Resolving S7-1200 PID_Compact Output Zero at 100% Input

Problem Overview

The PID_Compact instruction (FB 1130 / FB 1131 / FB 1132 depending on TIA Portal version) on the S7-1200 platform can enter a permanent "inactive" or "stopped" operating mode in which the manipulated variable collapses to 0% and the controller refuses to restart. The four most common triggering conditions observed in the field are:

1. The process value (Input) reaches 100% of the configured measuring range while the manipulated variable is also at 100%.
2. The PLC is power-cycled while the process is far from setpoint, especially with a temperature loop where PV has cooled to ambient.
3. Output saturation persists beyond the configured watchdog time, with no path for integral recovery.
4. The analog input or scaled PV leaves the configured InputScaling limits, triggering the safety transition.

The reported symptom is identical in all four cases: Retain.CtrlInactive = TRUE, Output = 0.0, and the controller cannot be brought back online without a CPU restart or a project re-download. The root cause is a combination of integral wind-up and the safety logic built into PID_Compact.

Affected Versions and Instruction Revisions

For S7-1200 G2 CPUs (6ES7 214-1xxxx-0SB0 and 6ES7 215-1xxxx-0SB0), the official PID functionality documentation in the TIA Portal Cloud describes the extended state machine that the V4.x instruction implements, including the explicit handling of the inactive-mode conditions described in this article.

Root Cause: Integral Wind-Up and the Inactive Safety State

The phenomenon of "controller wind-up" predates the S7-1200 and was first documented on pneumatic and electro-mechanical controllers. It occurs when the integral component of the PID algorithm continues to accumulate the control error while the physical actuator has already reached its travel limit. The integrator grows without bound, and when the process value finally changes, the accumulated integral term is so large that no realistic change in the error can reduce it in a reasonable time. The controller effectively becomes a fixed-output device, ignoring the process.

In PID_Compact the failure is amplified by a second mechanism: when the instruction detects that the input and the output are both at 100% (a "coincident saturation" condition), the safety logic transitions the controller to the Inactive state. The output collapses to 0%, the controller holds the integral term at its last value, and the only way out is a mode change or a CPU restart.

Mathematical Model

The discrete PID_Compact integrator maintains the following state across scans:

I_sum(n) = I_sum(n-1) + (Kp / Ti) * e(n) * Ts

Where:

• I_sum = internal integrator state (tag sRetain.fIItVal)
• Kp = proportional gain (sBackUp.r_Gain)
• Ti = integral action time in seconds (sBackUp.r_Ti)
• e(n) = error (setpoint minus process value) at scan n
• Ts = sampling time of the PID_Compact instance

When the output is clamped at 100% for many scans, I_sum grows linearly with the error. If the loop has been at setpoint 180 °C with PV at 30 °C (a typical heating scenario), the integrator is fed 150 °C of error on every scan. Even with a modest Kp/Ti = 0.05 and a one-second Ts, the integrator can accumulate 7.5 units per second, requiring tens of minutes to "unwind" once the process finally overshoots the setpoint.

PID_Compact Mode State Machine

The instruction moves through the following operating states. The transitions that lead to the Inactive state are the ones that lock the output at 0%.

Diagnostic Indicators

When the controller is locked in the inactive state, monitor the following tags in the PID_Compact instance DB (right-click the DB in TIA Portal and select "Monitor / Modify" with all instance tags visible):

PID_Compact Error Code Reference

Solution 1: Configure Output Limits Below 100%

The most direct fix reported in the field is to limit the maximum output to 99% (or any value just below 100%) so the safety logic never sees the coincident-saturation condition. This is the approach Siemens documents in their PID_Compact configuration guidance.

1. Open TIA Portal and load the S7-1200 project containing the affected PID_Compact instance.
2. Navigate to Project Tree → PLC_x → Program blocks → System blocks → PID_Compact instances → [Instance DB].
3. Open the instance DB and switch to the Configuration tab.
4. Expand Basic settings → Output settings.
5. Set Output upper limit to 99.0 (or 99.5 if 99% still causes the trip in your process).
6. Set Output lower limit to 0.0 unless the actuator has a non-zero minimum (e.g., a 4-20 mA valve with a 4 mA minimum).
7. Compile (Project → Compile → Software) and download the project to the CPU.
8. Either power-cycle the CPU or trigger a PID_Compact re-initialization (set Retain.CtrlInActive = FALSE and re-assert bSetPointValue).

Solution 2: Match InputScaling to the Real Process Range

Configure the input scaling so that the scaled PV cannot exceed 100% under any realistic process condition. Mismatched scaling is the second most common cause of the inactive state.

