Low Altitude Temperature Profile

This project is to make a mathematical model of the temperature at low altitudes, and how it changes as the sun warms the earth. I’m basing the model around the one dimensional heat equation. Here's the output from the model with cold boundaries and a hot initial center, linearly distributed along the left wall (time is to the right):



The actual writing of the model was about as exciting as coding usually is… pretty boring to hear about. Thus, I’ll describe that process and how it works when I get a little more time, but at the moment here are the stories of testing the model using the rockets.

I started with 2 rockets, I currently have none. Thursday we went out and launched the first rocket without the computer in it to make sure it went where we wanted. It didn’t. In the past I’d used rockets with 4 motors and two stages, but separated, on either side of the body. As you can see here, this was not the case on the test rocket:



Here the bottom two motors were taped together… this turned out to be a bad idea. Apperently the motors do not burn at the same rate exactly, so when one finished and ignited its other stage, it was cast off… while still being attached to the other. Thus the bottom motor ripped off before it had lit its upper stage, leaving the rocket underpowered. Besides that, the launch rod we used in the test wasn’t nearly strong enoough, and the rocket ended up leaving the pad at about a 45-degree angle. This wasn’t so bad until the previously-explained stage switch, in which the sudden change in acceleration made the payload and thus the CG shift forward.This shifting mass caused the rocket to further rotate, and when the (single) upper stage motor fired, the rocket was pointed horizontally along the ground, about 200 feet up. The rocket continued missle-style for the last 3 seconds of the burn, then found its way into the trees at around 250 MPH. When we found it, there were peices strewn in a 25 ft radius of the actual payload.



The payload was intended to split from the body at the end of the burn. Unfourtunatly it got stuck, and the pressue inside the fusalage went instead out a different way: through the sides. So the payload came down with a section of fusalage, parachute still stuffed inside. As it went through the trees, the fins were stripped off. The wooden nose cone must have hit a tree, because I’ve never seen dents like that in such a large piece of wood…

This left us with 3 lessons to learn for the “real deal”: 1) Don’t strap the bottom stages together, 2) Make sure the payload detaches smoothly, 3) use a much stiffer launch rod. We did these things, and the first (and unfourtunatly final) launch went smoothly, though we had to knock it over about 30 degrees due to high wind. Thus the final altitude was around 2500 feet. As the rocket fell, the fusalage detached from the payload, and its wherabouts are currently unknown. However, the payload was succesfully recovered and read about 3/4 miles from the launch rod, near the intended landing spot:



Taking into account the issues the previous models had, I built a new rocket that was meant to be the final, perfect launch. This time rather than having such a wide and heavy body with a full BX24 motherboard in it, I soldered my own board, which included an FM transmitter for tracking. It was predicted to use half the engines (3 stages, no clusters) and go 120% as high.

As I was testing the board I built, I noticed an issue. The temperature sensor was probed every half-second, and the radio was pulsed every second. When the two happened to coincide (a reading while the radio was transmitting), the value read by the computer was very far off. I did not originally make this connection, and thought my board was simply bad. I hooked a probe up to the positive of my battery, and poked the pin that was reading the temperature sensor. This should have returned a full HIGH value, or a 1024. Instead, the chip crashed, and never came back.

Woops… I had forgotten that the battery (9v) was going into a 5v regulator before the computer. Thus, the highest voltage expected by the chip was 5v. All the manuals with the chip indicate that reverse polarity will instantly kill a chip. When I applied the 9v, the comparator tested it against 5v, allowed current to flow backwards, and beasted the poor device.

It was 3am on launch day, and I didn’t have another chip. I decided that I might as well launch the rocket, because rockets are cool. I was rewarded with the most awesome landing yet:



Epic!