State of Wind Sensor Instability

While reviewing the telemetry, I noticed a pattern of erratic State of Wind readings that correlated to when I was working on the clock. This was isolated to the LED case lights being on. The State of Wind is measured using an Sharp IR distance sensor, which sits under the pedestal table top looking downward looking at the top of the weight. In addition to sensitivity to input voltage noise, these devices also have significant spike in output voltage due to a ~40mA inrush current when the IR LED fires. I added a 10uF ceramic capacitor between ground and VCC and between ground and signal, and a 49uF electrolytic capacitor between ground and VCC. Together they function as decoupling capacitors and a simple low-pass RC filter and drastically reduce the EMI. Furthermore, in software the State of Wind is now set by taking 5 readings and averaging the middle three. There is now no discernable difference between the readings with the case LEDs on or off and the readings are very stable.