted some, but they felt just too expensive as a simple counter weight. I got lucky enough and the same day, I bought a pair of used clock weights for a few Euros.

But they were too heavy. 

I was super annoyed that the nice looking weights are basically useless to me. The problem was that by chance, the water bottles had the exact weight that when powered off, the motors would neither spin forward nor backwards, without any additional breaks installed. This was super handy for long prints where I needed or wanted to power the bot down. 

Before buying another set of weights I figured that I wanted an adjustable solution, which I could make heavier or lighter depending on my needs. And since I love modeling stuff up on my computer, I started drawing shapes that I could 3D print and use as weight.

After a few hours, I knew I wanted a pine cone. I knew it should be hollow so I can fill it up with weights and I knew it should be adjustable (without a weight permanently enclosed in the print). But modeling such a shape up in CAD is really difficult. So I turned to my old love Blender and got there super quick. But in fact, with so many intersecting pieces, it was impossible to get the shape water tight (no open faces). And this is a basic requirement. Also I struggled to put 3D printable threads inside such that I could open and close the cone. So back to Fusion 360. After several restarts from scratch and almost one entire day of work, I was finally there. 

A pine cone. 

It looked marvelous and is super easy to print. Only two parts, minimal preparation and minimal clean up after print. The threads work nicely and the weight I needed could be easily achieved with spare coins I had waiting in a jar.

I also tried smaller versions at 75% scale & 50% scale to see if the clearances for the threads would work and they do.

The file is now als available on Cults3D for purchase for 1.49€

So if you have an inquiry for an intriguing shape that needs to remain 3D printable, feel free to write to me about it.

For almost two years, I was part of the team that developed the Augmented Reality (AR) game called Marshmallow Mystery. It was the brainchild of the theater group GRIPS from Berlin Germany. Their kids program “Rakete jetzt!” imagined the story of a summer camp, where mysterious things are going on. They drew the prototype for the map, did the voice acting and even doubled as the characters from the game. Their idea was that the story should become an interactive and immersive app for Android and iOS, such that other children can play what they imagined. 

When the project was outlined, their project manager approached me and asked me if I would like to volunteer as a software developer. All and all the project was supposed to be small and unpaid, but also only for a couple of hours of work that were estimated initially. But as for every good IT project, the scope became huge and the resources were sparse. This had as consequence that the project felt for a long time like it would never see the light of day.

The start however was good, lots of constructive meetings, lots of exciting ideas as well as positive energy and some early prototypes. But that motivation was only short lived. After a couple of months, the software development came to a standstill. Mostly, because the development team was overwhelmed by the amount of work it would take to get this project done, as well as the unrealistic deadlines. More and more time passed by, and fewer meetings were held by the developers because no progress was made. On the content creation side, things were looking better. Progress was slow but definitely there. The concept and story was done, the 3D models came along nicely and the audio recordings of the children sounded awesome.

After more than one year after starting the software development, not a lot had been archived. I felt really bad. So much time was invested by all of the other people but no playable prototype was in sight. Only a huge collection of assets, waiting to be turned into a game. I felt like that needed to change. After all, I said I would do it. The existing code base however made debugging near impossible so I asked the other developers if they would mind if I started from scratch and then I got going. At this point, I had no idea how much work actually had to be done to turn everything into the game that all of these people had in mind, but I was committed.

So I got to work. In the beginning only a few hours per day but then for a couple of weeks, I put all of my energy into this project as I saw it slowly coming together. The script for the game were a solid 80 pages of text that had to be translated into an enormous state machine. All combined with AR, multiple scenes, stages and different times at which content could be found and interactions be triggered.

When I started to post regular updates on my progress, the game designer and one of the GRIPS responsibles got back at me and also started helping by testing and building all the required Animations to make even better progress. 

There was one more important deadline. Since the funds that were required for this project (e.g. to print the physical maps which doubles as markers) came from somewhere, there were people expecting results as well. And after all this time, for that day, we actually managed to have a prototype ready that allowed for a complete playthrough. One and a half years after the software development had started.

After that, I had to start working full time on another project but the game designer kept the good pace going and published the game to the App Store as well as the Play Store. It also got shown off at a game fair where a lot of positive feedback came in.

Find the Android version here

Find the iOS version here

Klick here to get to the project page

In the end, more than 18 people worked for almost two years on the game & assets. 8 kids and 3 adults lent their voices to the characters. Approximately ~900 commits were done to the production repository of which ~300 were code commits, of which I did a good ~230.

So if you have an idea for an awesome game with innovative game mechanics and are looking for a committed developper, feel free to hit me up.

I moved some time ago, and one of the moving gifts was an ice cream machine. It works by crushing frozen fruits into sorbet. Quite a lot of fun to be honest. But unfortunately, after only a few servings, we found a major weak point of the maschine. The gears that are used to reduce the motor rpm to get the required strength to crush the frozen fruits are made from plastic. Especially the lower stages that turn slower and have to endure a lot of power are therefore under a lot of strain. One of these gears sheared off its teeth when some of the used fruits were probably still too hard.

Some time spent disassembling the machine and searching online for the matching replacement gear revealed that we were out of luck and this was probably a custom made gear with no available replacement parts. If there weren’t the magic of CAD and 3D printing. After designing for a bit in Fusion360, I had recreated the exact gear. To ensure a good fit, I printed it from PETG which showed that the tolerances on the first iteration were kind of tight, but the second iteration fit like a glove.

That said, I did not have a lot of trust in my 3D printed gear so I ordered the CAD model online, made from 3D printed sintered stainless steel. But since the plastic gear was already in the machine, I figured I might just use the ice maker as usual and see how much the PETG gear can take. And what can I say, the gear was working perfectly. Even so good that when the metal gear arrived I did not bother installing it because why change a winning team. 

