Smart Koi Pond Project
Project Overview
If you enjoy working on your pond setup, you can easily turn an ESPHome or Arduino board into a simple Koi pond monitor. In this guide, I’ll walk through how I read the water temperature, humidity, the pump and UV light, plus the WiFi signal and general uptime — all shown directly in Home Assistant.
You’ll get the wiring diagram, the basic setup and the exact ESPHome YAML I’m using. It’s not a difficult project, and it works whether you have a small backyard pond or a larger Koi setup. The whole idea is just to keep an eye on what’s going on in the water, so you notice any changes before they become a problem.
Step-by-Step Guide
- Connect the ESP32/ESP8266 board to your sensors.
- Upload the ESPHome YAML configuration to your device.
- Integrate the device with Home Assistant.
Materials
- In this example i used a NODEMCUV2.
- DHT22 AM2302.
- Two wired NTC (5 kΩ NTC-thermistor).
- 5 kΩ reistor.
- Optional: you can add a 10K pull-down resistor on the DHT22 data pin if the signal is unstable.
Project Images
ESPHome KoiPond Sketch
esphome:
name: pond
esp8266:
board: nodemcuv2
logger:
api:
ota:
platform: esphome
password: "ac416d0e41631e70a0f2fbfdfe4ccca5"
wifi:
ssid: !secret wifi_ssid
password: !secret wifi_password
ap:
ssid: "Vijver Fallback Hotspot"
password: "alz4yUCJt1Fe"
captive_portal:
sensor:
- platform: uptime
name: "Uptime Pond"
# DHT22 temperature + humidity
- platform: dht
pin: 5 # GPIO5 (D1)
temperature:
name: "Pond Temperture"
humidity:
name: "Pond Humidity"
model: AM2302
update_interval: 15s
# ADC
- platform: adc
id: source_sensor
pin: A0
filters:
- multiply: 3.3
update_interval: 15s
# ADC voltage to risistor (NTC)
- platform: resistance
id: resistance_sensor
sensor: source_sensor
configuration: DOWNSTREAM
resistor: 5000
name: "NTC Resistance Sensor"
- platform: ntc
sensor: resistance_sensor
calibration:
b_constant: 3470
reference_temperature: 20
reference_resistance: 5000
name: "Pond Water Temperture"
- platform: wifi_signal
name: "Pond WiFi"
id: wifi_signal_db
update_interval: 60s
entity_category: "diagnostic"