Recently Browsing 0 members
No registered users viewing this page.
In this thread I will endeavour to build an MQTT system to read ambient weather conditions outside and in the observatory and control the roll-off-roof etc.
I recently posted my design for a weather station in this section.
As I had bought several pressure/humidity sensors, as well as ir temperature sensors and ESP32 development boards, I wondered how small a weather monitor could get. The sensors are quite small, and so is the micro controller. Such a weather monitor wouldn't incorporate wind speed measurement nor a rain detector, since these take up more space. But otoh, there is seldom rain without clouds, so if you detect clouds, you should be safe.
Here it is, a miniature (9.8 x 5.9 x 2.7 cm) weather monitor. The device has built in wifi, is powered from a micro usb contact and is compatible with the INDI Weather Watcher driver.
(the mat underneath has a 1 inch grid pattern)
plastic box 9.8 x 5.8 x 2.7 cm a piece of V-board, in my case with copper islands rather than strips ESP32 development board with male headers MLX90614 ir temperature sensor with I2C interface BME280 environmental sensor with I2C interface micro usb cable and power adapter, or a powerbank for wireless operation Assembly is really easy and involves drilling a hole in the box, soldering the components in place and wiring to the ESP.
The finished monitor in place. As this is a box with a click lid, I used silicone to seal it. The holes on the sides and bottom are drilled at an angle to keep rain out.
As I built it, the electronics will heat the BME slightly, and because it is mounted inside the casing, it will be slow to reach ambient temperature should this change abruptly. Adding more holes near the ESP would take the inside temperature down. Otoh, temperature readings don't have to be that accurate, and you could use the MLX ambient reading for more accuracy.
Here's the INDI control panel for the weather monitor
(Wind and rain are simulated, because I was testing the driver when I took the screen shot)
Here is how it looks in Ekos scheduler. The red marker indicates that weather conditions are bad. In this case clouds = 100 %. If the tickbox next to "Weather" is checked, Ekos will allow weather conditions to control an imaging sequence.
And in the ROR driver
(I know it says Dome, but the ROR driver is derived from the dome driver, and it's still under development. Besides, this is the simulator driver.)
The code for the esp is on my github page:
You need the files:
bme280.py mlx90614.py boot.py (replace the ssid and password with your own, or comment/uncomment lines to create an access point) main_mini.py (which you will have to rename to main.py before uploading to the esp board)