IPEM S3-AI | KIT | ESP32-S3 ATM90E36A IoT Ethernet Mains Power Energy Monitor

Name: IPEM S3-AI | KIT | ESP32-S3 ATM90E36A IoT Ethernet Mains Power Energy Monitor
Brand: DitroniX
SKU: IPEM S3-AI-Board
Price: 99.00 GBP
Availability: InStock

£99.00 GBP

What is IPEM S3-AI?

IPEM S3-AI is an acronym for IoT Power Energy Monitor

IPEM S3-AI is a compact and powerful STEM Community board which is designed for IoT monitoring of mains power circuits.

Incorporating the popular ATM90E36A, this high-precision, poly-phase energy metering IC is designed for advanced power monitoring in residential, commercial, and industrial systems.

By incorporating the powerful Espressif S3 MCU, with it's Dual-core Tensilica Xtensa LX7 processor (up to 240 MHz), you are able to take advantage of it's built-in vector instruction sets to accelerate machine-learning math operations.

LLM and AI Integration TinyML & Offline AI capability by running quantized, lightweight models locally (e.g., TensorFlow Lite Micro) for offline tasks.

IPEM S3-AI Also features a separate dedicated SPI port for the Ethernet.

What is a Power Energy Monitor?

A power energy monitor is a device that safely samples the AC voltage and current flow, from one, or more, electrical phases, and from this, derives a range of accurate mains power measurements which can then be used for smart metering, or energy monitoring purposes.

These individual measurements can then be extrapolated into a range of useful real-time data such as total RMS power, active, re-active power, harmonic power, power factor, phase angles, frequency, etc.

The purpose of extracting the power energy data within Solar Inverter installations, as an example, will enable you to much better manage what you do with the available energy, ways of storage, EV Charging, divert under certain conditions, and control flow from, and to, the grid.

All this data is far more extended information than the basic electric meter functionality, allowing for smart IoT systems and greater efficient use of our own energy. It may sound complicated but it does not need to be and all depends on your setup, use-case and what you want to achieve.

Where can IPEM S3-AI be used?

The IPEM S3-AI board can be used in a number of projects and installations such as:

Homes
Education
Offices
Commercial
Factories
Farms
EV Charging
Smart Metering
Greenhouses
Factories
Industry
Solar Farms
Wind Farms
Off Grid Systems

Top Level Features

Espressif ESP32-S3-WROOM-1U-N16R8

Xtensa® dual-core 32-bit LX7 microprocessor (Clock frequency up to 240 MHz)
ESP32-S3-WROOM-1U-N16R8
Edge AI and AIoT Computing Applications
Dedicated vector instructions designed to accelerate neural network (NN)
16 MB Quad SPI Flash and 8 MB PSRAM (Pseudo Static RAM)
512 KB SRAM and 384 KB ROM
2.4 GHz Wi-Fi (802.11 b/g/n up to 150 Mbps), Bluetooth® and Bluetooth 5 Low Energy (LE), with long-range support
AI Acceleration vector instructions that accelerate neural network computing and signal processing workloads
Ultra-Low Power (ULP) Co-processor allows the module to remain in deep sleep while performing simple tasks, preserving battery life

Microchip ATM90E36A

High-performance poly-phase energy metering
Wide Dynamic Range 6000:1
Supports 3-phase 4-wire, 3-phase 3-wire, 2-phase, split-phase systems and singles phases
High Accuracy Measuring active energy to within ±0.1% and reactive energy to ±0.2%

Ethernet (on Dedicated SPI Port)

W5500 - Ethernet Controller with Full TCP/IP Stack
SI3402 - Power over Ethernet (PoE+ Isolated)

RS-485 / DMX

CS48520AD - High-speed, half-duplex RS-485
For local data gathering and interfacing

Real Time Clock

DS3231SN - Accurate RTGC with Battery Backup

MicroSD

On-board MicroSD Holder (Board Underside)

OPTO Output

Galvanically Isolated Output

ADS1115

DC Voltage Sensing
0-3V3 or 0-75V Inputs

Board Power

USB C 5V
DC Input 5-20V
AC Input 8-20V
PoE 37 to 57V

DIN Mountable

External Interface (12W)
SPI Interface (6W)

Overview

RS-485

To interfacing to other Smart Meters, Devices and Smart Batteries which have an RS-485 MODBUS interface, the IPEM S3-AI includes a standard industry interface which allows you to collect, or control, other devices on the RS-485 bus.