After quite a few servings of ice cream however, the PETG gear finally gave in but in a different way than I had expected. For this gear, the teeth did not shear off but the gear split along the layer lines between the smaller and the bigger gear. But the replacement was easy and the metal gear just perfectly now

If you need a custom replacement part that is stronger than 3D printed plastic with a really intricate shape, write me an email and let’s see if we can fix another machine.

I recently changed my apartment, to something bigger and nicer. But combining two households basically doubles the things you have to stow away. 

The household room is perfect for such a matter. It allows you to put everything out of sight. To maximize its usable space, a well outplanned storage solution is crucial. With the dimensions of the room we got to work with, and the fact that this small space also needs to contain the washing machine, there was no easy and off the shelf solution available to efficiently use up all the space.

But since I call myself a maker, and my new address is conveniantly directly accross a hardware store, I knew how to deal with this kind of problem: building a custom solution. For the layout of the appartment I had already a Fusion 360 document with the blueprint and I started to design different shelf layouts.

The criteria were the following:

• Must use up the available space as efficiently as possible
• Should use mostly off the shelf components from the hardware store
• Cannot block access to the main water valve, or the electrical cabinet
• Contains the washing machine without touching her to avoid vibration
• No holes in the wall

So after playing around with different options, I came up with the following design. It is easy to build and uses the same standard board 6 times, two of the boards cut in half in the hardware store. Everything is on stilts such that the washing machine fits perfectly underneath.

I then went to the hardware store and got all the materials I planned to use

And when I came home, I started assembling the shelves

Overall, I really like how everything turned out. Even though it has little to do with my usual projects, it still shows how you can use your 3D skills (In this case the use of a CAD software) to archive different things in life. And it is so much more rewarding to claim that you designed and built your shelf from scratch than screwing something together that lots of people own as well.

If you want to plan a custom shelf or a piece of furniture, feel free to reach out to me via email using this link.

This is definitely one of my most elegant projects. For a long time, I wanted to get my feet wet with jewelry. Since my workspace at home is not suited for metal work, I knew that I had to use an external service for such a project. My father already had prototypes for his model trains made out of metal with an online service which worked great for him. This gave me the confidence to also give it a try. But since these unique metal pieces cost a bit more than your average 3D printed model, and I am not a big fan of wearing jewelry myself, I wanted something that my girlfriend would like to wear. So I explained my idea to her, which she liked, and one evening we sat together and brainstormed on what could be her next earrings. One of the first ideas was to use the 3D models of our favorite animals, which I had previously created and used in her valentines gift, and let them sit minimalistically in a fine, otherwise empty ring. This allowed for bigger earrings without lots of added weight and a good contrast of the model.

I then ordered them online at https://www.sculpteo.com/ in a process called wax casting. First, the models are printed in wax with the help of a special 3D printer, and these wax models are then casted in sterling silver. Their process is perfect for jewelry since it can replicate very fine details in a given 3D model and the silver cast comes out nearly perfect.

Only were the earrings minimally scratched when I received them, so I bought some microfiber cloths and jewelry polish from France and started polishing them up as much as possible. In the end, they looked so shiny that one person on Reddit even accused them of being fake 🙂

But of course, I could not give them to my girlfriend in a boring bag, so I had to design a cute box for the jewelry to come in. So as always, I opened Fusion 360 and modeled away. The design is quite basic and rather functional. The lid and the case are held together by small magnets that I inlaid in the 3D prints. To prevent the silver from getting scratches, I lined the inside of the box with microfibre cloth.

And one sunny day before giving the gift away, I took it with me on a walk and created a few nice pictures of the final state of the project. 

As of the time of writing this blogpost, I don’t yet know if my girlfriend likes her present but I definitely hope so since this is the most time I have ever spent on a project that I intended to give away.

If you also have an idea for a custom piece of jewelry and are looking for help, please contact me using this email address.

The Titanic’s engine project felt never quite finished for me. Even though she looked great, she did not meet one of my main criteria for success for this project, namely to run by herself on air. I tried many things, such as sanding, more or less lubrication, different brands of PLA, different valve timings and so on but I never got her to run smoothly on all four cylinders. In my search for possible solutions, I figured that one last resort would be to scale the entire engine up by a certain factor. My tolerances would stay the same, but much more air could get used to generate force to overcome friction. This turned out to be the right assumption. After printing the first cylinder at 200%, which immediatley ran without a hiccup just after assembly, I was confident that the engine would work as a whole as well.

https://www.youtube.com/watch?v=ZOpecbVx1Ww

My small airbrush compressor was by far not powerful enough to get this model going. Therefore I resorted to my vacuum cleaner. And instead of hooking the up to the inlet as in my previous air pressure engines, I connected the vacuum to the outlet. To control the speed of the engine, I had to design a pressure reducer. For this, I choose to design a tube with hundreds of small holes, which could get blocked by an overlaying piece that rotates to vary the number of blocked holes. This proved to be extremely effective and it is great fun to rev the Titanic’s engine

To add something for the engine to move, I decided to model one of the outboard screws of the Titanic as well. This was quite fun since I haven’t modeled such a complex organic shape as the propeller blades before. I connected them in the same way to the hub as they got connected on the real ship

One thing that did not work well when scaled up were the eccentric valve timing disks. For my smaller builds, they were simply concentric rings that did produce some friction, but negligible due to their overall size. When scaled up however, eight of them produced so much friction that the engine would barely turn over. I experimented with different ball bearings but came to the conclusion that adding ball bearing balls into my 3D printed parts would take less of a hit on the overall look and would work absolutely perfect. 

