Tasmota SML Images

Tasmota Firmware with SML and Script Support

This repository provides specialized Tasmota firmware images designed for reading smart meters (Stromzähler) using the SML (Smart Message Language) protocol. These images are pre-configured to support advanced scripting capabilities, allowing users to visualize consumption data and emulate smart energy devices without the need to compile their own firmware from scratch. The project targets a wide range of ESP32 and ESP8266 hardware, providing tailored binaries for different flash sizes and feature requirements.

Choosing the Right Image for Your ESP

The firmware is categorized into two primary variants based on the scripting engine provided:

• _tas: This variant includes the classic Tasmota Scripter and Google Charts support, maintaining compatibility with existing script-based configurations.
• _tc: This variant features the TinyC VM and a browser-based IDE. It replaces the classic scripter with a faster, C-subset virtual machine that operates significantly more efficiently.

For ESP32 users, generic images are available with Ethernet support, alongside specialized builds for variants like the C3, C6, S2, S3, and Solo1. ESP8266 users are provided with specific bundles for 1MB and 4MB flash devices, including optimized versions for energy-monitoring plugs and Shelly/EcoTracker emulation for smart battery systems.

TinyC: A High-Performance Scripting Alternative

TinyC represents a significant shift in how logic is implemented on Tasmota devices. Developed as a C-subset compiler and virtual machine, it allows users to write and execute programs directly on the ESP within a web-based IDE. Because TinyC compiles to bytecode, it runs up to ten times faster than traditional text-based interpreters. On ESP32 hardware, it leverages FreeRTOS to run in a dedicated background task, ensuring that blocking operations do not interfere with Tasmota’s core networking and system functions.

In TinyC builds, the SML driver is activated via code using tasm_rule = 1;, which triggers the loading of meter descriptors from the filesystem. To support the IDE and the storage of bytecode, these images utilize a custom partition layout that expands the available filesystem (FS) at the expense of the application area, providing up to 1344 KB of space for scripts and data on standard 4MB ESP32 boards.

Advanced Partition Management with chkpt

Transitioning between different firmware variants often requires adjusting the device’s partition table. This project utilizes a runtime partition manager called chkpt (or TinyCChkpt in TinyC builds). This tool allows users to reconfigure the flash layout—such as shrinking the application partition to expand the filesystem—directly from the Tasmota console without requiring a physical USB connection or factory flash. This workflow is essential for “in-place” upgrades of sensors already deployed in the field.

Custom Build Environment and Optimization

For developers who wish to compile their own versions, the repository includes configuration files for PlatformIO. A key highlight of the build system is the integration of ccache, a compiler cache that significantly reduces build times. By utilizing a second-stage preprocessor mode, ccache can reuse object files across different environment builds (like switching between _tas and _tc variants), often resulting in a 60–75% reduction in total compilation time.

The build process also includes specialized scripts for generating “factory” images for ESP32, which bundle the Safeboot recovery partition and bootloader into a single binary for initial flashing.

Testing and Simulation Tools

To facilitate development and debugging of energy monitoring scripts, several testing tools are recommended. A PowerShell-based UDP/HTTP/PING tester is provided to verify the emulation of Shelly and EcoTracker devices.

Additionally, an SML Simulator tool for Google Chrome allows users to simulate various smart meters. By connecting a USB IR read-head to a PC, developers can generate SML data streams and send them directly to a Tasmota device to test meter descriptors and visualization scripts in a controlled environment.

Debugging and Stability

For troubleshooting unexpected reboots, the project documentation outlines a process for analyzing crash dumps. By connecting the ESP via USB and monitoring the serial output, users can capture “Guru Meditation Errors.” When combined with the .map and .asm files generated during a custom build, these addresses can be traced back to specific functions in the source code, allowing for precise identification of issues like buffer overflows or stack protection faults.