Projects

One of the best ways to learn is by doing, and thats what I've done here. The goal of this project is to learn more about compressors, and ultimatly find a compressor design that best suits my future ambitions of building a small turbojet engine. This project serves as a helpful tool for gaining insight into complexities of integrated systems, while also being usefull in the long term.

The compressor shaft is driven by a 750 watt BLDC motor, which is reduced to spin the shaft up to 10,000 rpm. The front of the wind tunnel is designed to be removable with two screws, providing access to the compressor blades. I designed the blades to be forward-mounted so they can be easily swapped out for testing. The motor and electronics are enclosed in a housing to prevent debree entering, and to protect the operator from any catastrophic failures.

There are a multitude of sensors to collect preformance data. RPM is measured with an optocoupler that reads an encoder mounted to the motor shaft. The RPM provides closed loop feedback for speed control. The speed is adjustable in increments of 100rpm, while actual rpm prints to the serial port. Current is measured with a hall-effect module - this measures the current into the motor driver. The output power of the motor is estimated using manufacturer-supplied data on motor efficency. Both total pressure and static pressure are measured using seperate pressure probes. The probes are placed about halfway into the tunnel to measure close to the average.

When I started printing with low-shore TPU (60A, 85A, 90A), I was constantly annoyed with having to set up my printer to top-feed the filament every time. This project is my solution to this problem, and funnily enough, just after I had finished the first draft of my design, Bambu Labs announced their own version!

This module functions by ramming the TPU filament into the PTFE tubes. In traditional printers, the filament is pulled through the tubes by the extruder drive. This doesnt work with TPU because of its high friction in the PTFE tubes. By ramming the filament instead of pulling it, it reduces the friction, allowing the extruder drive to pull the filament into the nozzle.

The physical design has gone through 3 itterations, and I unfortunatly didnt take picutres of the first itteration, but the fundimental design hasnt changed. The filament is pushed by two wheels lined with TPU to increase the friction. The wheels are mechanically linked, but can be released with levers to unload the filament. The ramming speed requires precise calibration; too little and the extruder drive cannot pull the filament, too much and the filament kinks in the tubes and becomes clogged. To solve this problem, I incooperated a wider space in the filament path, which allows the filament to start kinking early. An optocoupler detects when the kinking starts and shuts the ramming motor off. Once the extruder pulls the filament straight, the motor starts back up.

This is a collection of mini-projects I've done while working at the UVM FabLab. This projects range from general organization, to specific machine upgrades.

The FabLab recently got a new filament recyling system. It grindes up discarded prints, melts them down, and spools new filament. The process has proven difficult to master, however. Many variables influence the extrusion rate, and with an acceptable tollerance of 0.05mm, getting usable spools has proven hard. I've managed to control one of these variables, the feed-rate, by designing a mixer for the scrap hopper. This addition has reduced diameter variation by ~0.5mm, but more work still has to be done.

The UVM Physical Plant placed an order for a new part after a cooling fan for an electric motor failed. I was given the melted part, tasked with replicating it in CAD, and printing out a new one. I simulated the loading and temperature conditions using FEA, modifying the part slightly to increase its strength. I was able to create a near-exact replica, and the replacement part has been in use for nearly a year

This is a quick overview of a special slicing technique that greatly improves the strength and rigidity of printed parts. I found this method while researching orca-slicer, and have since tuned it to be more reliable and even stronger.