BMW N55 133E10 and 135704 with poor performance but no check engine light: what should I test?
A test-first N55 Valvetronic plan for 133E10 and 135704, covering fault context, live valve-lift data, wiring and mechanical drag.
A 2014 BMW 435i xDrive owner found 133E10 and 135704 returning immediately after clearing, with much worse performance but no check-engine light. I would treat that combination as a repeatable Valvetronic control problem, not permission to order a motor or eccentric shaft. The first job is to learn when the DME deactivates the system and whether the failure begins in the command, electrical path, feedback or mechanism.
Valvetronic fault codes 2014 435i xDrive N55 poor performance no CEL: what should I test?
The exact scan-tool text matters. In the owner report, 133E10 was shown as “Valvetronic system: deactivated, adjustment fault too frequent,” while 135704 was shown as “Valvetronic system: warning threshold value, feedback control deviation exceeded.” I would save that wording with the DME identification because code descriptions can vary across software and scan tools.
Taken together, those descriptions say that adjustment has failed repeatedly and the feedback loop has deviated far enough for the controller to react. They do not identify the failed component. The motor can be unable to move, its integrated position information can be unreliable, the harness or terminals can interrupt power or signals, or the eccentric-shaft mechanism can resist movement.
BMW's N55 training material explains why those possibilities overlap. Valvetronic III uses a brushless servomotor to turn a worm drive and move the eccentric shaft. The sensor used to determine motor and eccentric-shaft position is integrated into that servomotor. The DME therefore judges motion through feedback coming from the same actuator assembly it commands, connected through the vehicle harness.
No check-engine light does not clear the system. A BMW-capable scan can expose manufacturer-specific and shadow information that does not request the dashboard lamp. In this case, the faults returned immediately and the car had a matching performance loss. That is stronger evidence than the lamp state.
Preserve the failure before disturbing it
I begin with a full vehicle scan and save the original report. For both faults, I want present or stored status, frequency, mileage, voltage and any environmental detail the tool provides. I also record companion DME faults. Supply-voltage faults, motor-phase faults, air-mass plausibility faults or several unrelated sensor faults would change the first test.
Next I define the complaint. Does the engine feel weak from the first cold start, only after warming up, only at idle, or only during a loaded pull? Does cycling the ignition restore performance? Is there clicking before start, a change in idle quality, or a drivetrain warning? I do not keep making hard pulls if power delivery is unpredictable.
A secondary Bimmerpost report is useful for method, not as proof of an N55 failure. That S55 owner saw 135704 after the engine became hot and reported requested and actual valve lift stuck at high lift. The different engine and symptoms prevent a direct diagnosis, but the report shows why I would log the fault under the exact temperature and load that produces it.
Before clearing anything, I make a baseline recording where the DME exposes suitable channels:
- Valvetronic or valve-lift requested value
- Valvetronic or valve-lift actual value
- Valvetronic status or activation state
- actuator current or related activation value, if available
- engine speed and load
- system voltage
- coolant and oil temperature
- throttle angle
Channel names vary. I do not invent a substitute because two labels sound similar. The useful capture shows request, response, system state and operating condition on one timeline.
Compare command, feedback and electrical integrity
I first watch what happens before deactivation. If requested lift changes and actual lift follows smoothly until both settle, that sample shows successful control. If the request changes but actual lift stays flat, the next branch includes the motor circuit, actuator and mechanical resistance. If actual lift jumps or drops without a matching command, position feedback or its connection deserves attention. If the DME declares Valvetronic inactive first and the values then park, I avoid treating the parked values as the original failure.
The connector and harness come next. With the ignition off and the correct service precautions followed, I inspect for oil or moisture, corrosion, displaced terminals, weak terminal fit, damaged locks and chafing. I trace the accessible loom where engine movement, heat or earlier valve-cover work could have stressed it. A clean connector shell is not proof that each terminal grips correctly.
Static continuity is only a screen. A strand-damaged wire or loose terminal can pass an unloaded beep and fail when the actuator is working. A technician should use the wiring diagram for that exact VIN and test the specified power, ground, motor and feedback paths under the conditions BMW calls for. I am not publishing generic pin numbers or resistance limits because the correct connectors and test values must come from current vehicle-specific information.
