Compost Monitor  (ESP8266 + MQTT)

Summary

Here is a quick little temperature monitor made to log compost temperature over time.  The monitor consisted of an ESP8266, two DS18B20 temperature probes and a battery.  I used Adafruit's IO service for publishing the  compost temperature and the outdoor air temperature.

Purpose

When hot composting, the temperature can be used to determine if maintenance is needed for the pile. The microbes inside generate heat, so if that temperature drops the pile probably needs to either be rotated or might be getting dry and need additional water.

I find it fascinating that a properly setup compost pile can reach temperatures approaching 150 degrees Fahrenheit. So, this project was done to provide a nice live website interface that showed the current temperature and the temperature plotted over time. 

DS18B20

The DS18B20 provides a couple of advantages to other temperature sensors I have used in the past. The first advantage is that you can run multiple devices on the same data bus. So 1 pin, 1 wire, etc.  for several temperature probes.  The other large advantage, is that you can easily and cheaply acquire the DS18B20 in an already sealed package that is "water tight". I do not know how water tight they actually are, but good enough for the compost pile for sure.

"Thirsty Microbes"

I started composting during the COVID-19 pandemic, and had built a few hot compost piles  before building this monitor. On day 2 of a new pile, I expect the temperature to be at least 140 degrees. In this pile, it was almost flatlined at 110 degrees. The only problem that I could think of for the low temperatures, was that maybe the pile was too dry. So I drenched the pile with a lot of water. I expected the pile temperature to drop for at least a day after adding that much water, but much to my surprise within 2 hours the pile went from being around 100 degrees to 150 degrees! I'm still amazed by this, and had assumed it was a fault in the sensor initially until my analog thermometer (and hand) confirmed that yes, it was indeed - really really hot. I'm still not quite sure if temperature is just the result of increased thermal conduction of heat that was already in the pile, or if the water rapidly increased microbial activity. I suspect the climb is more about thermal conduction, but personally I like the idea that it was just due to thirsty microbes :)

Battery

I order to power the device, I used a standard 5v battery pack. The first one that I tried had a low power mode which would keep the battery pack discharging power for 3 hours. This was somewhat inconvenient, so I tried finding another battery pack that would continuously discharge if it detected load. I ended up a battery pack that I found on Amazon for $30. Since the monitor would only draw enough load when it was publishing data, the second battery pack would occasionally auto-off on me so I added an external LED to both indicate that the thing was running and add a tiny bit of extra load on the battery. The LED is one of those that has an internal IC so it annoyingly changes colors and blinks. This LED was amusing in the evening since it would appear as a random dot of changing color just floating in the air.  The second battery pack had a built in solar cell for "emergency use". I was curious if it would generate enough enough extra pixies to offset the compost monitor but I never tested.  The compost pile is fairly well shaded but the battery pack would last 4-5 days running my sensor.

Awesome nerdy details about the physics behind composting:

http://compost.css.cornell.edu/physics.html

More information on how to hot compost:

https://deepgreenpermaculture.com/diy-instructions/hot-compost-composting-in-18-days/

Guide to hooking up DS18B20

https://lastminuteengineers.com/ds18b20-arduino-tutorial/