If the configured UpperPointIn exceeds the real sensor maximum, the scaled PV will rise above 100% on overrange conditions and trip the safety logic even though the physical sensor is still in spec. Always scale to the worst-case realistic PV plus a 5% engineering margin.

Solution 3: Mode Switching for Manual Unwind

When the controller has wound up, switching to Manual mode allows the integral term to drain to the value that matches the current output. The PID_Compact supports bumpless transfer: when you switch from Manual back to Automatic, the integral term is initialized to the manual output value, eliminating the bump on transition.

Structured Text Example for Manual Unwind

// Manual unwind when PV reaches the upper limit
IF "PID_DB".Input >= 99.0 AND "PID_DB".Output >= 99.0 THEN
    // Force manual mode to let integral drain
    "PID_DB".eMode := ePIDModeManual;
    "PID_DB".bSetPointValue := FALSE;
    "PID_DB".ManualEnable := TRUE;
    "PID_DB".ManualValue := "PID_DB".Output;  // Hold current output
    
    // Wait for unwind (or use a process-specific timer)
    IF "unwind_timer".Q THEN
        // Switch back to automatic with bumpless transfer
        "PID_DB".ManualValue := "PID_DB".Output;
        "PID_DB".eMode := ePIDModeAutomatic;
        "PID_DB".bSetPointValue := TRUE;
        "PID_DB".ManualEnable := FALSE;
        "unwind_timer".IN := FALSE;
    END_IF;
END_IF;

Ladder Logic Equivalent

|   Input >= 99.0   Output >= 99.0   unwind_done   |
|---|----[/]----[/]----[/]------------( S )---|
|   "PID_DB".eMode = ePIDModeManual            |
|                                               |
|   unwind_done   unwind_timer.Q                |
|---|----[/]------|------[/]----------( )---|
|   "PID_DB".eMode = ePIDModeAutomatic          |

Solution 4: Power Cycle Recovery Configuration

When the PLC loses power with the process far from setpoint, PID_Compact can enter the inactive state on the next startup. The fix depends on the desired restart behavior.

Option A: Configure Restart Mode

1. Open the PID_Compact instance DB Configuration tab.
2. Navigate to Basic settings → Behavior at CPU restart.
3. Set Mode at restart to one of: Inactive (operator must bring online), Last active mode (resume where left off), or Automatic (force automatic on restart).
4. For temperature loops that cool at restart, choose Inactive and let the operator verify the process before going automatic.

Option B: Initialize Setpoint from Process Value in OB100

For loops where the process is always allowed to be at PV on startup, pre-load the setpoint to the current PV in the warm restart OB. This prevents the controller from trying to drive a 150 °C delta immediately after a power cycle.

// OB100 - Warm restart
// On power-up, set setpoint equal to current process value
IF "PV_Scaling_OK" THEN
    "PID_DB".Setpoint := "PID_DB".Input;
    "PID_DB".eMode := ePIDModeAutomatic;
    "PID_DB".bSetPointValue := TRUE;
END_IF;

Option C: Configure Retain Settings

Mark the critical state tags as Retain in the instance DB properties so the integral state survives the power cycle. The S7-1200 classic supports up to 10 kB of retain data; the G2 supports more. Verify the retain budget on your CPU before enabling.

Solution 5: Anti-Windup Implementation

PID_Compact includes built-in back-calculation anti-windup that activates when the output reaches the configured limits. The behavior is governed by these parameters:

The built-in anti-windup is passive (back-calculation with gain = 1.0). It only works if the output limits are properly configured. If the limits are 0-100% and the process regularly hits them, the integrator will still wind up.

Explicit Anti-Windup in Structured Text

For critical applications where the built-in anti-windup is insufficient, add explicit clamping logic in the OB that calls the PID_Compact block:

// Explicit anti-windup clamp on the integrator state
IF "PID_DB".sRetain.fIItVal > "PID_DB".Config.OutputScaling.UpperPointOut THEN
    "PID_DB".sRetain.fIItVal := "PID_DB".Config.OutputScaling.UpperPointOut;
END_IF;

IF "PID_DB".sRetain.fIItVal < "PID_DB".Config.OutputScaling.LowerPointOut THEN
    "PID_DB".sRetain.fIItVal := "PID_DB".Config.OutputScaling.LowerPointOut;
END_IF;

Verification and Commissioning Tests

After applying any of the above fixes, run the following test sequence in TIA Portal with the CPU in RUN and the PID_Compact instance online:

Test 1: Step Response

1. Set setpoint to 50% of measuring range.
2. Confirm controller is in Automatic mode (eMode = ePIDModeAutomatic).
3. Apply a process disturbance (load change, setpoint step, or actuator pulse).
4. Observe: Output responds within 2-3 sample times.
5. Pass criterion: Output settles within ±5% of steady-state value within 10 × Ti.

