The shutdown controller reduces the fuel demand, the servo regulator converts that demand into current, the servo valve meters hydraulic oil, the actuator moves the stop-ratio valve, and the resulting gas pressure returns as P2 feedback. Follow that path in order. At 2,000-2,300 RPM, the reported sequence is a 0.5-0.8% separation between SRV command and feedback, an SRV position of 14-18%, continued P2 regulation, and a trip about eight minutes after the shutdown command.
Where does the shutdown control path stop?
The SRV is part of a pressure-control loop, not simply a position-control loop. The controller moves the valve as required to make actual P2 equal its reference. A visible command-to-position error can therefore be either the fault or the loop's response to a pressure disturbance.
| Path element | Reading to capture | Branch decision |
|---|---|---|
| Shutdown and speed control | Speed, fuel demand, shutdown elapsed time | If fuel demand stops decreasing before the speed plateau, trace the upstream shutdown logic. If demand continues downward, follow the command to the servo regulator. |
| Servo regulator |
SRV command, feedback, and servo current |
If command changes but current does not respond, inspect configuration or output circuitry. If current responds, continue to the hydraulic stage. |
| Hydraulic actuator | Valve motion and current required at each stroke position | Abnormal current confined to part of the stroke indicates sticking, binding, leakage, or a servo-valve problem. |
| Gas-pressure loop |
P2 reference and actual pressure |
If actual pressure follows reference, position error alone does not prove bad pressure control. If pressure remains high, determine why fuel flow is not falling. |
Use the controller's configured signal identifiers when collecting these points; no tag addresses or communications ports are specified. The relevant path is the internal control, electrical, hydraulic, and process chain rather than an assumed network fault.
Is the position error electrical or mechanical?
Layer one first. Inspect the LVDT wiring, terminations, shielding, excitation, and both feedback channels before changing calibration. Trend each channel separately where the control system exposes them. A simultaneous disturbance on both channels points toward real actuator motion or shared excitation; one unstable channel points toward its transducer, wiring, or input path.
| Observation | Meaning | Next check |
|---|---|---|
Both LVDT signals move with the physical stem |
The feedback system is probably reporting real motion | Compare servo current with motion through the 14-18% region. |
| One channel jumps while the stem remains steady | Channel-specific electrical or transducer fault | Correct wiring, termination, or the affected LVDT; then calibrate if its output or mounting changed. |
| Feedback changes but an external position measurement does not | Calibration or feedback scaling mismatch | Verify calibrated feedback against physical travel. |
| Command changes and the stem hesitates or oscillates | Hydraulic or mechanical response problem | Inspect servo current, oil condition, filters, actuator leakage, and binding. |
A 0.5-0.8% error is a useful trigger for investigation, but its significance depends on whether it is steady, oscillatory, channel-specific, or concentrated at one stroke position. Preserve synchronized trends rather than comparing separate snapshots.
Does servo current expose a hydraulic restriction?
On this actuator arrangement, negative servo current opens the valve against the closing spring. Compare the current needed to hold several reference positions, paying particular attention to the 14-18% region where the shutdown disturbance appears.
If the current becomes more negative only in that region while nearby positions require normal current, the regulator is working harder to overcome a local mechanical or hydraulic defect. Recurring causes include actuator sticking, internal binding, worn actuator surfaces, leaking O-rings, or a servo valve that cannot meter oil smoothly. If current changes normally but neither position nor pressure responds, check hydraulic supply and the actuator manifold before changing control settings.
GE-design heavy-duty gas turbines use lube oil as actuator hydraulic fluid. Contamination can obstruct the servo valve's small internal passages and create hysteresis or unstable motion. Review the latest oil-cleanliness test and inspect the actuator manifold's last-chance filters. A neglected filter can restrict flow; a ruptured filter can release trapped contamination into the servo valve and actuator.
Is P2 regulation masking the actual fault?