Here are some stats for this build:

Scale:
1:37.5

Size of the engine: 
50 cm x 30 cm x 16 cm

Weight of the engine:
3.3 kg

Diameter propeller:
28 cm

Weight propeller:
210 g

Design time:
~300 h

Print time:
~200 h

If you also have an idea for a 3D printed machine, please do not hesitate to contact me using my email address

Valentine’s Day is a special occasion that requires a unique and thoughtful gift for that special someone. This year again, instead of buying a gift, I decided to design and build a custom 3D printed gift. The idea I had in mind was an enclosure for a circular mirror. To add a personal touch to the design, I included a small bird and a fox overlaying the lower part of the mirror.

Here is the result:

Furthermore, to make the mirror more portable, I designed it to have a strong but collapsable stand.

The first step in the design process was to create vector graphics for the animals in Inkscape, which I then exported as SVG files.

Then I designed the 3D model of the enclosure using Fusion360. Finally, I imported the SVG’s to add the geometry for the animals.

Once the model was complete, it was time to 3D print the enclosure. I used my Prusa Mk3S+ 3D printer, which is a reliable and precise printer that produces high-quality prints. I chose to print the enclosure using REDLINE’S Mystic Silky PLA filament, which gives the print a nice multicolored finish.

The final result was a stunning 3D printed enclosure that perfectly encased the circular mirror. The bird and fox overlay added a unique and personal touch to the design, making it a one-of-a-kind Valentine’s Day gift.

In conclusion, designing and building a custom 3D printed enclosure for a mirror is a fun and rewarding project. Whether you’re making a gift for a loved one or creating a custom accessory for your home, 3D printing is a versatile and exciting technology that is well worth exploring.

If you’re seeking a reliable and experienced 3D printing professional for your next project, please don’t hesitate to get in touch with me.

Have you ever come across a broken appliance that you couldn’t bear to get it thrown away? That’s what happened to me recently when I found a broken desk fan outside my neighbor’s apartment. They had placed it outside to be thrown out, but I saw the potential in it and knew I had to save it. The blades had snapped off, but I was determined to fix it and make it usable again. But rather than start searching for a replacement part and buy a new fan, I decided to take matters into my own hands and fix the broken blades using my 3D printer.

First, I had to design the replacement blades in a CAD software. I used Fusion 360 for this purpose. This was no easy task, as I had to make sure the blades were the exact same size and shape as the original ones. But with some trial and error, I was eventually able to create a design that fit perfectly in the metal enclosure of the fan.

One of the biggest challenges I faced during the design process was the fact that the complete assembly of the fan blades with its hub was larger than the surface of my 3D printer. This meant that I had to split the design into multiple parts and print them separately.

First, I designed and printed the hub, which was relatively simple. Then, I modified the blade design so that they could be attached to the hub once they were printed. This required me to add a tapered base to the blades and corresponding negative shape on the hub.

Next, I printed out the 3 blades using my 3D printer with white PETG filament. This took a few hours, but the end result was worth it. The printed blades were strong and durable, and fit seamlessly onto the fan.

I must say that I was pleased with the results. Overall, this project was a fun and rewarding experience that taught me the joy of fixing things rather than throwing them away.

If you’re inspired by this project and want to discuss similar ideas, don’t hesitate to contact me.

The idea for this project came from my younger sibling. When he was visiting me in Berlin during the summer of 2022, he talked about how it would be cool to have a guitar style midi input device with clear tactile feedback. He already tried a variety of other midi guitars but he would often get wrong or missing notes with them because of the way they sense the user’s input.

Back then, we were only brainstorming for a while and with no real outcome but the idea stuck with me for the weeks to come. Every now and then, I was searching for different parts and button options to make this idea come together and finally decided to build a prototype to see if this idea is even practical. So I ordered all the parts and put them aside.

Only a few days later, I caught Corona. Not too bad though. But I knew I could not leave my apartment for a few days so it was the perfect time to get going with this project.

First I started to mess around with the buttons that I ordered, to find out if I can read them accurately with decent latency and if I could convert the presses into midi signals.

Then I started modeling up the case in Fusion360 and started 3D-printing out prototypes, to see if  the PCBs as well as all the calculated number of cables would fit.

When I was sure that I had all the big pieces figured out, I started soldering 8 perfboards together and mapped out where the buttons would go. I used a real guitar as a reference for this step. The biggest amount of work in this project was definitely the soldering job. Each guitar site has its 21 distinct notes plus one button for triggering and another button for muting the site. This means I had to solder a total of 138 buttons to the perfboard. Then I connected all the ground wires for all the buttons together with common lines and added 138 tiny cables, one for each button. Last, I soldered all 144 wires (138 button wires +6 ground wires) to the pin headers that should go into the 6 microcontrollers.

Since the ESP32 microcontrollers I used for this project were 3 digital inputs with internal pull up resistors short, I had to add external pullup resistors to these lines. 

Because I only ever wanted to connect one microcontroller to my PC, I realized that I need to connect all 6 controllers together in some way or another. The easiest solution I came up with was to daisy chain them. I used their Serial2 interface for this purpose. Each controller passes the data they receive on their RX line forward via their TX line and they also add their own midi packages to this message chain. The last controller in line then sends all the packages via the Serial interface to the PC.

Here I found that the Serial2 communication started to glitch out a lot and I therefore added pullup resistors to the RX lines of the Serial2 interface of each microcontroller to mitigate this issue.

To turn these serial messages into midi commands for my digital audio workstation (DAW), I used a method that I learned from the Switch & Lever’s video “Building a MIDI Controller Using Arduino” (https://www.youtube.com/watch?v=JZ5yPdoPooU&t). In this project, he uses the Hairless MIDISerial Bridge (https://projectgus.github.io/hairless-midiserial/) as well as Tobias Erichsen loopMIDI tool (https://www.tobias-erichsen.de/software/loopmidi.html) to convert bytes from the serial input into valid midi messages.

