While I've got years of coding experience and a good basic understanding of electricity my digital electronics experience began a few weeks ago when I picked up this amazing little device called an Arduino. It has placed some of the ideas I've been kicking around in my head for years, within my reach.
I have a question about the temperature sensor I'm using that will relate to most of the sensors I wind up using. It is one of the little transistor-shaped temperature sensor triodes and has TMP 36GZ printed in miniscule letters on the back. I've got it wired into a basic thermometer displayed on a Hitachi 2x16 LCD.
Two things:
First, the displayed temperature jumps around a lot. I was wondering if this is noise on the reference voltage or noise in the sensor, or both. Is there a preferred way to stabilize the measurement or am I looking at implementing a digital filter.
Second, when I turn on the LCD backlight the temperature goes up about 10 degrees. It has nothing to do with an actual temperature change so I'm guess it is pulling down the voltage on the bus or maybe over-taxing the current specs of the Arduino. What's the best way to handle this? Would I use a transistor to maybe drive an amplifier or something?
Thanks! This little board is one of the slickest things I've seen in a while. Endless possibilities!
I posted the same issue a few hours ago! Mine is changing 10 degrees when I write low to a digital pin on a relay board. I'm guessing the answer is adding components somewhere but I have no idea what or how.
How do you power your Arduino and LCD ?
For stabilizing temperature readings, you can take, for example, 100 samples then calculate average. It will smooth the reading.
waski:
How do you power your Arduino and LCD ?
For stabilizing temperature readings, you can take, for example, 100 samples then calculate average. It will smooth the reading.
I have the Vilros Starter Kit which provides a small breadboard. I'm running the Arduino from a USB port and powering the breadboard (and the LCD) from the 5V Arduino pin. Maybe I should power the LCD from a separate source? You're laughing at me, aren't you?
I'm using a related temperature sensor, and find that the output noise is about 2 degrees in amplitude (looking with an oscilloscope).
A combination of low pass filtering and software filtering (simple averaging is ok) gave me the stability I needed.
You are probably seeing the AREF voltage sagging when you load the supply with the LCD backlight, as well as some ground drift with load.
Using an external supply for your load will help, but gets tricky when you want the instrument to be stand alone. Using an external reference voltage (AREF) is better, but read the caveats in the documentation so you don't let the smoke out.
Everyone makes simple mistakes - allways 