Plot P2 reference, actual P2, SRV command, both position-feedback channels, servo current, speed, fuel demand, and shutdown elapsed time on one time base. The order of movement identifies the initiating stage.
| First change in the trend | Interpretation | Decision |
|---|---|---|
P2 reference stops decreasing |
The pressure loop is being told to maintain pressure | Trace the reference selection, shutdown sequence, and active control limitation. |
Reference decreases but actual P2 remains high |
Fuel flow is not decreasing as commanded | Follow servo output through the valve, actuator, and gas path. |
| Servo current changes before position oscillates | The controller is driving the disturbance or correcting another measured change | Compare the current change with reference and feedback inputs. |
| Position moves before servo current reacts | An external hydraulic, mechanical, or pressure disturbance is moving the valve | Inspect actuator integrity, servo stability, oil, and supply pressure. |
| Trip occurs near eight minutes | The trip is time-correlated with the shutdown sequence | Read the first-out alarm and active trip logic; elapsed time alone does not identify the initiating condition. |
The decisive process question is why fuel flow is not decreasing enough for speed to continue falling. Do not tune the position loop merely to hide a pressure-loop response.
Should calibration or null bias be changed?
For the majority of GE-design heavy-duty gas turbines with TMR Mark control systems, the typical null bias current setting is 2.67. The stated adjustment limits are 1.3-4.0. A value other than 2.67 is justified only after an external measurement shows that physical position does not match calibrated LVDT feedback.
Measure valve travel with an independent instrument, such as a dial indicator with a magnetic base, and compare it with the calibrated position. Small differences of a few thousandths of an inch are not significant for the SRV pressure-control function. Position accuracy matters less than stable pressure response because the regulator moves the valve to whatever position produces the requested P2.
| Maintenance action | Configuration action |
|---|---|
| Replace only a new servo valve | Enter and save the default null bias value of 2.67 before operation. Do not run autocalibration solely for this replacement. |
Replace or reposition an LVDT
|
Run the applicable calibration because feedback output or geometry changed. |
| Remove and reinstall or refurbish the actuator/valve assembly | Calibrate after confirming the physical stroke and transducer installation. |
| Find a physical-position versus feedback mismatch | Correct the mechanical or calibration cause before considering a null-bias adjustment within 1.3-4.0. |
Refurbished or repaired servo valves require separate scrutiny because their failsafe-spring adjustment may not match a new manufacturer-supplied valve. Do not use null bias to compensate for sticking, contamination, leakage, or an incorrect spring adjustment.
How should the resolving branch be tested?
- Retrieve the first-out trip indication and correlate it with the eight-minute shutdown timestamp.
- Record synchronized high-resolution trends for speed, shutdown demand, fuel demand,
P2reference and actual pressure,SRVcommand and feedback, bothLVDTchannels, and servo current. - Inspect feedback wiring and compare each
LVDTchannel with independently measured physical travel. - Exercise the valve through the operating range using the approved maintenance function. Look for extra negative current, hesitation, or oscillation specifically through 14-18% travel.
- Test lube-oil cleanliness, inspect the last-chance filters, and examine the actuator and servo valve for leakage, sticking, binding, or wear.
- Restore the null bias setting to 2.67 when replacing only a new servo valve. Calibrate only when an
LVDT, its mounting, or actuator stroke has changed. - Repeat a controlled shutdown and verify that fuel demand, actual
P2, valve response, and speed decrease continuously through 2,300-2,000 RPM without the prior command-feedback separation or trip.
FAQ
Why does the GE Frame 9E SRV fluctuate at 14-18%?
Compare servo current with physical valve travel in that stroke region. A localized increase in negative current indicates sticking, binding, actuator leakage, servo-valve trouble, or contaminated hydraulic oil.
Why does P2 remain controlled when SRV command and feedback differ?
The SRV regulates pressure, so the controller moves it to the position required to match actual P2 with its reference. Determine whether the pressure reference, actual pressure, servo current, or valve motion changes first.
Do I autocalibrate after replacing an SRV servo valve?
No, not when only a new servo valve is replaced. Save the default null bias value of 2.67, then perform the final verification: repeat the shutdown and confirm stable P2, continuous speed decay through 2,300-2,000 RPM, and no trip.