QYS-EFI-16 - Clean CKP and CMP Signals Are Not Proof of Correct Cam/Crank Correlation
Prove each signal first. Prove the relationship second. Assign the cause only after the failed layer has been isolated.
If CKP and CMP each produce clean, believable signals, engine synchronization and mechanical timing must be correct.
Each qualifying signal proves only its own validity at the tested point and condition. Correct CKP-to-CMP correlation is a separate proof question.
Purpose
This QYS separates four diagnostic questions that are easy to collapse into one: CKP signal validity, CMP signal validity, CKP-to-CMP correlation validity, and cause assignment after a correlation failure.
Its job is to stop clean individual signals from being promoted into synchronization proof and to stop a failed correlation result from being promoted into a parts diagnosis.
Core Idea
A crankshaft position signal and a camshaft position signal can each be electrically credible, stable, and recognizable while their reference events are still wrong relative to one another.
What the symptom seems to prove: Two clean waveforms mean the crank and cam are synchronized correctly.
What it actually proves: Two qualified waveforms establish two individual signal findings. They do not establish the relationship between those signals unless the relationship itself was captured and compared under authorized conditions.
Correlation is its own evidence layer. It requires valid source signals plus synchronized relationship evidence and the correct exact-vehicle interpretation authority.
Why This Matters In The Real World
This matters because a no-start, hard start, sync fault, timing code, or VVT complaint can push a technician toward mechanical timing even when the only thing actually proved is that both sensors are producing believable outputs.
The opposite mistake is just as expensive. Once a correlation comparison fails, it is easy to jump straight to a chain, phaser, reluctor, sensor, or controller. The failed comparison proves a relationship problem. It does not identify which part created it.
What Is Actually Failing
At the correlation layer, the failure is not automatically a failed CKP sensor or a failed CMP sensor. The nonconforming condition is the measured relationship between the crank reference event and the cam reference event under the tested operating state.
That relationship can be disturbed by signal-path faults, shifted or damaged targets, incorrect mechanical timing, variable-cam position errors, phaser behavior, learned or interpreted reference errors, or other application-specific causes. Correlation tells you where the evidence failed, not which cause is guilty.
The Big Diagnostic Trap
The primary trap is false proof by addition: CKP passes, CMP passes, therefore correlation passes. That conclusion adds evidence the tests never produced. Individual validity and relationship validity are different questions.
The secondary trap is false cause assignment: correlation fails, therefore the chain jumped or the sensor failed. A failed relationship comparison narrows the diagnostic layer, but component identity still requires isolation.
The System Chain You Have To Think Through
Think through the evidence in layers instead of treating the entire synchronization system as one test:
- Qualify the CKP signal at the required point, operating condition, and test setup.
- Qualify the CMP signal independently under the required point, operating condition, and test setup.
- Capture CKP and CMP in a synchronized same-condition record when the authorized correlation method requires direct event comparison.
- Compare the relationship against exact-vehicle authority, including the applicable engine state and commanded or achieved variable-cam position when relevant.
- If correlation fails, isolate the cause layer before condemning a sensor, circuit, target, reluctor, phaser, mechanical timing component, or controller.
Skipping a layer turns evidence into assumption. Each layer must earn the conclusion assigned to it.
Why This Gets Worse Under Real Conditions
Real operating conditions can move the relationship while leaving each individual signal looking clean. Heat, vibration, speed, oil pressure, phaser movement, chain dynamics, or intermittent target and circuit behavior can change phase or event timing without creating an obviously ugly waveform.
Separate captures can also fool you. A clean CKP capture from one event and a clean CMP capture from another event may prove individual validity, but they cannot prove the exact relationship that existed between the two signals during one synchronized event.
What This Failure Looks Like On The Vehicle
A correlation-layer problem can show up with symptoms that overlap several failure families:
- Crank/no-start or extended crank with individually believable CKP and CMP activity.
- Cam/crank correlation DTCs or synchronization status changes without a clearly failed individual sensor signal.
- Intermittent start, stall, or run-quality complaints that appear only at certain temperatures, speeds, or VVT states.
- A sync PID or learned offset that looks suspicious but does not, by itself, identify the physical cause.
- Mechanical-timing suspicion even though the available evidence has not yet separated mechanical phase from signal-path or control-state errors.
These symptoms justify correlation proof. They do not bypass it.
What A Good Tech Should Prove Instead
A strong diagnostic path proves the layers in order:
- Prove CKP validity using the authorized criteria for that signal and condition.
- Prove CMP validity independently using the authorized criteria for that signal and condition.
- Use synchronized same-event evidence when the correlation method depends on relative event timing or waveform phase.
- Use exact-vehicle correlation authority and account for the operating state, including variable-cam command and achieved position when applicable.
- After a failed correlation result, isolate signal path, target or reluctor position, phaser behavior, mechanical timing, and controller interpretation as separate cause questions.
The goal is not to collect clean waveforms. The goal is to prove exactly which diagnostic layer is valid, which layer failed, and what remains unproved.
The Field Rule
A symptom is not proof of a cause, and two valid inputs are not proof of a valid relationship between them.
Things Learned From This Asset To Stop Doing
- Stop treating a clean CKP waveform as proof that the CMP signal or correlation is valid.
- Stop treating a clean CMP waveform as proof that the CKP signal or correlation is valid.
- Stop combining separate non-synchronized captures and calling them correlation evidence.
- Stop treating a correlation DTC, sync PID, or learned offset as automatic proof of the failed component.
- Stop condemning mechanical timing, a sensor, a phaser, a reluctor, or a controller before the failed layer has been isolated.
Things Learned From This Asset To Start Doing
- Start labeling CKP validity, CMP validity, and CKP-to-CMP correlation as separate proof questions.
- Start preserving same-event and same-condition evidence whenever relative timing is the question.
- Start using exact-vehicle correlation authority instead of generic waveform appearance.
- Start accounting for commanded and achieved variable-cam position before judging a correlation relationship.
- Start treating a failed correlation result as the beginning of cause isolation, not the end of diagnosis.
Summary
A valid CKP signal proves CKP signal validity. A valid CMP signal proves CMP signal validity. Even when both are clean and believable, neither finding proves that the crank and cam reference events have the correct relationship.
Correlation must be proved with synchronized, same-condition evidence and the correct exact-vehicle authority. When variable cam timing is active, the applicable commanded and achieved state must be part of that interpretation.
If correlation fails, the relationship has failed the authorized comparison. The failed component is still an open question until the cause layer is isolated.
Stay in the loop
Get new drops in your inbox
New QTS builds, tools, and garage stories — straight to your inbox. No spam. Unsubscribe anytime.
Read the Privacy Policy.