PROFIBUS DP (Decentralized Periphery) is a serial fieldbus communication protocol designed for fast, deterministic data exchange between a central controller and distributed field devices — I/O modules, drives, sensors, actuators, and remote terminal units. It is the dominant variant of PROFIBUS and the backbone of tens of millions of installed automation systems worldwide.
In the IEC standards framework, PROFIBUS DP is defined as Communication Profile 3/1 (CP 3/1) within Communication Profile Family 3 (CPF 3) of IEC 61784-1:2014. It operates on RS-485 (or fiber optic) physical media and uses asynchronous transmission — a key distinction from PROFIBUS PA (CP 3/2), which uses synchronous MBP transmission.
The core design goal of PROFIBUS DP was speed and simplicity. Where PROFIBUS FMS (its predecessor) was a complex general-purpose protocol, DP was streamlined to do one thing extremely well: poll remote I/O devices as fast as possible in a deterministic cycle. That focus is why DP dominated the market and FMS was eventually retired.
Table of Contents
PROFIBUS DP in the IEC Standards Framework
PROFIBUS DP is standardized across two IEC documents:
- IEC 61784-1:2014 — defines the Communication Profile (CP 3/1), specifying which parts of the IEC 61158 series apply
- IEC 61158 Type 3 — the underlying technology stack (physical, data-link, and application layers)
The IEC 61784-1 standard is explicit: PROFIBUS DP (the application layer protocol) is identical for both CP 3/1 and CP 3/2. What differs is the physical layer — RS-485 asynchronous transmission for CP 3/1, versus MBP synchronous transmission for CP 3/2 (PROFIBUS PA). This shared protocol design is what allows PA field devices to be integrated into a DP network via a segment coupler with no protocol changes at the controller level.
Physical Layer: RS-485 and Fiber Optic Options
CP 3/1 supports four physical layer variants, each identified by a code number in the device’s GSD file:
| Code | Medium | Notes |
|---|---|---|
| 0 | RS-485 copper | Standard; optional IS version (RS-485-IS) |
| 2 | Plastic optical fiber (POF) | Short distances, low cost |
| 3 | Glass fiber (multimode or singlemode) | Long distances, high immunity |
| 4 | PCF fiber (polymer-clad silica) | Intermediate option |
RS-485: The Most Common Choice
RS-485 is a balanced differential signaling standard using shielded twisted-pair cable with a characteristic impedance of 150 Ω. Key characteristics for PROFIBUS DP:
- Cable: violet sheath, two-core screened twisted pair
- Connectors: 9-pin D-Sub (DB9) or M12 circular
- Topology: linear bus with terminators at both ends (220 Ω pull-up/pull-down, 110 Ω terminator resistor)
- Maximum stations per segment: 32 (without repeaters)
- Maximum stations total: 126 (with up to 9 repeaters in series)
- Each repeater counts as one station on each segment it connects
Baud Rate vs. Cable Length
PROFIBUS DP is unique in that the cable length limit depends on the selected baud rate. A single baud rate applies to the entire network — all devices must operate at the same speed.
| Baud Rate | Max. Cable Length per Segment |
|---|---|
| 9.6 kbit/s | 1,200 m |
| 19.2 kbit/s | 1,200 m |
| 93.75 kbit/s | 1,200 m |
| 187.5 kbit/s | 1,000 m |
| 500 kbit/s | 400 m |
| 1.5 Mbit/s | 200 m |
| 3 Mbit/s | 100 m |
| 6 Mbit/s | 100 m |
| 12 Mbit/s | 100 m |
For most factory automation applications, 1.5 Mbit/s or 12 Mbit/s is chosen. Process applications with longer cable runs may use lower speeds.
RS-485-IS: Intrinsically Safe Variant
For PROFIBUS DP in hazardous areas, an intrinsically safe version of RS-485 — RS-485-IS — is available. It limits bus current and voltage to below ignition thresholds. This is less common than PROFIBUS PA’s MBP-IS for process instrumentation but is used in discrete automation applications in Zone 2 / Division 2 areas.
