The Basics

In principle, I needed two oscillators with electric fields. These electric fields would use the human body as an additional capacitor. The changes in capacitance would change the frequencies, which would be processed by a heterodyning mixer. Finally, the data would be graphed and stored.

The Prototype

To build my first oscillator prototypes, I utilized CMOS (Complimentary Metal oxide semiconductor) chips. The two oscillators were based upon a single 4093 chip. This circuit as seen, was placed on a breadboard along with a 9 volt battery.

Circuit 1

The chips happened to burn out later in the week. I replaced the chips and these two also failed for unknown reasons. Upon concluding that more stable chips were needed, Transistor-Transistor-Logic(TTL) chips were used. A Schmitt Trigger NAND gate oscillator was researched to replace the CMOS. This was the TTL 74LS132n chip. A Schmitt trigger was necessary for various reasons. It facilitates rapid switching of voltage allowing the oscillator to function and prevents rise/fall time limitations. Schmitt trigger chips have the hysteresis function to prevent unwanted switching by storing memory, crucial for maintaining the stability of the oscillator and preventing any drifting. To replace the CMOS 4077 mixer chip, an XNOR oscillator was selected. This was the TTL 74LS266n chip. Because TTL chips require less power, a regulator was necessary to drop voltage to 5v. The lm7805 regulator was selected because of its 1.5 ampere current capability, internal current limiting mechanisms, and thermal shut-down feature that allows it to be immune to accidental overload.

The regulator was inserted, yet no signal was received. After repeatedly checking the wires and components, I knew that it was not a circuitry problem. Upon much pondering, I concluded that the problem was the resistors. TTL chips tend to draw more power and I changed resistors to compensate. I added new resistors and received an output.

The next problem was even more confounding. The frequencies of both oscillators were exactly the same. I tuned the variable oscillator but the frequency changed only slightly. Both oscillators seemed to be working as one, by leaking frequencies through the power lines. To fix this problem, I operated the oscillators on separate chips. Then, I placed decoupling capacitors across the chips as well as on power lines. This would provide temporary power to the chips and cause a slight delay that would prevent syncing.

The Final Design

Implementation

I tested various different capacitors to find the best capacitance. A wide range of frequencies was obtained, but the best frequency was made by the 1nf capacitor. It provided the best sensitivity. I implemented the ThereNIM by hoisting antennas from the ceiling. Two antennas were used to account for sleeping on various sides. An L-plate antenna was used for bi-axial monitoring. The patient can be monitored when sleeping on any side. The data was plotted with an Amplifier connected to the visualizer program on the computer. The graphs were recorded over night.

Graphs from ThereNIM

All images on: https://googledrive.com/host/0B-45o0GIQ9ZQb0NMNFdUcDFsWFU/