My name is Grant Trebbin and I'm electronics and technology enthusiast from Brisbane, Australia, and I'm looking for work. Completing three years of an Electrical Engineering degree at the University of Queensland is one of my proudest and most enjoyable achievements. I find pulling things apart to see how they work preferable to actually using them in most cases, and am transfixed by anything technical.
Currently my job doesn't involve anything remotely technical, and that's something I desire to change. I don't have a degree, but through both my formal and personal studies, I have an extensive technical knowledge base to draw on, which I believe is wasted running a Woolworths meat department. Have a look around at some of my projects, and if you know of any full time permanent electronics or related technical work in Brisbane that I'd be suitable for, I'd love to hear from you. My contact details are at the bottom of this page or in my resume above.
Studying Electrical Engineering with a major in communications and biomedical engineering means I have a good foundation in RF and signal processing technologies and their implementation in hardware. I've also had experience using Python, C, and MATLAB to perform image and signal processing operations, run simulations, and program microcontrollers.
Portfolio
I've selected 10 posts from the blog that demonstrate my skills and highlight the type of projects I like to work on. There's a mix of theoretical and practical topics that use everything from handtools to software. I hope you have as much fun reading them as I did creating them.
1. Harmonic Elimination Pulse Width Modulation
The HEPWM process generates special waveforms that control or eliminate harmonics in a switched waveform. This can be incredibly useful in power generation where DC needs to be converted to an AC waveform. Being able to control the harmonics means you can reduce the size of physical components like filters. Out of everything I've done on the blog this is the work I'm most proud of. It's incredibly niche and esoteric, I guess that's why I love it. When you Google the topic I'm one of the leading results, not because I invented the technique, far from it, but I think I've helped to make it more accessible for others to use in their own projects. I wade through the Fourier Series so you don't have to.
For some inexplicable reason I became interested in how the focus coil in a CRT tube works and wanted to simulate it, but to do that I'd need to be able to determine the magnetic field from a current loop. Most physics students have calculated the field magnitude on the axis of a current loop, but if you want to know the magnitude of the field anywhere else the maths becomes incredibly complex. To explain the derivation of the mathematics involved in the situation I put together a small animation. I'm incredibly proud of this video. Not only did it take a lot of work to get done, it's my most popular Youtube video with almost 6000 views at the time of writing this.
While researching magnetic fields I came across a quadrupole configuration that's used to focus beams in a synchrotron and was curious how it worked. Basically, two specially designed electromagnets are placed in the path of an electron beam to focus it in one plane and then the other. Using some cobbled together code I was able to demonstrate this in Octave. The video below shows how the beam spreads in one plane, but focuses in the other while passing though the first lens and then converges after passing though the second.
I wanted to test out the new MCP9808 temperature sensor from Microchip, and needed a breakout board. Up until this point I was hesitant to use any surface mount components, but decided it would be a good learning experience to make a small board with some input protection for the sensor. The temperature sensor comes in two different packages, an MSOP and a DFN. I wanted to challenge myself by mounting the DFN, but wasn't sure if I could. To be sure, an alternate footprint was placed on the board for a MSOP package. It wasn't needed though as I was able to mount the DFN. This was also my first experience using Kicad.
Manually placing surface mount parts on a PCB can be difficult at the best of times, but if you're assembling a high density board with small parts, tweezers can be problematic. With this in mind I wanted to create a manual version of a pick and place machine. A custom made brass hand piece is connected to a vacuum pump controlled by a foot switch. The tip of the hand piece has a luer lock connector so different tips can be easily attached. Depending on the tip used, you can then pick up a small PCB or as in the image below, an 0603 resistor.
A while back, a local supermarket released a set of collector cards with pictures of animals on them. As an added feature they had a bar code on the back that allowed them to be swiped through a device that played the sound the animal on the card makes. I was intrigued and did a small teardown of the device. How the card number is encoded in the barcode was also reverse engineered. I went one step further and desoldered the memory from the board and read it into a computer. Unfortunately I was unable to determine how the sounds were encoded. My original goal was to replace the audio samples with clips of Ralph Wiggum saying strange things.
My father has some mobility problems and needs help getting around. He has a winch to raise and lower him out of the bath, but the switch that controls it wore out. Getting a replacement was near impossible due to the strange nature of its operation. It allows the winch to be connected to a battery in a forward and reverse direction, but when in the centre position it shorts out the motor, effectively braking it allowing him to get on to the platform. It isn't a safety critical situation as failure only causes the platform to slowly lower when weight is placed on it. To replace the switch I came up with a solution that uses relays to replicate the dynamic braking operation.
A long time ago I had a crazy idea to calculate the time it took to get between certain locations in the city of Brisbane. I started by writing a program to visualise the data. The project didn't really go anywhere. At that time the public transport data I needed wasn't easily obtainable and I had enough trouble writing my visualiser. I'm pretty happy with the visualiser though. It's written in C++ and uses OpenGL to render map tiles as you move about the city. The level of detail of the rendered tiles depends on your location, height, and in what direction you're looking. Data from the Open Street Map project was used to render the map tiles. Unfortunately the program was lost when I upgraded a computer. I've since learned about backups and version control. ;-)
After having a lot of trouble with a pump connected to a rain water tank at home, I was relieved to finally find a solution to the problem after a couple of failed attempts at fixing the motor. This was a case of troubleshooting and learning at the same time. I had a basic understanding of induction motors but this forced me to learn a bit more.
This wasn't a post I put too much thought into, but as it turns out, it's become my most popular. While building a chicken coop for my sister I needed to join some PVC pipe to a fitting that was a different size. The only things I had on hand were a heat gun and a coke bottle. It's here for completeness and to show that sometimes simple solutions to problems are often the best.
Expanding the end of the pipe
Fitting in place
Get In Touch
The best way to contact me is via the email address in my resume, but I'm reasonably active on Mastodon as well. Using Facebook, or YouTube is fine too, I may just take a little longer to respond.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.