Fiber Optic
Fiber eliminates ground loop problems and provides immunity to electromagnetic interference — important in environments with heavy motor drives, welding, or high-voltage switchgear. Glass fiber segments can span up to 15 km between repeaters. Topologies supported include bus, star, and ring. Ring topology can be configured redundantly for high-availability applications.
Bus Access: Token Passing + Master/Slave
PROFIBUS DP uses a hybrid access method at the data-link layer, combining two mechanisms:
Token Passing Between Masters
In a multi-master network, active stations (masters) pass a logical token between themselves, forming a token ring. Only the station currently holding the token may initiate communication. The token rotates around the ring in order of station address. This ensures that every master gets predictable, fair access to the bus — no master can monopolize it.
The key timing parameter is the Target Token Rotation Time (T_TR) — the configured maximum time for the token to complete one full rotation. This parameter, set during network configuration, directly determines the worst-case cycle time for the network.
Master/Slave Between Masters and Field Devices
Slaves (field devices) are always passive. They cannot initiate communication. A slave only responds when addressed by the token-holding master. This makes the timing of slave responses entirely predictable — the basis for PROFIBUS DP’s determinism.
The primary communication service is SRD (Send and Request Data with Reply): the master sends output data to a slave in the same telegram that requests input data back. The slave responds with its input data. This combined send/receive in one transaction is highly efficient.
For broadcast outputs (e.g., SYNC and FREEZE commands), the SDN (Send Data with No Acknowledge) service is used — the master sends to all slaves simultaneously with no individual response required.
The DP Versions: V0, V1, and V2
PROFIBUS DP is not a single fixed specification. It evolved in three versions, each adding capabilities while remaining backward compatible.
DP-V0 — The Foundation
Every PROFIBUS DP device must implement DP-V0. It covers the core cyclic exchange:
Cyclic I/O data exchange — the Class 1 master polls each assigned slave in a fixed, repeating cycle. Output data (from controller to device) and input data (from device to controller) are exchanged in every cycle. The maximum payload per slave is 244 bytes input and 244 bytes output.
Diagnostics — every slave reports a diagnostic status on each cyclic response. The diagnostic data includes: station status bytes (is the slave responding correctly?), master address (which master owns this slave?), and device-specific diagnostic information. Standard diagnostic flags cover configuration errors, parameter faults, and device hardware failures.
Parameterization — the master sends a parameter set to each slave during startup. This configures the slave’s behavior before cyclic data exchange begins.
Configuration check — the master verifies that the slave’s actual I/O configuration matches what was configured in the engineering tool. If there is a mismatch, the slave goes into error state and does not exchange data.
DP-V1 — Acyclic Communication
DP-V1 adds acyclic communication alongside the normal cyclic exchange. This is the foundation for advanced device management.
Class 1 master acyclic services — the controlling master can read or write any parameter on any slave, at any time, without disrupting the cyclic exchange. This is used for setpoint changes, mode switches, or reading extended diagnostic data during operation.
Class 2 master access — an engineering station, asset management system, or handheld configurator can connect to the network as a Class 2 master and access any slave acyclically. This enables full remote configuration and diagnostics without interrupting production.
Alarm handling — DP-V1 slaves can push unsolicited alarm messages to the Class 1 master. Alarm types include process alarms (process variable out of range), diagnostic alarms (device failure), status alarms, and update alarms. The master must acknowledge each alarm. This replaces the polled-diagnostic model of DP-V0 with an event-driven approach that is much more efficient for high device counts.
DPV1_Status bytes — the first three bytes of the user parameterization data take on specific meaning in DP-V1: DPV1_Status_1, _2, and _3 enable alarm communication and other extended features. A device with DP-V1 features must set the DPV1 attribute to TRUE in its protocol configuration (as referenced in IEC 61158-6-3).
DP-V2 — High-Performance Extensions
DP-V2 adds features specifically for demanding applications like motion control, coordinated drives, and large distributed I/O systems.
Slave-to-slave communication (DXB — Data Exchange Broadcast) — a slave can publish data directly to other slaves on the bus as a publisher, with those slaves subscribing to that data. This allows direct device-to-device communication without the data passing through the master. Latency is dramatically reduced for applications like synchronized drive coordination.
