I think about the engine as one continuous path: intake, compression, combustion, turbine, and exhaust. Every section depends on the one before it, which is what made this project much harder and more interesting than designing isolated parts.

The spinner centers incoming air and hands it to the first compressor stage without turbulence. The inlet lip radius matters more than it looks; I learned that a bad curve here shows up as instability everywhere downstream.

The compressor made small dimensions feel important. Blade pitch, spacing, and tip clearance decide whether I actually gain pressure or just build a very complicated fan.

The combustor forced me to think about two opposite goals at once: keeping a stable flame while protecting the metal around it. That tradeoff changed how I thought about airflow and materials.

The turbine has to recover enough energy to keep the compressor turning without taking more from the exhaust than it needs. It is the part that made the engine's energy balance feel real to me.

The reheat section is a direct trade: more fuel for more exhaust velocity. Designing around a fast exhaust stream made flame stability and temperature control the questions I kept returning to.

The nozzle is where the work upstream becomes useful exhaust velocity. I like this stage because pressure, temperature, and geometry finally show up as something I can observe directly.

The idea started with a simple observation: light becomes visible when it has something to scatter from. I want to use that effect to place enough bright points in space that a shape feels physically present.

I want the display volume to stay open, so I use thin sheets of moving air instead of glass walls. The airflow contains the mist while still letting a hand pass through the drawing.

Three ultrasonic misters in the base atomize water into a fog fine enough to hang in the air. Density is a tradeoff I am still tuning: too thin and the dots are dim, too thick and beams wash out before they cross.

I placed diode modules around the chamber so the full volume stays reachable. Their default alignment passes through the center, which gives me a repeatable reference while I tune the steering.

The galvo mirrors turn a fixed beam into an addressable point. I found that accurate timing matters more than simply adding power, because a brighter dot is still useless if it appears in the wrong place.

I chose a rotating wireframe cube as the first target because it makes alignment errors obvious. If the twelve edges stay straight and connected, I know the steering and timing are close.

The glove tracks orientation and finger movement, then small vibration motors respond when a fingertip crosses the cube. I want the interaction to feel immediate even though the object itself is only light.

This design uses motion to turn one surface into an apparent volume. My job is to light each point at the exact angle where it belongs, fast enough that the image appears to stay still.

The motor and base have to hold a steady speed without shaking the image apart. Balancing the rotor and keeping the shaft aligned are just as important as the electronics driving it.

Power and data still have to reach the LEDs while the rotor is moving. I use a slip ring to cross that rotating joint without wrapping wires around the shaft.

I use a thin double-sided LED blade so the display refreshes the volume twice per revolution. Keeping that blade balanced and aligned is just as important as controlling the pixels.

The encoder gives me one trusted reference each revolution. Every voxel is timed from that pulse, so even a small amount of drift shows up as a visible smear.

When the timing is right, the moving surface becomes hard to notice and the lit points appear to hold their position. That is the moment I am working toward with this prototype.

I designed the clip to keep the emitters and detector aligned around a fingertip while staying simple enough to test repeatedly. The signal processing happens on the small board above it.

I use red and near-infrared wavelengths because blood and tissue respond differently to each one. Comparing those responses gives me more information than a single LED could provide.

Light does not travel cleanly through a finger; it scatters through tissue before reaching the detector. That messy path is exactly why calibration and mechanical alignment matter so much in this project.

The photodiode produces a very small current, so I amplify it before digitizing the signal. Most of what I measure is a steady background; the pulse is the small changing part I need to preserve.

I compare the changing and steady parts of each wavelength instead of trusting raw brightness. The ratios reduce some device and tissue differences, but they do not remove the need for careful calibration.

The readout lets me inspect the waveform while I test the sensor. My focus right now is repeatability and comparison against reference measurements, not pretending a prototype is already a clinical instrument.