Skip to content

Why BuckPow

Most IoT dashboards answer: "What is happening right now?"

BuckPow answers engineering questions.

The Problem

Building low-power devices involves guesswork.

  • "How long will my battery last?"
  • "Which firmware version uses less power?"
  • "Is my solar panel big enough?"

Traditional IoT dashboards show live voltage and current. They don't help you compare, benchmark, or understand energy behavior over time.

The Gap

Energy measurement is treated as an afterthought. Developers wire up a multimeter, take a few readings, and make decisions based on snapshots.

What's missing:

  • Session-based recording — Compare "before" and "after" firmware changes
  • Energy accumulation — Calculate total Wh, not just instantaneous W
  • Benchmarking — Overlay multiple sessions to find differences
  • Reproducibility — Run the same experiment twice and compare results

BuckPow's Approach

BuckPow treats every measurement as part of an engineering experiment.

Measurements → Sessions → Benchmarks

  1. Measure — Collect voltage, current, power, energy from edge devices
  2. Session — Group measurements into named experiments
  3. Benchmark — Compare sessions to find energy differences

Example Workflow

1. Flash firmware v1.0 → Start session "FW v1.0 baseline"
2. Run workload for 1 hour
3. Stop session
4. Flash firmware v1.1 → Start session "FW v1.1 optimization"
5. Run same workload for 1 hour
6. Stop session
7. Benchmark → v1.1 uses 23% less energy

Who It's For

User Use Case
Makers Optimize battery life for wearables
Researchers Profile TinyML inference energy
Engineers Validate power supply designs
Educators Teach energy-aware programming
Hobbyists Characterize Raspberry Pi energy consumption

Design Principles

Self-Hosted

Your data stays on your server. No cloud dependency.

Open Source

MIT licensed. Modify, extend, deploy anywhere.

Hardware Agnostic

Start with ESP32 + INA219. Add Raspberry Pi, Linux agents, MQTT devices later.

Session-Oriented

Every measurement belongs to a session. Every session is an experiment.

What Makes It Different

Feature Traditional Dashboard BuckPow
Live characterization Yes Yes
Session recording No Yes
Energy accumulation No Yes (Wh)
Benchmark comparison No Yes
Device threshold alerts Sometimes Yes
CSV/XLSX export Sometimes Yes
Self-hosted Varies Yes

Roadmap

Planned features:

  • Raspberry Pi power agent
  • Linux power agent
  • INA226 support
  • MQTT device ingestion
  • PZEM-004T support
  • Multi-user collaboration
  • API webhooks
  • Grafana integration

Try It

git clone https://github.com/arifnd/buckpow.git
cd buckpow
docker compose up -d

Open http://localhost:8000 and start measuring.