Isochronous mode — all devices on the network are synchronized to a common clock. Cyclic I/O, application processing, and data exchange all occur at precisely defined time offsets within each bus cycle. This is essential for high-performance motion control (position, velocity, torque synchronization) where jitter in the order of microseconds matters.
Time stamping — events in slaves can be tagged with a network-wide timestamp, enabling sequence-of-events recording with sub-millisecond resolution across the entire network.
Upload/download of load regions — large data blocks (programs, recipes, configuration data) can be transferred to slaves in the background without disrupting the cyclic exchange.
Master and Slave Architecture in Detail
Class 1 Master (DPM1)
The primary controller — typically a PLC, DCS, or motion controller. The Class 1 master:
- Owns a defined set of slaves
- Performs the cyclic I/O exchange with those slaves
- Sends parameterization and configuration data to slaves during startup
- Manages slave diagnostics
- Holds the token during its token rotation period
A Class 1 master must initialize each slave before data exchange begins. The initialization sequence is: Reset → Parameterization → Configuration Check → Data Exchange. If any step fails, the slave does not enter data exchange and the master reports an error.
Class 2 Master (DPM2)
An engineering or asset management station. The Class 2 master:
- Can read/write parameters on any slave acyclically (DP-V1 required)
- Does not own slaves — it shares bus access via the token ring
- Can coexist with one or more Class 1 masters on the same network
- Is typically a handheld configurator, PC-based engineering tool, or asset management server
Multiple Class 2 masters can be present simultaneously. Each holds the token in its turn and performs its acyclic transactions within that window.
Slaves
Passive field devices. Slaves:
- Never initiate communication
- Respond only when addressed by a master holding the token
- Have a unique bus address (0–125 for slaves; 0–126 total including masters and repeaters)
- Are either compact (fixed I/O) or modular (configurable slots)
Modular slaves allow the controller to define which modules are active in which slots, making them highly flexible for varied I/O requirements. The GSD file describes all available modules and their I/O data formats.
The GSD File in Detail
Every PROFIBUS device must have a GSD file (General Station Description). The GSD is a standardized text file per ISO 15745-3 that an engineering tool reads to add the device to a PROFIBUS network.
The GSD contains three sections:
General specifications — vendor name, device name, hardware/software revision, supported baud rates, timing parameters (T_SDR min/max — the minimum and maximum station delay response times the device can handle).
Master-related specifications (masters only) — maximum slaves that can be connected, upload/download capabilities.
Slave-related specifications — I/O channel count and type, parameter sets, available modules for modular devices, diagnostic text definitions, bitmap files for visualization in the engineering tool.
The GSD format is flexible. It supports both fixed parameters (like the vendor name) and option groups (like selectable module configurations). Bitmap images of the physical device can be embedded for display in configuration tools.
Each certified PROFIBUS DP device must have a type-specific GSD and pass conformance testing through PROFIBUS International. This guarantees that any compliant engineering tool can configure the device without vendor-specific software.
Application Profiles
Beyond the base DP protocol, PROFIBUS International defines application profiles — standardized behavior specifications for specific device types or application domains. These profiles run on top of PROFIBUS DP and ensure that devices from different manufacturers behave consistently at the application level.
PROFIdrive
The most widely used application profile after PA. PROFIdrive standardizes the interface between motion controllers and drive systems (frequency inverters, servo drives, spindle drives). It defines:
- Drive parameter objects and their numbering
- Standard control words and status words
- Velocity, position, and torque control interfaces
- Multiple application classes from simple speed control to synchronized multi-axis motion
PROFIdrive uses DP-V2 isochronous mode for the highest-performance synchronized motion applications.
PROFIsafe
Adds a safety layer on top of PROFIBUS DP for functional safety applications (SIL 1–3 per IEC 62061, PL c–e per ISO 13849). PROFIsafe works by adding a safety protocol wrapper around the normal DP I/O data. Safety devices (e-stops, light curtains, safety relays) exchange safety I/O alongside standard I/O on the same cable. The safety protocol provides:
- CRC-based error detection across the entire communication path
- Message counter to detect lost or repeated messages
- Watchdog monitoring
- Separate safety controller (F-CPU) that processes safety I/O
This eliminates the need for separate hardwired safety loops in many applications.
