SINAMICS DCM Reversal Braking: Resolving 4Q Drive Drift Issues

David Krause12 min read
SiemensTroubleshootingVFD / Drives
Licensed PE Working through this on a live machine? A Maine-licensed engineer can take it from here — included with IMD hardware, by the hour for everything else. Book an engineer

Overview

The SINAMICS DCM (DC Master) is Siemens' modular six-pulse or twelve-pulse converter for DC motors in the 6 kW to 1,800 kW range. In 4-quadrant (4Q) operation, two antiparallel thyristor bridges (B6C) allow motoring and regenerating torque in both directions of rotation, making the unit suitable for reversing drives such as wire-draw blocks, coilers, payoffs, hoists, and revolving-blade bar cutters.

Field reports on steel-bar cutting lines indicate that a 4Q SINAMICS DCM, while running stably in one direction, drifts past the target position on a reversal command: the drive does not brake to a full stop within the required window and the blade misses its index point. This article consolidates the parameter, command-source, and commissioning practices that eliminate the drift, with the application diagram shown in SINAMICS DCM Field Reversal (Application Description, Siemens Support) used as the architectural reference.

Problem Description

Symptom profile observed on 4Q reversing applications:

  • Forward-to-reverse command: motor continues to rotate in the original direction for hundreds of milliseconds, with measurable overshoot before deceleration begins.
  • Setpoint polarity change (e.g. ±10 V analog input): actual speed does not follow the new setpoint, or the drive enters a free-wheel condition.
  • Application: a revolving-blade bar cutter requires a deterministic stop at a defined angle. A drifting rotor causes the blade to land in the wrong position, producing scrap or a mechanical jam.
  • Encoder feedback is fitted (typical pulse count 1024 or 2048 ppr), so the issue is not an uncontrolled open-loop ramp.
Safety: DC link voltages on a SINAMICS DCM exceed 900 V DC at 3 AC 690 V supply. Lock-out/tag-out, wait at least 5 minutes after mains removal for DC-link discharge (consult the warning label on the unit), and verify with a properly rated meter before touching terminals.

Root Cause Analysis

Drift on a 4Q reversing SINAMICS DCM almost always traces to one of three interacting sources:

  1. OFF-command wiring and pulse cancellation. If the controller issues OFF2 (pulse cancellation) at the same instant the speed setpoint is reversed, the firing pulses are removed, the motor is effectively torque-free, and a load with stored kinetic energy continues to rotate (free-wheel). The rotor is then driven by load inertia alone and overshoots the target.
  2. Setpoint filtering and ramp-function generator (RFG) mismatch. An analog setpoint arriving through a heavily smoothed input, or an RFG with an excessively long ramp-down time (p508) and wide rounding constants (p509, p510), introduces a delay between the setpoint change and the torque reversal required for regenerative braking.
  3. Torque limit and field configuration. Incorrect torque limits (p501/p502), current-controller tuning, or a field-weakening setpoint that does not match the speed range can leave the drive in field-weakening during the brake phase, reducing available braking torque at the higher speeds where braking is most needed.

For 2Q applications the same problem is solved by the "Braking to defined speed" function block described in the Siemens DCM Field Reversal PDF. On a 4Q unit the regenerative bridge is always present, so the diagnosis focuses on command sequencing, RFG, and torque limits rather than field reversal.

Prerequisites

Before changing parameters, confirm the following:

  • Commissioning tool: STARTER (≥ V5.4) or SINAMICS Startdrive (V15.1 or later) installed on a PC connected via PROFIBUS, PROFINET, or the X21 RS232 service port of the SINAMICS DCM Control Unit (CUD).
  • Firmware: SINAMICS DCM firmware V1.4 or later (current release line: V1.6 HF3 as of the SINAMICS DCM product support page on Siemens Industry Online Support). Earlier versions had different default values for p503 and p504.
  • Drive order (article number): 6RA80xx-xxxxx-xxxx on the rating plate, which confirms 4Q (B6C) versus 2Q (B6A) topology.
  • Motor nameplate data complete and entered in p100 through p114 (armature, field, thermal, encoder).
  • Encoder type confirmed: HTL/TTL on X177 of the CUD, or DRIVE-CLiQ encoder via the SMC30 (option K50).

Command and Setpoint Source Mapping

Verify that the binary commands and the speed setpoint are bound to the correct terminals and that the polarity is interpreted as a true bipolar command, not as enable with separate direction bits.

Parameter Default Recommended (4Q reversal) Function
p600.0 (Control word source) 1 (terminals) 1 or 2070 (PROFIdrive) Source for STW1
p601 (Speed setpoint source) 1 (analog input main) 1, 6 (motor potentiometer), 2070 Source for n-setpoint
p654.0 (OFF1 source) r722.0 r722.0 / bit field Coast-to-stop / ramp stop
p655.0 (OFF2 source) r722.1 r722.1 Pulse cancellation (no torque)
p656.0 (OFF3 source) r722.2 r722.2 / PROFIdrive STW1 bit 2 Fast stop, max braking
p650.0 (ON command) r722.0 r722.0 Run enable

Setpoint interpretation rules:

  • If p601 = 1 (analog) and p600.1 = 0, a unipolar 0–10 V signal expects a separate direction bit; a bipolar ±10 V signal gives direct speed and direction.
  • For 4Q reversing, prefer bipolar ±10 V on the main setpoint (X177, AI0) so that one parameter change (p600.1 = 1) and one analog wire deliver signed speed reference.
  • If a current setpoint (4–20 mA) is used, set p600.1 = 0 and use a separate digital input for direction; the drive cannot infer direction from a unipolar current signal.