Last step was to finish the assembly and put everything together. To strengthen the neck of the guitar, I added a 5mm steel rod to it which I hotglued on the inside. This makes everything much more sturdy and solid. Also, I decided to add hotglue to all the connectors so that the cables do not rip off easily and also provides electrical insulation. This is definitely a trick I will use in the future.

All in all, It went together quite smoothly, and I was extremely pleased with how the instrument performs when connected to a DAW. The only thing that remains is to learn how to play it 😉

Things I would improve on for a V2.0:

• Buy more expensive buttons (These cheap buttons sometimes click but do not make good contact)
• For the lower/upper notes, I would choose a different spacing since the current spacing makes it hard to play certain chords 

Here a short video demonstration of the guitar in action

https://www.youtube.com/watch?v=vJrge6eaxFY

If you want to build your own copy, or just want to look into the code, I published everything under the MIT open source license.

https://gitlab.com/0x3b29/electronic-guitar

If you have a project that involves complex hardware and software systems and would like to discuss it with me, I would be happy to hear from you. Feel free to contact me.

This is a small project I created during my Christmas holiday 2021. It started when I found an old toy of mine and I thought, that it might be fun to recreate one that is easily 3D printable. So I checked out the springs that my dad got in his supplies and started working on it.

After two evenings of tinkering, I had a first working prototype which looked like this

The Prototype

https://twitter.com/0x3b29/status/1474484127982305281

Even tho this prototype already looked a lot like the final model, there are many fine details that changed which now make the model print better, allow the head to jump back to the correct orientation and facilitate the assembly.

The Final Model

https://www.youtube.com/watch?v=8dRado33eSc

I shared all the files for this project under the CC0 license. You can find them at the link below

>>> Link to PrusaPrinters <<<

Finally, a gallery of images that show the whole assembly process step by step:

And if you print one, I’d love to see a picture of it.

If you have an idea for a unique and exciting toy that you would like to bring to life, I would be thrilled to help you with the prototyping process. Please contact me using my email address.

The idea for this project came from a friend of mine, while we were walking over a Christmas market in Berlin. We were looking at many decorative objects and at some point she asked me if it would be possible to 3D print a design that she saw on the internet. Well, the answer was yes and the next day, I started working on the idea. First I designed a small piece to see how thin I can go with the letters, and what the required clearances for the pieces are to not fall out of the board. When I had everything dialed in, I started with a big board and followed with a set of letters.

After assembly, I noticed that it was rather hard to make the sign stand up on its own so I thought about those stands for plates and that they could hold the sign beautifully. So I went ahead and also designed one of these and the project was completed.

https://www.youtube.com/watch?v=smvL3He8y68

If you have an idea for a unique and innovative gadget that you would like to bring to life, I would be thrilled to help you with the 3D printing process. Please contact me using this email.

This is not just a small update. Almost every part of the project got redesigned and many technologies changed. This is such a major leap forward that for me it feels like a new project altogether.

After the update, the combined error was reduced from 4-5 cm / 100000 coordinates to literally zero. The speed of which the bot is able to plot at got upped from 1 dot/s to almost 4 dot/s. Last, the precision at which a coordinate can get reached on a 3m x 3m test canvas got down from ~1 cm to 0.1 mm

Here is a list of the biggest changes and what they mean to the project in no particular order

Encoded DC motors -> Stepper motors

The first version of the project used encoded geared DC motors. They had the drawback that the microcontroller needed to precisely count all the rising edges of the motor encoders. Any missed edges would result in an accumulating error. During communication and other blocking tasks, those interrupts would get ignored quite a few times which lead to an offset of a few centimeters after only 100000 reached coordinates. 

The new stepper motors however are indirectly controlled by the microcontroller via a TMC2209 stepper motor driver. This driver can calculate microsteps which leads to super smooth movements and almost no audible noise. Due to the nature of sending complete steps to the motor drive, no steps can ever get lost in the process. Also, using a library that comes with ramps, the bot can now move much faster from one set of coordinates to another, without ever under or overshooting.

Different types of wire -> GT2 belts

In the beginning, we wanted to make the bot as affordable as possible, and hence decided to go with pulleys with simple rope or wire. We tried different materials, each with its own challenges and were never quite happy with the result. E.g. nylon had a lot of stretch to it. In regimes of higher tension, the rope would stretch in a way that was not easily compensable by adding or subtracting a simple factor. Thin steel wire on the other hand always tried to clump up on the spool, thus changing the diameter of the spool in unpredictable ways which resulted in undesired patterns over the images.

The Dot Bot 2.0 uses 10mm wide GT2 belts. These belts have little to no stretch to them, and in combination with the geared and smooth pulleys, they have no slip whatsoever.

Spool of wire -> Gravity tensioned secondary belt holder

The bot was always planned to have a drawing area of approximately 10 x 10 meters. To achieve this, a lot of additional length needed to be stored somewhere. The first version of the bot did this directly on the axis that was connected to the motor. In an ideal world, every N revolutions of the motor you would assume a slightly changed diameter and everything should remain precise. But the wire and nylon always tried to spool on top of the highest point on the spool making this assumption not work at all

The new system however stores the additional length in a secondary system. This system is gravity tensioned by a weight which allows for a smooth addition and release of the belt. With this system, the belt can now be almost endless with no further modifications. Right now, each side holds 10 meters of said belt.

1 microcontroller per side -> 2 microcontrollers per side

The first Dot Bot used one microcontroller per side, to do all of the wireless communication with the server as well as commanding the motor and counting the interrupts. We found out that using Wifi on a microcontroller involves a lot of blocked time, where the processor cannot react to interrupts or produce signals to command the motor. 