PROFIBUS PA Device Profile
Standardizes process field instruments — pressure, temperature, flow, level, valves, analyzers. Defines function blocks, parameter structure, and the measurement value + status byte format. Used specifically with CP 3/2 (PROFIBUS PA) physical layer, but the profile itself is an application layer specification built on the shared DP protocol.
Bus Cycle Time: What Determines Speed?
The bus cycle time is not a fixed parameter — it depends on several factors that engineers must consider during design:
Number of slaves — more slaves means more polling telegrams per cycle.
I/O data volume per slave — larger data payloads take longer to transmit.
Baud rate — higher baud rate reduces telegram transmission time directly.
Number of masters — each master needs its share of the token rotation time.
Acyclic transactions — DP-V1 acyclic reads/writes consume part of the token holding time and extend the overall cycle.
Target Token Rotation Time (T_TR) — the configured budget for one full token rotation. Set too tight and acyclic transactions will be throttled. Set too loose and cycle time is unnecessarily long.
As a practical example, at 12 Mbit/s with 32 compact slaves (4 bytes I/O each), a typical bus cycle time is 1–3 ms. At 1.5 Mbit/s with the same configuration, it is around 10–25 ms.
Network Design: Practical Considerations
Termination — both ends of the RS-485 trunk must be terminated. Unterminated networks produce signal reflections that cause communication errors, especially at higher baud rates. Most PROFIBUS connectors have a built-in termination switch that can be enabled at the physical end of the bus.
Address assignment — address 0 is typically reserved for programming devices or default Class 2 masters. Address 1 is often the Class 1 master. Slaves use addresses 2–125. Gaps in the address space are allowed but increase GAP polling overhead — the master periodically checks unused addresses for newly connected devices.
Stub lines — PROFIBUS DP does not support spur/stub lines at higher baud rates. At 12 Mbit/s, even a short stub (30 cm) can cause reflections severe enough to disrupt communication. All devices must be connected to the bus trunk directly or through the connector’s integrated circuitry.
Repeaters — extend the physical segment. Each repeater counts as a station on each of the two segments it bridges. With 9 repeaters in series, a network can span up to 10,800 m at 9.6 kbit/s (9 segments × 1,200 m).
Optical fiber — preferred when EMC is a concern (drive cabinets, welding environments), when ground loops exist between control cabinets, or when long distances are required. The network topology is more flexible with fiber — star and ring configurations are common.
Diagnostics and Troubleshooting
PROFIBUS DP has one of the richest built-in diagnostic frameworks of any fieldbus — another reason for its long-term success.
Station Diagnostics (DP-V0)
Every slave returns up to 6 bytes of standard diagnostic data on each cyclic response:
Station Status 1 — indicates: extended diagnostic data present, device is not ready for data exchange, configuration fault, parameterization fault, master address not correct, slave does not support requested functions.
Station Status 2 — indicates: static diagnostic (device requires a new parameterization), slave deactivated, slave has changed its response time, master/slave assignment issue.
Station Status 3 — indicates extended diagnostic available (more than 6 bytes of diagnostic data queued).
Master Address — the address of the Class 1 master that owns this slave. If a different master attempts to parameterize a slave already owned by another master, an error is set here.
Ident Number — a 16-bit identifier unique to the device type, used to verify GSD compatibility.
Extended Diagnostics (Device- and Channel-Specific)
Beyond the standard 6 bytes, slaves can report extended diagnostic data up to 244 bytes. Three formats are defined:
- Device-related — general device faults (hardware failure, temperature, supply voltage)
- Identifier-related — which module slot has the fault
- Channel-related — which specific I/O channel has the fault, and the fault type (wire break, short circuit, overrange, etc.)
Channel-level diagnostics allow a fault on a single I/O point to be isolated without removing the device from the network — a major advantage over simple analog I/O.
