I have a Dallas Semiconductors 18B20 temperature sensor on a RobotLinking board. It's part of the RobotLinking 37 Sensors Kit. The board has a LED to indicate activity. It resembles the attached image tempSensor.jpg.
Here's what's weird. All the guides I can find for the 18B20 say that the pin order is GND, DQ, VDD. That should mean pin 1 goes to GND, pin 2 goes to a GPIO, and pin 3 goes to 3.3V or 5V (this one can take either). Yes, I am looking at the sensor's pins from its bottom.
If I do this, the board's LED is lit solidly. I get 0 degree readings, the same as if no device was connected. (Infinite resistance?)
To make this work, and get valid temperature readings, I have to hook it up this way: sensor GND to the GPIO, sensor DQ to GND, and sensor VDD to 5V or 3.3V. The LED only flashes when the temperature is actually read.
Here's a twist: if you look at the circuit traces on the RobotLinking board, it looks like the paths between the male breadboard connectors and the sensor's pins 2 and 3 are reversed. This means if you think you are connecting to pin 3 via the male connectors, you are in fact connecting to pin 2, and vice versa. The path from pin 1 on the sensor to the male pin 1 where it plugs into the breadboard is straight, so pin 1 probably means pin 1.
When I described the above connections I made, I accounted for this apparent circuit pathing weirdness.
Another twist is in the manual, the sensor is soldered to the middle leads. My sensor is instead on the furthest leads from the pins, like in the attached image.
The manual with this kit didn't explain any of this. I'm at a loss to understand my observations.
