Resolving Siemens S7-315 Profibus Bus Fault with Danfoss FC300

1. Problem Description

A Profibus DP network connecting a Siemens SIMATIC S7-315-2DP/PN CPU to a Danfoss VLT FC 300 drive develops an intermittent bus fault as soon as the drive is commanded to run. The configuration downloads to the PLC without error, the FC 300 acknowledges Profibus synchronisation, and the first cyclic data exchange appears healthy. The moment the user writes the start control word (CTW) into the drive's PZD input area, two events occur simultaneously:

• The FC 300 front display shows Alarm W34 ("No Profibus communication").
• The S7-300 diagnostic buffer (open in STEP 7 via PLC → Module Information → Diagnostic Buffer) appends the entries "Distributed I/O: station failure" and, a few hundred milliseconds later, "Distributed I/O: station return".

The bus LED on the Danfoss Profibus option then re-illuminates, the warning W34 self-clears, and PLC/drive communication is re-established. As soon as the CTW is written back to a stop value and the motor has decelerated to zero speed, the diagnostic buffer appends the same pair of entries again. A critical field observation is reported in the troubleshooting session: disconnecting the motor output conductors from the drive terminals U/V/W makes the bus fault vanish instantly, even while the drive is still commanded to run. This single test narrows the root cause to load-side electromagnetic interference rather than configuration, addressing, or GSD issues.

2. Affected System Components

3. Diagnostic Buffer Interpretation

The two S7-300 diagnostic events are the master firmware's response to a complete loss of Profibus frame acknowledgement from the slave for at least one full watchdog interval. The events are not configuration errors — they are runtime faults. The exact event IDs and meaning are:

Because the two events appear in immediate succession, the slave is not permanently down — it is being knocked off the bus for tens of milliseconds during the load transition. That rules out a duplicate address, GSD parameter mismatch, or watchdog configuration error (those would generate persistent station-failure entries, not a paired failure/return).

4. W34 Alarm on the Danfoss FC 300

The Danfoss FC 300 Profibus manual maps W34 to a "No communication with the drive" warning. It is raised when the drive's Profibus option detects that no valid DP telegram has been received for longer than the configured telegram failure time (parameter 8-07). The warning is non-latching by default; once telegrams resume it self-clears. Two key W34-related parameters on the FC 300 are:

5. Root Cause Analysis: Motor-Load-Coupled Bus Fault

The decisive field test is: "when I pull out the motor output of the inverter, the error of PLC disappears instantly." With the configuration, cable type, address, and connector hardware all already ruled out, this is the textbook signature of common-mode conducted and radiated noise from the inverter's output stage. The mechanism is:

1. The drive's IGBT bridge switches at 2–8 kHz with sub-microsecond edges, producing high dV/dt at the motor terminals (typically 3–8 kV/µs).
2. The cable-to-motor parasitic capacitance and bearing/capacitive coupling to ground force a high-frequency common-mode current back through the PE conductor and the cable shield.
3. Part of that common-mode current flows onto the Profibus cable shield, especially if both shields share the same backplane ground path or run in parallel trays.
4. The common-mode voltage shifts the Profibus reference ground above the receiver's common-mode input range (Siemens Profibus transceivers tolerate typically ±7 V on the bus pins), corrupting one or two frames and triggering the station-failure/return pair.

6. Profibus Physical Layer Verification

Before changing the program, perform the following field checks. None of these require STEP 7 modifications.

1. Segment voltage & termination. With the bus running, measure between pins 5 (GND) and 6 (VP, +5 V) at the master and the drive connector. A healthy bus reads between 4.75 V and 5.25 V DC. Voltage below 4.5 V indicates too many terminations or a shorted stub.
2. Terminator count. Exactly two terminators must be ON in any single segment, at the first and last physical node. The terminator is the built-in 220 Ω + 390 Ω resistor network inside the D-sub connector, activated by the slide switch. A drawing of the two-termination rule is shown below.

Two-termination rule: ON at first and last physical node, OFF in between.

3. Shield bonding. The Profibus cable shield must be clamped 360° at every connector with the metal strain-relief saddle in contact with the shield, not the drain wire alone. Lift the shield only at one end (the master) if a low-impedance ground path is uncertain; otherwise bond at both ends.
4. Trunk length. At 1.5 Mbit/s the maximum trunk length is 200 m. A 2 m segment like the one described here is well within spec, so distance is not the cause.
5. Connector pin check. Pins 3 (B-line) and 8 (A-line) on the 9-pin D-sub must show clean ±1 V differential eye; pins 5 (DGND) and 6 (VP) the 5 V supply. A bent pin is a frequent commissioning defect.