Therefore, we switched to a two microcontroller system. One for coms, and the other one to control the motor. Both controllers are connected via a serial connection and with a very basic communication protocol, the secondary microcontroller only gets an absolute amount of steps to go and reports back once its duty is done. This ensures that the motor can run as quickly and as smoothly as possible.

12V -> 24V

Initially, we planned to have the stations running off 12V DC. This seemed to be a good compromise between availability of batteries and losses. But with the new motors that can draw up to 3A, 12 V was simply too little. For testing, the stations are connected to a bench power supply through 30m of copper wire and the measured voltage drop was approximately 10%. Now with 24V, less power is wasted over the lines and the stations get a much more stable input voltage.

Smaller images (A3) that were plotted while testing

Larger complete images ~2.5 m x ~1.5m

If you are looking for a developer who can help you bring your unique and creative ideas to life, I would be delighted to collaborate with you. Please contact me via this email.

When I worked on the first version of this project during my Christmas holiday 2020, it was also my first CAD project ever. This meant that I did not use many of the possibilities that a CAD tool offers and also for the sake of simplicity, I designed more of a Titanic inspired triple expansion engine than a historically accurate copy. So when I got back to Berlin, I spent more time researching the Titanic’s engines and I realized all the potential that this project still had.

In a first attempt to change everything to a more realistic model, I tried to refactor the old model but quickly realized that I needed to start from the beginning. This was also a great point for a new approach. To get as close to the original as possible, I bought myself books such as “Ocean Liners of the Past: “Olympic” and “Titanic”” and “Titanic the Ship Magnificent”. Both contain high quality plans of the engine rooms as well as the steam engines that propelled the Titanic. I imported those and other resources I found online in my CAD program, aligned everything and started modeling. After approximately 100 hours of design work, I was happy with the new engine and started printing.

https://twitter.com/0x3b29/status/1450210156961320962

The inside measures, tolerances and clearances of the engine were close to the prototype ones, but I was unable to get the completed engine to run. After lots and lots of trial and error, I figured that some PLA has better gliding characteristics than others. This was the reason I then started to reprint almost all of the moving pieces in a different brand PLA. Unfortunately, the main body and the base were still providing too much friction such that I was never able to get this engine running on more than two cylinders from compressed air.

https://www.youtube.com/watch?v=j6hWbHcUT2Q

But nonetheless, she is a beauty to look at and I am sure that I will revisit this project once again sometime in the future.

If you have a gadget project that requires specialized expertise and a high level of attention to detail, I would be excited to work with you on the 3D printing process. Please feel free to contact me.

Story:

Usually, the most interesting ideas form while brainstorming over a few beers. This is also true for the dot bot project. Although the original concept used to be the construction of a robot which uses spray cans to paint murals and was supposed to be called spray bot.

This version was supposed to only be a prototype to get going with the software development. Also most of the parts were supposed to be scrapped and reused in the final version of the project.

Luckily for the bot though, we quickly fell in love with how the project turned out using permanent markers which lead to the decision to spare this bot and simply build the spray bot from ground up. Also, we decided that we are continuing the project as a two separate hardware branches running on the same software.

This project was created in collaboration with falko.berlin

First painting

This was the very first run with a picture after everything came together. Nothing was tuned in, lots of wobble and overshooting of coordinates

Second painting

With all the learnings from the first tests (and a few more unspectacular tweaking sessions), we decided to give the bot a second go on the same picture. This time with a much higher resolution and against a sturdy wall.

And the result is quite impressive. This picture consists of 44000 individual dots that got applied over the course of 14 h. The canvas used are 4 Din A3 pages taped together. The painted area is approximately 70 cm high and 50 cm wide.

Here a few more pictures that document the states after 3 h, 6 h, 8 h and 12 h of printing

Closeup

In the left picture is one of the two stations that control the position of the gondola which is shown in the right picture

How a picture becomes a dot painting:

Preparation

• Both stations need to be installed at the same height
• The distance between both stations needs to be measured and entered into the configuration mask
• The gondola needs to be positioned at the center of the drawing area
• The vertical distance from the pen to the imaginary line between both axis needs to be measured and entered as well
• The height and width of the drawable area needs to be entered

Drawing

• For each pixel, their X and Y coordinates are taken
• Some basic math is applied to scale the coordinates to the real world’s width and height values
• Then, each side is treated like a triangle, where we know the width and the height. Therefore, we are able to calculate the diagonal distance or the rope length
• Each rope length is send via a websocket to the respective station
• Both stations try to match the requested rope length as closely ass possible
• After the stations are satisfied with their position, they send an OK to the server which, after both stations replied, forwards the drawing command to the gondola
• The next pixel is chosen from the image

Here a close up shot of the single dots that get painted by the bot:

Technical information

On the hardware side, both stations consist each of a geared down DC motor with an encoder (with high precision at 1800 steps per rotation), an H-bridge that allows us to run the DC motor in both directions at different speeds, a 12 V to 5 V converter and an ESP8266 microcontroller. The gondola consists of a power bank, a servo motor, a servo control board, a gyroscope sensor, an ESP8266 microcontroller and a permanent marker. It hangs on two strings from the stations.

On the software side, we used a node.js webserver that communicates with a frontend which has been developed using Vue.js. The microcontrollers use websockets to communicate bidirectionally with the webserver. They e.g. send their position to the server which sends commands to motors.

On the mechanical side, all the necessary parts have been designed in Fusion360 and 3D printed on a Prusa MK3s 3D printer from PETG.