Test 2: Saturation Recovery

1. Drive the output to 99% (raise setpoint above PV).
2. Hold for 60 seconds at saturation.
3. Lower setpoint below PV to force the output down.
4. Observe: Output responds within 5 sample times, no transition to inactive state.

Test 3: Power Cycle

1. Set setpoint far from process value (e.g., setpoint 180 °C, PV 30 °C).
2. Power cycle the S7-1200 (STOP / RUN or power removal).
3. Observe: PID_Compact resumes in the configured restart mode.
4. Pass criterion: Output drives toward setpoint without entering inactive state.

Test 4: Bumpless Manual-to-Auto Transfer

1. Start in Automatic mode with output at 50%.
2. Switch to Manual mode (eMode = ePIDModeManual).
3. Change ManualValue to 30%.
4. Switch back to Automatic mode.
5. Pass criterion: Output remains near 30% during transition, no bump > 2%.

Troubleshooting Matrix

Platform-Specific Considerations

S7-1200 Classic (CPU 1211C / 1212C / 1214C / 1215C / 1217C)

• CPU firmware V4.0 or later required for PID_Compact V2.x with the configurable InputScaling.
• Use TIA Portal V15.1 or later for the full feature set including scaled input monitoring.
• Maximum number of PID_Compact instances per CPU is firmware-dependent: 1211C supports 6, 1212C supports 8, 1214C supports 16, 1215C supports 16, 1217C supports 16.
• Minimum recommended SampleTime = 0.1 s for reliable operation with PID_Compact V1.x and V2.x. V3.x on newer CPUs can run at 0.05 s.

S7-1200 G2 (CPU 1212G / 1214G / 1215G / 1217G)

• New CPU firmware supports PID_Compact V4.x with the extended state machine described in the TIA Portal Cloud PID functionality documentation.
• Improved restart behavior with separate warm and cold restart handling.
• Extended diagnostic information in the instance DB including per-transition state flags.
• Multi-instance support for code reuse across multiple PID loops.

Comparison with Other S7-1200 PID Instructions

PID_3Step uses position feedback from the valve and does not exhibit the wind-up failure mode described in this article. If your process is compatible with PID_3Step (i.e., you have a motorized valve with position feedback), it is the more robust choice for applications that experience long saturation periods.

Field Examples

Example 1: Injection Molding Barrel Heating

A plastic injection molder used PID_Compact to control barrel zone temperature with setpoint 230 °C and ambient start temperature 25 °C. The integrator accumulated over 4 hours of operation per shift. On overnight cool-down, the PV dropped to 25 °C while the integrator remained at its wound-up value. The next morning, the controller immediately went to 100% output, the PV overshot to 245 °C, and the controller tripped to inactive.

Applied fix: Output upper limit reduced from 100% to 99%, setpoint initialized from PV in OB100, and integrator clamped to output limits in the user program.

Example 2: HVAC Supply Air Temperature

An AHU supply air loop with setpoint 13 °C and outdoor air at 35 °C saw 100% cooling valve demand in summer. After a power outage, the controller would not restart and displayed Retain.CtrlInactive = TRUE.

Applied fix: InputScaling adjusted so the scaled PV cannot exceed 100% on overrange, and the restart mode was set to "Last active mode" with the integrator state marked as Retain.

Example 3: Pump Discharge Pressure

A booster pump station with PID_Compact controlling VFD speed on a centrifugal pump. Demand spikes during peak hours drove the output to 100% for several minutes. After each spike the controller went inactive.

Applied fix: Output upper limit set to 95% to leave headroom for the VFD ramp-down, and the integral time was increased from 5 s to 20 s to reduce integrator accumulation rate.

Best Practices Summary

1. Always set output upper limit to 99% (or 99.5%) instead of 100% on critical control loops.
2. Match InputScaling to the real process range plus a small engineering margin; never let the scaled PV exceed 100% under any realistic condition.
3. Mark integrator state tags as Retain so the controller resumes cleanly after a power cycle.
4. Initialize the setpoint to the current PV in OB100 for loops that start far from setpoint.
5. Use bumpless transfer (set ManualValue = current Output) when switching from Manual to Automatic.
6. Monitor sRetain.i_wError and sRetain.i_wWarning in the HMI for early warning of the inactive condition.
7. For processes with frequent or long saturation, consider PID_3Step for motorized valves or implement external anti-windup clamps.
8. Run the four commissioning tests (step response, saturation recovery, power cycle, bumpless transfer) before handing the loop over to operations.

FAQ