PROFIBUS Station Address and HSA

Back to Blog

Station Addresses 0 to 126

Each PROFIBUS node - every master and every slave - carries a station address in the range 0 to 126, and it must be unique on the bus. Address 127 is reserved as a broadcast address and is not assigned to a device. Two nodes with the same address is a classic commissioning fault: they collide, and the symptoms range from one device never appearing to intermittent corruption, so a unique-address discipline is the first thing to confirm when a slave will not come online.

Addresses are set either by hardware switches on the device, by software during configuration, or in some cases assigned by a Class 2 engineering master. Low addresses are often reserved by convention for masters and diagnostic tools, with slaves numbered above, though the standard does not mandate a particular scheme. The key rule is uniqueness, and keeping a clear address plan for the bus prevents the duplicate-address problem before it happens.

What the HSA Controls

The HSA, Highest Station Address, is a parameter set on the active stations - the masters. It tells the token logic the highest address at which another master might exist, so the token rotation only bothers looking for potential new masters up to that address rather than all the way to 126. When a master finishes its turn, part of passing the token involves checking whether a new master has appeared in the address gap ahead of the next known one, and the HSA bounds that search.

This has a real performance consequence. Set the HSA far higher than your highest actual master address and the token logic wastes time probing empty addresses looking for masters that will never be there, which lengthens the token rotation and slows the bus. Set it just above your highest master address and rotation stays tight, while still leaving room for one more master to join if you plan to add a diagnostic tool. Note the HSA concerns masters only - slaves can exist at addresses above the HSA, because the token search is about finding active stations, not polling passive ones. This token-rotation behavior connects directly to the polling model used in other master-slave field protocols, where address planning similarly shapes scan efficiency.

An Address Plan That Survives Expansion

The duplicate-address fault is almost always a planning failure rather than a device failure, and a written address plan is the cheap prevention. A workable convention:

1. Reserve the lowest addresses for permanent masters - the controlling PLC and any standing supervisory master.
2. Leave one spare address immediately above the masters for a temporarily attached engineering or diagnostic tool, and set the HSA just above that spare.
3. Number slaves in a contiguous block that mirrors physical order along the trunk, so the address itself tells you roughly where the device hangs.
4. Record every assignment, including reserved spares, in the loop documentation before a device is powered.
5. When a device is replaced, verify the replacement's address before it goes on the bus, since spares often arrive with a different default.

The standard also gives address 126 a particular role: it serves as the delivery default for slaves that accept their working address by software assignment, and a device sitting at 126 is waiting to be addressed rather than exchanging data. Finding a node parked at 126 during commissioning usually means the address-assignment step was skipped.

Symptom Table: Address and HSA Faults

Address and HSA mistakes each have a characteristic signature, which makes a small symptom table worth keeping in the commissioning folder:

Symptom	Likely cause
New master never joins the token rotation	Its address sits above the HSA configured on the existing masters
Bus noticeably slow while every device is healthy	HSA set far above the highest real master, wasting rotation time probing empty addresses
Two devices fault intermittently, or one never appears	Duplicate station address
Slave present and healthy but never polled	Address configured in the master does not match the address set on the device
Device visible but not exchanging data	Device parked at the delivery address awaiting assignment

Changing Addresses on a Running Bus

Many devices read a switch-set address only at power-up, so flipping the switches on a live node changes nothing until the device restarts; whether an address change takes effect immediately is device-specific, per the manufacturer's datasheet. That timing detail is the root of a familiar confusion where a corrected address appears not to have fixed anything, because the device is still answering on the old address until its next power cycle.

Any address or HSA change on a bus that is controlling live equipment should be treated as a controlled intervention: the token rotation is disturbed while stations renegotiate, and I/O exchange with affected slaves stops during the change. Schedule it, follow the site's management-of-change and permit procedures, and let qualified personnel judge whether the process can tolerate the interruption. The wider context of how the bus is structured and how masters and slaves share it is in the PROFIBUS overview.

Frequently Asked Questions