The entire system has been designed in a way that every station as well as the gondola can work self-sufficient on a single battery with no cables that need to be installed. The cables in the picture were connected to a 12 V bench power supply for easier analysis but are in fact not needed. All the data transmission is wireless using the WIFI capability of the ESP8266 ESP-12F NodeMcu Mini D1 boards. This means the system can be installed in a matter of minutes.

More examples

Self portrait:

This is the first painting we produced after exchanging the nylon wire for thin steel wire. Therefore we wanted to see if the accuracy holds. With only 26 cm by 30 cm, this is much smaller than we planned to go.

Left: Plotted image
Right: Source image

Accomplishments

We are honored that Autodesk, the company behind Fusion 360, asked us to use this project as a showcase for their CAD software on Instagram.

Also, we got great feedback on Reddit as well as on Twitter by many people around the world.

• Twitter first working state
• Reddit showcase
• Reddit resulting picture

And finally, we got featured on the great show called HotMakes

https://www.youtube.com/watch?v=-pszrIKWrK8&t=835s

If you have an idea for an innovative and customized Arduino project that you would like to bring to life, I would be thrilled to help you with the development process. Please contact me using this email address.

In March 2021, I got contacted by the founder of the immersive startup VRetro. Their business case: touristic binoculars … with VR!

When we first met, a lot of conceptual and software work had already been done and their current state was that they were getting funding for prototypes and preparing for a big public test run in front of the Brandenburg Gate in the heart of Berlin. The tech demo they had on their VR headset looked nothing but spectacular. The only downside, it was hard for people to grasp how this would look and feel, not being strapped to your head but mounted on a pillar in front of a sightseeing attraction.

To fill in this gap, I started to mock up a prototype of a 3D printable enclosure for their VR hardware to have a full size physical prototype at hands. After getting their approval that I was on the right track (as well as one of their VR devices to get all the measurements right) I started to refine the prototype to a state where it would fit all of their requirements.

These were:

• Should fit their VR hardware snugly
• Recognizability that this was a pair of touristic binoculars
• Able to hold a power bank for additional battery life
• A sturdy mounting point for a standard 4/3” screw
• Accessibility of the buttons
• Some nice handlebars for good controllability

This is the final design I came up with. To overcome the size limitation of standard 3D printers, I decided to split up the CAD model into 3 separate parts that could then get joined back together with a press fit and some screws:

And here is what the first fully 3D printed prototype looks like:

As soon as the pandemic situation is getting better, they will announce a date at which the prototype can be publicly tested in front of the Brandenburg Gate. I will share the date here on this page.

If you are looking for a developer who can help you with the design and 3D printing of your gadget project, I would be happy to collaborate with you. Please contact me using this link.

2020 was quite the boring year. To ease things up a little, I bought myself a Kalimba from that online marketplace we all overused by a lot. Learning to play a new instrument is always a lot of fun, but as with every new hobby, you sooner or later get to a point where the learning curve flattens, and you cannot feel that quick progress from the beginning anymore. At that point, I thought there must be a way to help me gain traction again. I imagined some sort of trainer, maybe a robot that can show you how to play.

For a long time, I wanted to do an electronics / microcontroller project again and thought that maybe I could combine these ideas. So back to that online marketplace, ordering the “most complete” Arduino starter kit available, a whole lot of these cheap blue servos and two servo controller boards, since some quick research showed that hooking lots of servos to an Arduino is otherwise pretty difficult.

After some days, all the packages arrived and I started tinkering around with my new Arduino board and some servos. I designed and 3D printed a few basic servo mounts and tested what arrangements would work the best. But soon, I discovered the first big challenge for this project, which was finding a way to simulate “fingers” in order to play the Kalimba notes. Every attempt to 3D print a tip that could play a note repeatedly failed because of material fatigue, since I only had PLA at hand. After some research, I found which materials are suited for bending and learned that nylon is a good fit. So I got creative and designed a way to fit pieces of zip ties (which are usually made from nylon) into my printed hinges and it worked perfectly.

Some of my 3D printed “fingers”:

And the prototype of the solution that solved the fatigue problem:

From here on, the project quickly started to gain shape and I found a fantastic tutorial on how to connect loads of servos to an Arduino: https://www.the-diy-life.com/connect-up-to-992-servos-to-an-arduino-using-just-2-pins/

Next I designed and 3D printed a custom PCB mount, to avoid electrical short and with some garden wire from the hardware store (earlier, I used straightened paper clips), my first working prototype was functioning.

For the first tests, I arranged the notes and the offsets for each song all by hand, and figured that I need a more efficient way to do this. So I started looking into midi. I even started writing my own midi parser when I stumbled across a beautiful open source project called DryWetMidi (https://melanchall.github.io/drywetmidi/). They did all the heavy lifting of parsing midi files so I only had to write a transcriber to port the parsed midi to the format which my Kalimba robot was using. I noticed that, without any deep knowledge about midi, I solved the core concept exactly the same. Play note X, wait amount Y, for polyphone notes, the second note is just played very shortly after the first note which we humans do not notice.

At this point, the project was more or less wrapped up, when a friend invited me for dinner and asked me to bring the robot with me, and I had to admit that the project is pretty dull while it’s not connected to a computer. Since I had lots of pieces left in my starter kit, I thought why not add an LCD screen and a rotary encoder to the project and make it have a menu where you can select from a few songs without it being attached to any PC. To get the cables out of sight and to make it portable again, I also designed and 3D printed a custom box where everything would fit in snugly.

A few months later, I was playing on my midi keyboard and thought why not use it to control my Kalimba robot. After an evening of coding and again with the help of DryWetMidi, I was successful and now the Kalimba can be played using a Keyboard because why not 🙂

All files are open source on my github either under CC0 or MIT.