Optimization Procedures

Parameter optimization must be run to populate the closed-loop controllers. The drive cannot reverse cleanly until these have been completed:

  1. p50051 = 1 – Armature current controller optimization with the rotor locked. The drive will pulse the armature, measure resistance and inductance, and write p110, p111, p112, p155, p156.
  2. p50051 = 2 – Speed controller optimization. The motor accelerates to a test speed (default 20 % of p101), runs the inertia identification, and writes p225, p226 (Kp/Tn) plus the torque precontrol factor p531.
  3. p50051 = 3 – Field current controller optimization. Runs at reduced field and writes p275, p276.
  4. p50051 = 4 – Field weakening characteristic (EMF controller). Required when p101 > p114 · supply voltage / 1.35.
  5. p50051 = 5 – Speed measurement optimization for encoder feedback.

Confirm each step ends with alarm A07980 cleared and the relevant r parameters non-zero. Do not skip step 2 (speed controller); without it, the regenerative current controller cannot stabilize the brake and the drive will overshoot.

Ramp-Function Generator (RFG) and Smoothing

The RFG is the dominant variable for the perceived "drift". Tighten it for deterministic stop:

Parameter Meaning Drift-fix value (bar-cutter application) Notes
p503 Ramp-up time (n=0 to n-max) 1.0 s Lower bound: current-controller bandwidth
p504 Ramp-down time (n-max to 0) 0.5 s Symmetrical with p503 by default
p505 Initial ramp-up rounding 0.05 s PT1-like smoothing of start
p506 Final ramp-up rounding 0.05 s Removes setpoint chatter at the start
p507 Initial ramp-down rounding 0.05 s Critical for reversal braking
p508 Final ramp-down rounding 0.05 s Reduce if stop window is short
p509 RFG bypass threshold 2 % of n-max Disable RFG near zero to avoid dead-band
p222/p223 Setpoint smoothing (PT1) 0 / 0 ms Remove if previously set high

For applications where the brake window is shorter than the RFG can deliver, switch p502.4 = 1 to use OFF3 (fast stop) on the reversing edge. OFF3 is a separate ramp (p513, p514, p515) that always commands maximum torque, regenerating the load into the DC link. Verify p518 (current limit for OFF3) is set to the drive's rated regenerative current (e.g. ±1.2 × I-armature rated) so the brake is not current-limited.

Torque and Current Limits for Regenerative Braking

Parameter Function Setting
p501 (positive torque limit M1) Forward motoring max 100 % (p100)
p502 (negative torque limit M2) Reverse motoring / forward braking max 100 %
p503/p504 alternatives Symmetric torque shaping Set p502 = p501
p518 (OFF3 torque limit) Fast-stop peak current 1.5 × I-rated, hardware clamps apply
p104 (field current setpoint) Constant field Match motor nameplate
p105 (rated EMF) Defines base speed / field-weakening corner Entered by p50051 = 4

Braking torque available is governed by:

M_brake = (2 / 3.14) · I_arma · Φ_field
where Φ_field is the flux produced by the field current p104.

During a fast reversal at high speed, the EMF generated by the spinning armature is higher than the DC-link voltage. If field weakening is in effect (operating above base speed), the flux is reduced, so the available braking torque drops at exactly the moment it is needed most. Either hold the field at p104 during the brake (configure p105 wide so base speed is at the application's top speed) or schedule field weakening only outside the reversing window.

STO / Safe Torque Off Behaviour

If the safety circuit is using STO via X122 of the SINAMICS DCM (terminals EP, 24V, STO-A, STO-B), removing STO at the same moment the run command changes to reverse will result in pulse cancellation followed by re-pulsing. During the OFF window the motor free-wheels, and the rotor can move several centimetres. Sequencing rules:

  1. Maintain STO closed (energized) throughout the reversal.
  2. Use OFF1 (p654) for normal stop, OFF3 (p656) for fast stop.
  3. Do not release STO and then re-engage the drive with the new direction bit; the CUD requires a minimum of 5 ms for safe pulse suppression/restoration, and the rotor will move during that window.

Mechanical Coupling Audit

Drift is sometimes mechanical, not electrical. Check:

  • Backlash in the gearbox or coupling. Backlash between the encoder and the motor shaft causes a fixed angular offset that can look like drift on reversal.
  • Encoder mounting: belt-driven encoders introduce slip; a non-torsionally-rigid coupling introduces wind-up proportional to acceleration.
  • Coupling alignment: misalignment of more than 0.05 mm at the encoder end can cause resonance in the speed loop and visible drift at zero setpoint.