7. PPO Type and FC Profile Configuration

Danfoss FC 300 drives communicate over Profibus with Parameter Process Data Objects (PPO). PPO types 1, 2, 5, and 6 are the standard drive profiles. PPO 1 is the most common and provides:

Confirm the following inside HW Config / STEP 7:

• Slave GSD file: Danfoss FC 300 ships with GSD file DAN_0F70.GSD (or equivalent — check the MCA 101 option's manual for the exact revision). Using a generic "VLT5000" or "FC300PROF" GSD on the wrong firmware revision can produce the symptom where communication starts and then drops on CTW activity.
• Selected PPO type: must match the FC 300's parameter 9-15 (PCD Config Write) layout. If 9-15 is left at default, PPO 1 is recommended.
• I/O length in HW Config: 4 words in / 4 words out for PPO 1 (or 12/12 for PPO 5, etc.). Mismatches cause one-shot aborts, but typically a configuration error, not a runtime drop.

8. CTW and STW Word Mapping (Danfoss FC Profile)

The values reported in the discussion, 0x047C for start and 0x043C for stop, are the canonical Danfoss FC profile control words. Verify that the PLC writes them to the correct PZD word offset of the slave's input area.

If the PLC is sending the right CTW to the wrong offset, the drive will read 0x0000, interpret that as "OFF1," and the motor will not start. That would not, however, generate W34 — so the bus fault is independent of the CTW value itself; the CTW simply happens to be the action that turns the motor (and therefore the noise) on.

9. Step-by-Step Resolution Procedure

Follow these steps in order. Each step should be verified before continuing.

1. Force a single-direction test. Disconnect the motor leads from the FC 300 terminals U, V, W. Run the program with 0x047C. If the bus fault no longer appears, the diagnosis of output-side noise is confirmed. Reconnect only after the next steps.
2. Inspect the motor cable. Use a shielded, symmetric 3-phase cable (e.g. CY-JZ or a screened VFD cable with 100% braid coverage). Confirm the shield is bonded 360° to the EMC gland at both the drive and the motor terminal box. A drain wire floating at one end is a frequent offender.
3. Increase the physical separation. The motor cable and the Profibus cable must follow a separate routing path with at least 200 mm continuous clearance and never share the same ladder/tray. Crossings must be at 90°. The 2 m distance is short enough that any capacitive coupling at low impedance is significant.
4. Install a VFD output filter. Add a Danfoss dU/dt filter (e.g. VLT MCC 101 or MCC 201 sine-wave filter) close to the drive output. This reduces the dV/dt and high-frequency common-mode content by 5–10× and is the single most effective cure when motor cable is short.
5. Add a common-mode EMC kit. On the FC 300, install the manufacturer-recommended EMC bracket kit and ensure the PE conductor returns to the drive's PE terminal through the shortest possible path, not through the backplate.
6. Insert a Profibus repeater or fibre-optic link. As a final hardening step, isolate the drive's Profibus node with a repeater (Siemens 6GK1 500-0AA00) or convert to fibre at the drive side. This breaks the galvanic common-mode path between the PLC and the drive, eliminating the coupling entirely.
7. Reconnect the motor and re-test. With the EMC and routing corrections in place, reconnect the motor leads. Cycle the drive from stop to several reference speeds. Monitor the PLC diagnostic buffer (open the buffer view and click Update after each run) — the failure/return pair should be gone.

10. Verification Tests

11. Preventive Measures for New Installations

• Specify a screened motor cable as part of the panel build, never a generic 4-core PVC cable.
• Add a 200 mm cable routing rule to the cabinet drawing for the Profibus and motor cables.
• For VFD-to-motor distances under 5 m, always include a dU/dt or sine-wave output filter in the BOM.
• For sites with multiple VFDs on a shared Profibus segment, plan for a Profibus repeater at the first drive and convert the segment to fibre for runs above 50 m.
• During commissioning, run a "motor-cable-disconnect" sanity test on every new VFD/Profibus pairing. It costs two minutes and immediately catches the issue documented in this article.

12. Escalation Path if the Fault Persists

1. Replace the FC 300 MCA 101 Profibus option card to rule out an internal option-card fault.
2. Verify the PLC's Profibus interface hardware (the DP port on the CPU itself).
3. Capture a Profibus frame trace with a Softing or Intrelix analyser and look for short bursts of corrupted frames that coincide with the motor current waveform (a spectrum analyzer on the Profibus shield will show the IGBT switching harmonics if the diagnosis is correct).
4. As a last resort, ground the Profibus shield only at the master and float it at the drive side — this breaks the common-mode current loop, but it must be evaluated against the site's EMC policy.