This RS-485 interface could also be used to control local CCTV on remote installations, or even send data from the IPEM S3-AI board over a single twisted pair, up to around 1,200 meters (4,000 feet), in distance. Ideal for farms and remote installations.

Calibration

The ATM90 series of monitor devices are proven for accuracy. The IPEM S3-AI core is the ATM90E36A which is used around the world in power monitoring instruments that also need to measure voltage, current, THD, DFT and mean power.

The IPEM S3-AI board requires little, to no, calibration and is easy to setup.

As the IPEM S3-AI board only safely sample the AC via a low voltage transformer, you are able to monitor varying AC (RMS) voltage mains around the World, such as:

USA and Canada which typically is 120 V or 230 V @ 60 Hz

With ranges around 114 V to 126 V

Europe and UK which typically is 230 V @ 50 Hz

With ranges around 216 V to 253 V

Japan which typically is 100 V @ 50 Hz or 60 Hz

Eastern Japan is 50 Hz, and Western Japan is 60 Hz

Should you wish to know more, see this Wiki

CT Clamps and Rogowski Current Transformer

IPEM S3-AI was designed primarily for the standard CT clamp such as the YHDC SCT-013 100A-50mA. Inputs for each channel 1, 2, 3, plus N are via Terminal Block Connectors.

You can also use Rogowski Current Transformers on any inputs. Each input is configurable using a jumper.

Mains Monitoring

All my boards are designed to be safe and easy to use - with NO live working, mains electricity parts or dangerous exposed high voltages.

The IPEM S3-AI uses only low voltage 12 V AC such as from a bell transformer and clip on CT Clamps.

Electrical Circuit Types

The IPEM S3-AI is capable of monitoring a number of variations in energy monitoring systems such as:

Accurately monitors both Import and Export
Single phase (Home, Office etc.)
Multiple Single Phases. (Different circuits or home/office rings, Inverters etc.)
Dual Phase (2 x Live -Typically for USA. i.e. 2 x 110V)
Three Phase Star (3 x Live and 1 x Neutral)
Three Phase Delta (3 x Live)

Data Output

Data provided from the IPEM S3-AI can be extrapolated through your code in any way you wish, in order to easily integrate with your IoT Home Automation system, or Solar Installation monitor.

Using the powerful ESP32-S3, you are also able to take advantage of AI features included with dual Xtensa LX7 cores, vector acceleration, and extensive PSRAM/Flash support.

All software and code is open source, community based and allows you to develop and integrate as you wish.

The main process typically used for publishing data (internally or externally), is via MQTT, or you could use other direct publishing routes such as to Domoticz, Zabbix or Home Assistant.

Data from the ATM90E36 includes:

Mains RMS Voltage (each phase)
Mains Frequency (each phase)
Import and Export Values

Mains RMS Current
Calculated RMS Power
Active Power (Absorbed or Used by the Load)
Calculated Total Active Power
Re-Active Power
Calculated Total Reactive Power
Apparent Power (Total Amount of Power Flowing from Source to Load)
Calculated Total Apparent Power
Fundamental Power
Harmonic Power
Power Factor
Phase Angle

Programming and Development

Development is primarily supported through Visual Studio Code (VSCode) and Arduino IDE. This also includes platforms like ESPHome.

A range of libraries are already available which support the devices on the IPEM S3-AI such as the:

ESP32-S3 N16R8
ATM90E36A
W5500
24C64
DS3231SN
TMP102
WS2812

Code support is available in my existing GitHub IPEM and FLiX repositories, with a specific core IPEM S3-AI code repository being provided as the board evolves, together with a new FLiX test and bring up firmware, which is in development.

Home Assistant is fully supported via ESPHome Firmware.

This is a worldwide community project and is able to be integrated in so many systems.