KNX Group Addressing
Group addresses are where KNX functions actually live. Here's how they work and how to structure them so a project stays maintainable for decades.
If individual addresses describe where a device is, group addresses describe what it does. Every switch press, dimming command and status update travels on a group address. Get the group-address structure right and a project stays readable for the next engineer years later; get it wrong and even a small system becomes a maze. This guide covers the model and a scheme that scales.
How group addresses work
A shared address links a sender to its receivers.
A group address is a logical channel. A push-button sends a telegram to a group address; every actuator subscribed to that address acts on it. One button can drive many actuators, and one actuator can listen to many addresses — all without any physical wiring change.
Group addresses are independent of topology. That decoupling is the heart of KNX flexibility: you can re-assign what a button does, or add a device to an existing function, purely in ETS.
- Sender (e.g. a button) writes to a group address
- Receivers (e.g. actuators) subscribed to it react
- One-to-many and many-to-one are both normal
- Each address carries one datapoint type (DPT) — switch, dim, %, °C…
Match the datapoint type (DPT) on both ends. Linking a 1-bit switch object to a 1-byte percentage object is the classic beginner bug — ETS warns, but it's worth understanding why.
The main / middle / sub structure
KNX group addresses use a three-level numbering.
Group addresses are usually written as three levels — main/middle/sub, e.g. 1/2/5. ETS also supports two-level and free structures, but three-level is the professional default because it gives two layers of grouping before the individual datapoint.
A common convention assigns the main group to a function category (lighting, blinds, HVAC), the middle group to a location or sub-function (floor, room, or 'switch vs dim vs status'), and the sub group to the individual datapoint.
- Main group — function category (e.g. 1 = lighting)
- Middle group — location or sub-function (e.g. by floor)
- Sub group — the individual datapoint
- Three-level (main/middle/sub) is the professional default
| Format | main / middle / sub |
| Main groups | 0–31 |
| Middle groups | 0–7 |
| Sub groups | 0–255 |
A scheme that scales
Consistency beats cleverness on large projects.
Pick one scheme at the start of the project and apply it everywhere. Whether you organise primarily by function or by building location, the rule is the same: it must be predictable, so anyone can guess where an address lives.
Name every group address descriptively (e.g. 'Lighting / Ground floor / Kitchen ceiling — switch'). ETS exports this structure; clear names are what make a handover document genuinely useful.
- Decide function-first or location-first — then never mix
- Name addresses in full, not cryptic codes
- Reserve ranges for future additions
- Document the scheme so the next engineer follows it
Common mistakes to avoid
Most group-address pain is self-inflicted.
The recurring problems are: no scheme (addresses created ad-hoc), mismatched datapoint types, and overloading a single address with unrelated functions. Each makes the project harder to extend and to debug.
Plan the address space like you plan the topology — before commissioning, not during. Time spent here is repaid every time the system is modified.
- No agreed scheme → unmaintainable project
- Mismatched DPTs → devices that don't talk
- One address doing too much → confusing behaviour
- No reserved ranges → messy later additions
Virasmart's engineers design the full group-address structure as part of every KNX project, so your installation stays clean and extensible. Talk to us before commissioning.
Designing a KNX project?
From topology and group-address structure to ETS programming and commissioning, Virasmart engineers KNX systems across the Baltics.