The kalimba robot with its firmware:
https://github.com/0x3b29/Electric-Kalimba

The tool to convert midi into my kalimba format:
https://github.com/0x3b29/Midi-to-Kalimba

The tool to connect the midi keyboard to my kalimba robot:
https://github.com/0x3b29/Midi-Keyboard-To-Kalimba

As an Arduino developer, I am passionate about creating unique and practical solutions for my clients. If you have a similar project in mind, I would love to discuss it with you. Please get in touch with me.

Also, during corona summer 2020, I came up with the idea to create a 3D printable magnetic chess board. Here, my intention was creating my own board that I could play wherever I was, without the need of a flat or steady surface. Furthermore, I had a very specific art style in mind.

So I started modeling everything in Blender 3D, sliced it in Cura and printed away on my trusty 3D printer, an “Anycubic I3 Mega”. For the magnets, the sliced gcode (the instructions file for the 3D printer) is post processed to have “filament change” pauses at the last layer before the cavity for the magnets gets overprinted. Furthermore, Everything is printed on a single extruder 3D printer with several files in multiple steps. This was tricky to figure out, but it is now all documented in the readme file on my GitHub where this project is available under CC0 : https://github.com/0x3b29/3D-Printable-Chess

But before attempting the first complete print, I had to tweak in the printer settings. This was when this cute little chess board got created. One thing is for sure, on this board, green always wins 😉

After I was happy with the prototype, I started the big print which took quite a few hours. Here is when the magnets in the board are printed over:

And this is the first complete board together with a Blender rendering of how it is supposed to look like:

And here are some more pictures of further boards I printed for friends:

And yes, they do get played from time to time 🙂

And the board is definitely rollercoaster proof!

As a 3D printing gadget developer, I am passionate about creating customized solutions for my clients. If you have a similar project in mind, I would love to discuss it with you. Please get in touch with me via email.

This project came to life during Christmas holiday 2020. After constructing my first 3D printable air pressured reciprocating engines earlier that year, I decided that I wanted to go further with this concept. Ever since I’ve been a kid I was fascinated by the Titanic and I thought that creating a 3D printable version of her engine would be an interesting challenge for me to solve.

To kick off the project, I started with gathering as much data as possible. Even this turned out to be more complicated than expected. But after quite a few hours, I had enough material that I felt comfortable starting the 3D modeling process. Like every other 3D printing project so far, I opened Blender and started modeling away but after only a few hours, I felt that Blender was not the right tool for the job this time. I saw my father working with Fusion 360 on an unrelated construction project and I figured that I should give it a try too.

The learning curve was quite steep since I have never used a CAD program before, but after the first day or so, I felt that this way of designing was speeding up my construction process significantly.

During the following three weeks I spent about 100 hours to model up as many details as closely to the original plans as possible but always with the idea in mind that it should be 3D printable with as little support material as possible.

Here is a short video of the finished engine running on pressurized air

And a few pictures of the engine in the snow

And a collection of parts that needed design changes 🙂

Also, here is a short video, that shows the Fusion 360 model with all the joints moving:

The files for this project are not published yet, but I plan to revisit this project sometime soon and create a complete guide on how to print and assemble each and every of the 150 parts using some superglue and about 80 M2 screws.

If you have an idea for a unique and practical custom mechanical part that you would like to bring to life, I would be thrilled to help you with the 3D printing process. Please contact me using the linked email address.

The idea for this project came up during a walk with a friend along the Spree in Berlin. We were talking about past and future projects and he asked me if it would be possible to create a high-resolution photorealistic image of his personalized turntable.

He wanted to use that image as a background for his DJ performance. For this purpose, the image should be printed on a large cloth, fixed on a frame and illuminated from the back.

I replied that the project would be feasible with Blender and that I could recreate almost everything with the appropriate reference material.

He then sent me pictures of his turntable and we looked up and bought a similar model online as good a starting point. That model tho was far from identical and so I started to change and add more & more details over the course of a few evenings to make the renderings look as close to the original as possible.

The major work that needed to be done consisted of

• Unwrapping and retexturing the entire model
• Exchanging the and retexturing the pickup
• Adding missing features such as the reset button
• Creating a custom record

The result is the following image, which was finally rendered at a 24k resolution:

The original turntable looks like this:

And the starting point we bought online looked like this:

If you need a professional and experienced Blender expert for your 3D modeling and animation projects, I would be delighted to work with you. Please contact me via email.

This was a fun little weekend project I developed sometime in September 2020. I played around with an Android App called Spectroid. Its only purpose is making sound visible as a spectrum over time. After analyzing quite some machines and drawing basic shapes while whistling, I figured that there must be a way to draw more complex shapes or even entire pictures.

I started up Unity 3D and found out that I could easily play two frequencies at a time. These frequencies would appear as bright lines in Spectroid. To draw more complex shapes than perfect lines, I had to come up with a simple algorithm to change these frequencies rapidly.

The basic idea of the algorithm is the following.

The algorithm selects the first line.

Then the two brightest pixels are selected. If there are multiple winners, two winning pixels will be selected at random.

The further left a pixel is, the lower the played frequency, the further right, the higher.

With the same speed as Spectroid draws a line, the algorithm selects the next line and repeats this process.

Here are some of the resulting images:

The code to this project can be found on my github account: https://github.com/0x3b29/Image-To-Sound

If you are looking for a developer who can help you with the design and development of your app prototype, I would be happy to collaborate with you. Please contact me by mail using this link.

This project came to life during the first months of the first Covid19 lockdown. Everyone had to stay at home and so it happened that the organizers of the cultural hackathon Coding da Vinci moved everything into the internet. I read about it on Twitter and asked a friend, who I worked with on a previous hackathon, if he would be in for another project and we agreed to take on a new coding adventure.