Pico Automotive's Valvetronic case study shows the value of measuring rather than translating a fault description into a part. Its BMW case involved different fault codes, but a motor-phase fault described as an open circuit turned out to involve winding imbalance, while an abnormal sensor-ground voltage kept the harness and control side in the investigation. The lesson applies here: the DME reports what it observed, not which component to buy.
Separate actuator trouble from mechanical drag
Once the external electrical path is credible, I look at how the mechanism moves. BMW's training document shows that the actuator drives the eccentric shaft through a worm mechanism and that engine oil is directed at this assembly. A motor, integrated sensor or internal actuator fault can produce bad motion or bad feedback. Resistance in the eccentric shaft, its support, the gear interface or related valvetrain can also prevent commanded movement.
Current behavior can help, but I do not use it alone. Rising effort paired with slowing or stalled movement supports a loaded mechanism or actuator. Little useful movement with an interrupted electrical pattern points back toward the circuit or motor control. Smooth movement with implausible position feedback raises a different question from a mechanism that audibly binds at a repeatable point.
A controlled workshop test should answer four questions:
- Does the DME issue a command before the fault?
- Does actual position follow that command smoothly?
- Do power and circuit behavior remain stable while movement is requested?
- Does the failure recur at a similar position, temperature or load?
If electrical checks pass but movement slows, sticks or becomes noisy at a repeatable point, mechanical inspection is justified. If the command or circuit becomes unstable first, opening the valvetrain is premature. I would also verify part and repair history. A previously replaced motor does not prove the motor is good, and it does not prove the eccentric shaft is bad.
How this differs from 135808 and 135A08
The related N55 article on 135808 and 135A08 starts with an electrical malfunction and an implausible position signal on a car that reportedly drove normally. This case starts with repeated adjustment failure, feedback-control deviation and poor performance. The components overlap, but the diagnostic emphasis changes.
For 133E10 and 135704, I want to capture the transition into deactivation and connect it to the performance loss. Requested versus actual valve lift, Valvetronic state, temperature and electrical behavior should be recorded together. For 135808 and 135A08, phase balance, position-signal integrity and actuator wiring are more prominent from the opening fault descriptions.
After a repair, I repeat the same cold, warm or loaded condition that produced the problem. I confirm requested and actual movement remain coherent, rescan the DME and verify the faults do not return. Clearing the memory and seeing no lamp at idle is not proof. A matched before-and-after capture is.
Questions owners ask
BMW N55 133E10 and 135704 with poor performance but no check engine light: what should I test?
Save the complete DME fault records, compare requested and actual valve lift during the complaint, inspect the Valvetronic actuator connector and harness, test the circuit under load using the correct vehicle wiring information, and investigate mechanical resistance only after the electrical path is credible.
Does 133E10 prove the N55 Valvetronic motor has failed?
No. The owner's scan tool described 133E10 as repeated adjustment faults causing Valvetronic deactivation. That result can follow an actuator fault, a circuit problem, unreliable position feedback or a mechanism that cannot move as commanded.
Why can 133E10 and 135704 be stored without a check-engine light?
A DME can retain manufacturer-specific or shadow fault information without requesting the dashboard lamp. The missing lamp does not make a repeatable fault false, especially when the car has a matching performance complaint.
Should I replace the eccentric shaft when requested and actual valve lift disagree?
Not from that disagreement alone. First prove actuator power, connection quality and feedback behavior. Mechanical drag becomes a stronger direction when the electrical path remains stable but movement slows, sticks, becomes noisy or fails repeatedly at a similar point.
Sources
- Reddit owner discussion: 2014 435i xDrive with 133E10, 135704 and poor performance
- BMW N55 Engine technical training manual
- BMW Valvetronic III functional description
- Pico Automotive BMW Valvetronic diagnostic case study
- Bimmerpost discussion: 135704 and temperature-dependent valve-lift deviation
- Related Learn article: N55 135808 and 135A08 electrical and plausibility tests