Step-by-Step Resolution

  1. Identify drive article number, e.g. 6RA8085-6DS22-0AA0 (4Q, 850 A, 480 V). Confirm 4Q topology with rating plate and the order code suffix DS/ES.
  2. Connect with STARTER or Startdrive. Back up the project (p0971 = 1, or use the "Upload from drive" wizard).
  3. Run optimization p50051 = 1, 2, 3, 4, 5 in that order with the motor uncoupled from the load.
  4. Re-couple the load and verify p501 = p502 = 100 %; check r019 (actual torque) at standstill is < 5 % of rated.
  5. Set the RFG as in the table above and lower p504 to 0.5 s.
  6. Confirm the analog input wiring gives a bipolar ±10 V on AI0 (X177.1 = +10 V, X177.2 = signal, X177.3 = GND).
  7. Program the controller to issue OFF1 + bipolar setpoint change as a single transaction; do not toggle STO between directions.
  8. Issue a small reversal (±10 % of n-max) and capture r021 (speed actual) in the trace; braking must reach the new setpoint within p504 + 2 × p507. If not, halve p504 and re-test.
  9. Move to the full reversal command; record r532 and r019 across the transition. Braking torque must reach p501 within 50 ms of the setpoint zero-crossing.
  10. Save the project to the CF card of the CUD (p0800.1 = 1, p0802 = 1) and to the engineering PC.

Verification

After the changeover, perform the following checks before returning the line to production:

Check Method Pass criterion
Setpoint response linearity Inject +50 %, then -50 %; read r532 No dead-band > 2 % of n-max
Braking torque envelope Trace r019 during reversal Reaches p501 within 50 ms
No free-wheel interval Trace r010 (status word) and r532 together Bits 2/6 remain inactive throughout reversal
Field stability Trace r037 (EMF) during reversal Within ±5 % of p105 · p104 / p105 ref
Mechanical drift Index marker on shaft, read in trace at n=0 < 1 encoder pulse
Alarm A07979 cleared Read r070 / r2110 No active alarms

Troubleshooting Matrix

Symptom Likely cause Parameter / Action
Motor free-wheels through reversal OFF2/STO cancelling pulses Check p655 and the X122 STO wiring; keep STO energized
Drift of several hundred ms Long ramp-down Reduce p504 to 0.3–0.5 s, lower p507/p508 rounding
Setpoint changes polarity but speed does not Unipolar input with no direction bit Switch to bipolar ±10 V or add a digital direction input
Brake torque collapses above 80 % of n-max Field-weakening active Re-run p50051 = 4, raise base speed to cover operating range
Oscillation during the brake phase Speed controller gain too high for inertia Re-run p50051 = 2, lower p226 Tn by 30 %
Alarm A07931 (drive blocked) on reversal Torque limit too low for inertia Raise p501 / p502 to 100 % each
Alarm A07944 (line failure) DC-link ripple during regen Check mains quality, reduce p518 (OFF3 torque limit) by 10 %
Mechanical backlash on reversal Coupling / encoder slip Inspect mechanical, re-couple

FAQ

Why does my 4Q SINAMICS DCM drift on reversal instead of braking immediately?

The drive is most likely entering pulse cancellation (OFF2 or STO) at the same instant the setpoint changes sign, so the motor free-wheels on load inertia. Re-sequence the OFF commands so OFF1 is used for the stop and STO remains energized across the reversal. If the problem persists, the regenerative bridge is current-limited by p501/p502 or by field-weakening above base speed.

What RFG ramp times should I use for a deterministic stop on a steel-bar cutter?

Start with p503 = 1.0 s, p504 = 0.5 s, and rounding constants p505–p508 = 0.05 s. If the brake window is shorter than 0.5 s, enable OFF3 via p502.4 = 1 and set p513/p514/p515 to match the mechanical stop window. The OFF3 ramp uses the drive's full torque limit p518, so it always brakes harder than OFF1.

Do I need field reversal for 4Q reversing with SINAMICS DCM?

No. A 4Q SINAMICS DCM has two antiparallel thyristor bridges, so reversal is achieved by re-firing the second bridge. Field reversal is only required for 2Q units, as described in the DCM Field Reversal application note. Using field reversal on a 4Q unit is unnecessary and will slow the response.

My drive enters OFF2 when I issue a reverse command from a PLC. What is wrong?

Confirm that bit 1 of the control word (STW1) is wired to a constant TRUE on both the PLC and the drive side, and that OFF2 is sourced to r722.1 (terminal DI 1) rather than to a PROFIdrive bit you are toggling inadvertently. The 1 in STW1.1 means "no OFF2"; setting it to 0 fires OFF2, which removes the drive pulses and causes the rotor to free-wheel on load inertia.

Which firmware version is recommended for SINAMICS DCM 4Q reversing drives?

SINAMICS DCM firmware V1.6 (current release V1.6 HF3) is the supported baseline. Versions prior to V1.4 had different default values for the speed-controller precontrol and the RFG rounding, and earlier behavior with bipolar setpoints could introduce a small setpoint dead-band near zero. Update via the CUD's CF card slot, following the upgrade instructions in the Siemens Industry Online Support FAQ for SINAMICS DCM.

Back to blog