# Sun-Seeking Sundial Self-Calibrates In No Time

Sundials, one of humanity’s oldest ways of telling time, are typically permanent installations. The very good reason for this is that telling time by the sun with any degree of accuracy requires two-dimensional calibration — once for cardinal direction, and the other for local latitude.

[poblocki1982] is an amateur astronomer and semi-professional sundial enthusiast who took the time to make a self-calibrating equatorial ‘dial that can be used anywhere the sun shines. All this solar beauty needs is a level surface and a few seconds to find its bearings.

Switch it on, set it down, and the sundial spins around on a continuous-rotation servo until the HMC5883L compass module finds the north-south orientation. Then the GPS module determines the latitude, and a 180° servo pans the plate until it finds the ideal position. Everything is controlled with an Arduino Nano and runs on a 9V battery, although we’d love to see it run on solar power someday. Or would that be flying too close to the sun? Check out how fast this thing calibrates itself in the short demo after the break.

Not quite portable enough for you? Here’s a reverse sundial you wear on your wrist.

# Measuring Magnetic Fields With A Robotic Arm

Learning how magnets and magnetic fields work is one thing, but actually being able to measure and see a magnetic field is another thing entirely! [Stanley’s] latest project uses a magnetometer attached to a robotic arm with 3 degrees of freedom to measure magnetic fields.

Using servos and aluminium mounting hardware purchased from eBay, [Stanley] build a simple robot arm. He then hooked an HMC5883L magnetometer to the robotic arm. [Stanley] used an Atmega32u4 and the LUFA USB library to interface with this sensor since it has a high data rate. For those of you unfamiliar with LUFA, it is a Lightweight USB Framework for AVRs (formerly known as MyUSB). The results were plotted in MATLAB (Octave is free MATLAB alternative), a very powerful mathematical based scripting language. The plots almost perfectly match the field patterns learned in introductory classes on magnetism. Be sure to watching the robot arm take the measurements in the video after the break, it is very cool!

[Stanley] has graciously provided both the AVR code and the MATLAB script for his project at the end of his write-up. It would be very cool to see what other sensors could be used in this fashion! What other natural phenomena would be interesting to map in three dimensions?