Common Fault Patterns
“Station not present” / slave drops out — check physical layer first: cable continuity, termination at both ends, correct baud rate in device, bus address uniqueness. A protocol analyzer (ProfiTrace, ProfiCore) can capture the actual RS-485 signal and measure eye diagrams, jitter, and signal levels.
Parameterization error — GSD version mismatch between what the engineering tool configured and what the device expects. Ensure the GSD file version matches the firmware version in the device.
Configuration error — the I/O module configuration sent by the master does not match the actual modules present in the slave. Check the hardware configuration in the engineering tool matches the physical hardware.
High cycle times — if the bus cycle is longer than expected, check for DP-V1 acyclic traffic consuming token-holding time, or check the T_TR setting. A Class 2 master connected during production can significantly impact cycle time.
PROFIBUS DP and PROFINET: The Migration Path
PROFINET (Industrial Ethernet) is the successor to PROFIBUS DP for new installations. However, with hundreds of millions of installed PROFIBUS DP nodes, migration is a long-term process — not a rip-and-replace.
The practical migration strategies are:
PROFIBUS DP proxy — a PROFINET device that acts as a Class 1 master on the PROFIBUS segment. All DP slaves appear as PROFINET submodules in the I/O controller’s device tree. No changes to the PROFIBUS field devices are required. This is the most common approach when upgrading a PLC or DCS to PROFINET.
Dual-port devices — some newer field devices support both PROFIBUS DP and PROFINET natively, allowing flexible deployment during a phased transition.
Coexistence — PROFIBUS DP and PROFINET can run in parallel indefinitely. Many sites operate both networks, with PROFIBUS DP serving existing field devices and PROFINET used for new installations.
Standards and Specifications Summary
| Standard | Content |
|---|---|
| IEC 61784-1:2014, CP 3/1 | PROFIBUS DP communication profile definition |
| IEC 61158-2, Clause 22 | RS-485 physical layer (asynchronous transmission) |
| IEC 61158-2, Clause 23 | Fiber optic physical layer |
| IEC 61158-3-3 | Data-link layer services |
| IEC 61158-4-3 | Data-link layer protocol |
| IEC 61158-5-3 | Application layer services |
| IEC 61158-6-3 | Application layer protocol |
| ISO 15745-3 | GSD file format |
| IEC 62061 / ISO 13849 | Functional safety (referenced by PROFIsafe) |
Key Numbers to Know
| Parameter | Value |
|---|---|
| Max baud rate | 12 Mbit/s |
| Min baud rate | 9.6 kbit/s |
| Max segment length @ 12 Mbit/s | 100 m |
| Max segment length @ 9.6 kbit/s | 1,200 m |
| Max stations per segment | 32 |
| Max total stations | 126 |
| Max repeaters in series | 9 |
| Max I/O per slave | 244 bytes in + 244 bytes out |
| Slave address range | 0–125 |
| Cable impedance (RS-485) | 150 Ω |
| Cable color (RS-485) | Violet sheath |
| IEC profile identifier | CP 3/1 (IEC 61784-1) |
| Physical layer standard | IEC 61158-2, Clause 22 (RS-485) |
Summary
The PROFIBUS DP protocol earned its dominant position in factory and hybrid automation through a combination of speed, determinism, simplicity, and a rich diagnostic framework. Its three-version evolution — from the basic cyclic exchange of DP-V0 to the alarm-driven acyclic communication of DP-V1 to the isochronous synchronization of DP-V2 — means a single protocol has served applications from simple remote I/O to high-performance synchronized motion control.
The shared protocol with PROFIBUS PA remains one of its most underappreciated strengths: a single controller, a single protocol stack, and a single engineering environment can manage both discrete factory automation and process field instrumentation simultaneously. For hybrid plants — pharmaceuticals, food and beverage, batch chemicals — this is a real operational advantage.
With PROFINET as the designated successor and proxy devices enabling painless integration of existing PROFIBUS DP installations, the protocol’s long-term support path is clear. For engineers working with existing installations or evaluating brownfield upgrades, a thorough understanding of PROFIBUS DP remains essential.