GitHunt
RO

RondaYummy/bms-monitor-ble

Simple application for controlling JK-BMS via Bluetooth low energy, Deye inverter, Tapo smart outlet

blebleakbluetoothbluetooth-low-energydeyedocker-composejikong-bmsjkbmsmodbuspwa-appspysolarmanv5rasberry-pitapotapo-p100uart

Devices overview System summary App settings

Supported devices

All JK-BMS models with software version >=6.0 are using the implemented protocol and should be supported.

  • JK_BD6A20S6P, hw 17U, sw 17.02
  • JK_BD6A17S6P, hw 11.XA, sw 11.48

All Deye models from this list are supported.

  • SUN-6K-SG01/03/LP1-EU/AU
  • Tapo P100
  • Tapo P110

BMS Monitor [BLE]

🔋 Monitoring System for JK-BMS and Deye Inverter

This application provides full real-time monitoring of your energy system.
It connects to:

  • JK-BMS via Bluetooth low energy (using the bleak library)
  • Deye Inverter via WiFi stick (using pysolarmanv5)

🧠 Key Features:

  • 🔋 Real-time monitoring of battery and inverter data

    • Battery voltage, current, power, SOC, SOH
    • Cell temperatures, internal resistance, cycle count, balancing
    • Solar generation, home consumption, grid import/export

  • 🌡️ Automatic alerts for temperature or imbalance

  • ⚙️ Dynamic power balancing with Tapo smart plugs

    • Dynamic Load Shedding: Automated power balancing to prevent inverter overload by switching Tapo smart plugs and reporting the number of devices disabled.

⚠️ WEB PUSH Notifications

Critical events (e.g. overheating, cell imbalance, low charge) are sent as push notifications to the PWA frontend.

📱 Progressive Web App (PWA)

The frontend is a PWA that works offline, supports mobile devices, and receives browser push notifications.

🚀 Why it's better:

  • No limitations of native apps (e.g. Bluetooth only from a phone)
  • All data in one place — accessible from any device
  • Runs autonomously on Raspberry Pi 5, with no cloud dependency, 24/7

Full control over your energy system — stable, local, and convenient.

⚠️ IMPORTANT: Before Using Deye Inverter and Tapo Sockets

To ensure stable system operation, you must assign static IP addresses to the inverter and Tapo sockets via your router settings. This prevents random IP changes after reboots and guarantees continuous connectivity.

✅ How to do this

  1. Open your router’s admin panel in a browser:

    http://192.168.31.1

  2. Navigate to the section:

    DHCP > Static Leases / Address Reservation / IP Binding

  3. Add the MAC address of the inverter's WiFi stick and assign it a static IP address. Example:

    • MAC: DC:4F:22:xx:xx:xx
    • IP: 192.168.31.39

  4. Add your Tapo smart plug the same way:

    • MAC: AC:84:C6:xx:xx:xx
    • IP: 192.168.31.110

  5. Save the configuration.

  6. Restart your router or devices to apply the changes.

🔁 After this setup, the IP addresses will no longer change and your system will maintain a stable connection with the inverter and smart plugs.

✅ This is highly recommended for all automation systems that depend on consistent local network addresses.

Deye

To connect to the inverter, you need to pass the following variables: INVERTER_IP, LOGGER_SN.

To connect to a Tapo outlet, you need to specify the outlet IP and EMAIL and PASSWORD from the official Tapo app.

⚙️ System Architecture

Graph TD
BMS[JK-BMS] --> PythonApp[Python Backend]
Deye[Deye Inverter] --> PythonApp
PythonApp --> DB[SQLite DB]
PythonApp --> PWA[Frontend (PWA)]
PythonApp --> Push[Web Push Notifications]

First steps:

To access your application from outside the local network, you need to get a static IP address from your Internet provider. And etup Nginx and SSL ( https://github.com/RondaYummy/bms-monitor-ble/blob/main/docs/nginx.md )

Make sure that the systemctl service is running:

sudo systemctl start bluetooth
sudo systemctl enable bluetooth

Build Docker:

yarn build

To copy static resources after a manual build:

yarn static

[PROD] - [DEV]

yarn prod
yarn dev

[AUTO] Deploy via PM2

Preparing:

echo "$(whoami) ALL=(ALL) NOPASSWD: /sbin/reboot" | sudo tee /etc/sudoers.d/reboot-nopasswd
chmod +x deploy.sh
npm install pm2 -g

[AUTO] Deploy start:

pm2 start ecosystem.config.js

Clear Database [Example]

docker ps
docker exec -it bms-monitor-ble-python-app-1 bash
rm /app/data/*.db
exit
docker compose restart

🚀 Motivation

The official BMS app left me deeply dissatisfied due to its lack of critical features and usability limitations. It didn’t provide real-time notifications for important events like low battery charge, missing charging sessions, or abnormal cell behavior. Monitoring these parameters while physically standing next to the BMS with a smartphone felt inefficient and outdated.

I envisioned a solution where I could access, monitor, and control my entire energy system from anywhere, without being tied to a specific location or mobile app.

So I decided to build my own system — starting with Python and a lightweight web interface that consolidates all the essential features the official tools lacked. My application adds:

  • 🔔 Web push notifications for overheating, cell imbalance, and battery issues
  • 🌍 Remote access via PWA, accessible from any browser or device
  • 📊 Advanced data visualization through real-time charts and logs

As the project grew, I expanded beyond the BMS:

  • 💡 TP-Link Tapo smart plugs were integrated to enable remote power control of connected devices (e.g. chargers, inverters, routers). Now I can automatically turn them on/off based on battery state or schedule.
  • ⚡ Deye Inverter support was added via Wi-Fi stick to monitor solar generation, grid consumption, and system balancing.
  • 🧠 All these components now work together as one intelligent system, running autonomously on a Raspberry Pi 5, without relying on the cloud.

The goal was simple: bring clarity, control, and automation to my energy setup — in a way that’s stable, open, and privacy-friendly.

This project is my attempt to create a fully local, integrated, and extensible energy management platform — something I believe every DIY energy enthusiast should have access to.

## 🗺️ Roadmap / Features

- [x] JK-BMS monitoring via BLE
- [x] Deye Inverter monitoring via WiFi
- [x] TP-Link Tapo plug integration
- [x] PWA frontend with offline support
- [x] Web push notifications
- [x] Dockerized backend
- [x] Automated power balancing to prevent inverter overload by switching Tapo smart plugs and reporting the number of devices disabled.
- [ ] Turning on Tapo smart plugs using timers.
- [ ] Dashboard with charts
- [ ] Export historical data (CSV / Grafana)
- [ ] Add multi-inverter support

🛠️ Tech Stack

Layer Technology
BLE Communication bleak (Python)
Inverter API pysolarmanv5
Smart Plugs TapoP100 fork
Backend Python / FastAPI / SQLite
Frontend Vue 3 + TypeScript + Quasar + PWA
Deployment Docker + PM2
Notifications Web Push API

Top contributors:

contrib.rocks image

💬 Feedback & Support

Feel free to open an issue if you encounter any problems, or suggest new features here:
👉 GitHub Issues

References

Languages

Vue40.5%Python37.9%TypeScript16.5%SCSS1.8%Shell1.5%JavaScript1.1%HTML0.5%Dockerfile0.2%

Contributors

RO
Apache License 2.0
Created January 14, 2025
Updated November 17, 2025
RondaYummy/bms-monitor-ble | GitHunt