After a few hours of online meetings, we found two more team members and we had the idea to bring event posters with augmented reality to life. We wanted to set up a platform where users can share the pictures they took during the event, and (after a sanity check of course) these pictures would appear in AR on the event poster.

For the final day of the hackathon, a WordPress plugin as well as a Unity App were developed to showcase the imagined functionality. Anyone can upload pictures to event posters, and App users are able to explore them with their smartphone in AR.

Furthermore, we tried to get everything working with WebAR, such that no app would be needed. But all the free WebAR options were performing poorly.

For this project, our team won the category “Most technical”.

My main tasks for this project were:

• Developing the WordPress plugin / backend
• Developing the AR app

Some facts about the project:

• Pictures in the AR will always reappear in the same spot
• The backend serves low resolution pictures for the AR view and high resolution pictures for the image viewer
• A caching algorithm was put in place such that images do not need to be redownloaded in the future.

The project page: https://codingdavinci.de/de/projekte/atondes-virtual-exploration

The entire project can be found on GitLab and is licensed under CC0 & MIT license: https://gitlab.com/atondes

If you have an idea for an innovative and customized app prototype that you would like to bring to life, I would be thrilled to help you with the development process. Please contact me here.

Sometime in 2020, I planned to create a big artwork on one of my apartment walls. Since I really like the look of tape art, I decided to order some electrical tape in bulk.

When the tape arrived, I already had an artwork in mind but I realized that I could not create the details accurately with the width of the tape that I got without scaling the artwork up to a size where my wall was just simply too small. The solution to this problem was clearly narrower rolls of tape, but how to cut the rolls I ordered earlier?

After checking what tools I had on my hands, I came across several packages of spare razor blades and I figured that they would do the trick. While trying to clamp them down securely to not cut my fingers, I decided that the only safe way to do this was a custom tool.

The idea was rather simple. A solid block that securely holds one or more razor blades in a way that they cannot slip out, even if a lot of force is applied and which is also narrow enough that the cuts would be as precise as possible.

Thankfully, I was able to quickly sketch this tool up in Blender and started the first print. And the result was just amazing. Straight cuts all around the rolls, down to the cardboard, with the possibility to cut rolls to the widths of 1/3, 2/3 or 50/50.

Here is a timelapse of me, using the tape that I cut to create the Berlin transportation grid:

And a picture of the finished mural:

I later also designed and printed a fitting cover such that I can put the tool away without the danger of later grabbing into the blade and hurting myself.

As a 3D printing gadget developer, I am passionate about creating customized solutions for my clients. If you have a similar project in mind, I would love to discuss it with you. Please get in touch with me via email.

Late 2019, during a holiday, I wanted to create a simple hot seat local multiplayer board game in Unity. The idea was to combine some of my favorite games such as The Settlers, Settlers of Catan and Heroes of Might and Magic and merging some of their concepts into a single game.

After a few evenings of development, a playable prototype was ready. It has:

• Procedural map generation (Land, mountains, lakes, forests)
• Hot seat local multiplayer
• A road building mechanic
• Resource gathering
• Dragon attacks
• Merchants that produce gold which is needed to win the game

The game itself is far from finished or polished, but it is fun to play and I learned a lot during the development process.

Here are some images that show what the current state of development is:

All the project files are available under MIT on my GitHub account: https://github.com/0x3b29/HexTexMex

If you have an idea for an exciting and engaging game prototype that you would like to bring to life, I would be thrilled to help you with the development process. Please contact me at via email using this address.

In 2019, the organization team of Coding da Vinci (https://codingdavinci.de/) asked a friend and me if we could show at the Frankfurter Buchmesse (www.buchmesse.de) how cultural hackathons work. For that occasion, they asked us to prepare small projects and present them live on stage as if we developed them during the 1h presentation.

The dataset that we were supposed to use were paintings from the Jüdisches Museum FFM (www.juedischesmuseum.de). To show that you can easily combine more datasets, I decided that I also wanted to use some data I produced during an earlier (real) hackathon. Namely by separating bugs from a dataset by the Museum für Naturkunde Berlin (www.museumfuernaturkunde.berlin) which is licensed under CC0 (http://gbif.naturkundemuseum-berlin.de/hackathon/Insektenkasten/)

The idea behind this project was to create new art from existing pictures. In this case arranging bugs in a 3D space such that the image reappears when viewed from the correct angle.

Again, all the project files are available under MIT or CC0 on my GitLab Account;:

https://gitlab.com/haxorpoda/bug-scapes

Here are some Examples:

If you are looking for a developer who can help you with the design and development of your computer-generated art project, I would be happy to collaborate with you. Please contact me via email using this link.

This is one of my older projects. I decided to show it in my references since it was the first complex IT project that I have worked on and because it influenced my career quite a lot. It was my final year high school project and it counted as an entire course towards my final grade.

https://www.youtube.com/watch?v=mna5FqNtQ0w

This video is much younger than the rest of this project. My brother captured it to preserve the project as long as everything was still in a working condition.

The Project itself consisted of two major parts: A webshop where people could register and order specific amounts of M&M’s as well as a microcontroller enabled, self designed and (with lots of help by my father) build machine. The machine was basically an air powered loop which could check single M&M for their color, and either put them in a can or direct them back in the hopper. Back then, Arduino had yet to be developed and so I worked on a bare ATmega32, which I programmed using a parallel port and communicated with via a serial port. The firmware was written in Bascom, and the software on the notebook to download the orders and control the machine in Delphi. The webshop had been developed in PHP & Java Script

For this project, I was rewarded with the highest possible grade, as the second student ever to do so in the history of this course.

If you are looking for a developer who can help you with the design and implementation of your automation project, I would be happy to collaborate with you. Please contact me by my email address which can be found here.