introduction

A tricorder is a portable sensing / scanning device first featured in the science-fiction TV series Star Trek (1967), which is set roughly 200 years into future. It follows humanity's first forays into deep space, exploring strange new worlds and meeting new life and new civilizations.

When visiting an uncharted planet, the protagonists want to measure several quantities such as the composition of the atmosphere, levels of ambient radiation, element in the soil, and scattering of light in the sky (and whatever else the episode's plot requires). To enable this scientific exploration, the aim of the tricorder device is to sense as many physical quantities as technologically possible in a handheld, battery-operated form factor. 

Tricorders have been depicted in three broad categories: medical, scientific, and engineering. This project deals with the scientific and engineering versions.

A few things make this particular implementation stand out:

• Sensing ability: The number of physical quantities it can measure, which is more than any project made before Detailed info ⎋

• Ease of use: Designed with the principle "If you can use a smart phone, you can use this tricorder". Uses the same intuitive touch functions (like swiping) and display elements (like icons) that are common in smartphone interfaces. There are only two fixed buttons: a "home"/"back" button and another that shows a "quick menu" which can be used for functions like taking screenshots, adjusting display brightness, or telling the device to go to sleep. These buttons are embedded within the LCARS graphics, and everything else is context specific. There is also the top bar which shows time and device stats like wifi name, display brightness, and battery percentage. It can be woken from sleep by touching the capacitive touch screen.

• Graphics hardware: As best as I can tell, this design has the highest density display ever put into a functioning tricorder project. Stats: 60 FPS max frame rate, ~254 pixels per inch,  720x720 px resolution,  18-bit color (!), 4.0" screen size, and a quirky square shape (Hyperpixel 4.0 Square w/ Touch). Uses the high-speed DPI interface for communication, instead of the slower and more common SPI interface. 

• Visualization quality: All real-time sensor readings are presented visually, many with customizable visualization options. Carefully coordinated color schemes and transparency make it significantly less of an eye-sore than previous UI designs. Other fine details, like anti-aliasing the shapes and using hardware-assisted frame buffering also improve the experience. Plotting is done mainly by using matplotlib and seaborn, which are the same libraries used for visualizing academic/scientific research. 

• Independently usable modules: The sensor module can be used independently in a pinch, or whenever I don't care about visualization. It has its own battery (2,500 mAh LiPo), two OLED screens (128x64), and four capacitive touch pads that allow menu navigation like using arrow keys on a keyboard. The battery can be charged while simultaneously being used by the Raspberry Pi, either by direct USB connection or separately by a wall outlet (Bluetooth). The Raspberry Pi portion handles the optical (IR) camera, Software Defined Radio (SDR), and data visualization. It has its own battery (10,000 mAh) with pass-through charging so it can be recharged while using the Pi.

These are just my educated guesses, since I'm not an economist or scientist:

• I suspect not enough people are interested in amateur measurement and exploration to justify the cost of R&D  and/or manufacturing. A mass-market product can't be cheap until large numbers of units are produced, but this is only possible if many people buy it in the first place. In this situation, economies of scale cannot be adequately established and the profitability (ultimate goal for companies) will remain questionable. This seems to be a version of the inventor's paradox.

• For any serious scientific/industrial measurement, it would be easier and almost certainly more accurate to just use separate devices such as range finders, microphones, thermal cameras, etc. In an integrated device, the sensors can (and do) interfere with each other and with the components of the device itself. For example, smartphones generally cannot tell you the temperature in your room because the heat from its internal components interferes with accurate measurement. Another example is smart watches not being able to determine accurate compass heading if the user is wearing a magnetic wristband. Work-arounds to such problems could probably be developed, but most buyers would not appreciate the added cost for something they don't really care about, and quite often don't want to care about.

A lot more than I had expected.

The technical side:

Making GUIs, working with vector graphics, designing (and procuring) PCBs, powering circuits with batteries, Bluetooth and BLE communication, standardized Human Interface Devices (HID) communication, standardized GPS communication, history and theory of measurement, to name just a few. A renewed appreciation for how integrated and monolithic the design of portable computers is (after much thinking, I concluded that a monolith is the most practical shape for this device). 

The human side:

During the course of testing (or bragging), I showed this device to a number of people from different walks of life, having different educational backgrounds and interests, and having varying levels of curiosity about technology. Some of the things I learned:

1. Surprisingly few people know that white light is actually made up of equal proportions of the visible colors, each with their own distinct wavelengths. Even more surprised to learn this is ultimately why the sky is blue and stop signs are red.
2. Even fewer people know what temperature really is (vibration of atoms), or that it is differs from one spot to another even if the spots are physically close.
3. Not everyone knows that normal glass blocks UV light. 
4. Barometric pressure increases before it rains (this was new to me too) and then settles back down to an equilibrium level determined by one's elevation above mean sea level. Additionally, this information can be combined with relative humidity readings to determine the absolute humidity of the air, i.e. how many grams of water are present in each m^3 of my air.