DIAGNOSTIC THEORY
Back to Explainers

The cam signal: how the computer knows which cylinder is which

The crank sensor tells the computer how fast the engine is turning. The cam sensor tells it which cylinder is about to fire. Learn what that signal carries, how the three sensor designs differ, and why a scan-tool reading can never prove the wire is good.

The cam sensor's one job: cylinder identification

The crank sensor tells the computer where the crankshaft is within one revolution and how fast it is turning. That is not enough to run a modern engine. A four-stroke cycle takes two full crank revolutions to complete - intake, compression, power, exhaust - which means the crank passes the same position twice per cycle. At that position, cylinder 1 can be approaching compression top dead center or exhaust top dead center, and the crank signal cannot tell the difference.

The camshaft turns at half crank speed - one cam revolution per two crank revolutions - so a mark on the cam happens only once per cycle. The camshaft position (CMP) signal gives the computer that once-per-cycle event: the sync mark that says which revolution you are on, and therefore which cylinder is approaching top dead center right now. That one piece of information is what makes sequential fuel injection, coil-on-plug firing order, and variable valve timing possible.

Why crank position alone is not enough

An engine can run on crank position alone if it does not care which cylinder is which. Batch-fire injection and waste-spark ignition do exactly that - fuel and spark delivered on a crank-only schedule, accepting that half the events land on the wrong stroke. Sequential systems refuse that compromise: each injector fires for its own cylinder's intake event, each coil fires for its own cylinder's compression event. That scheduling demands a cylinder-ID reference the crank cannot provide.

When the computer loses the cam signal, most systems fall back - batch-fire fuel, waste-spark or fixed timing, VVT disabled - and many will still run, poorly. That fallback is why a bad CMP signal can produce a no-start on one vehicle and a runs-rough on another. The signal's job is identification, not motion; the engine can turn all day without ever telling the computer which cylinder it is looking at.

Three designs, three tests: Hall, variable reluctance, magnetoresistive

Cam sensors come in three electrical designs, and the test you run depends entirely on which one you are facing. Identify the design from service information before touching a lead - the wrong test on the wrong design produces readings that mean nothing.

The practical rule: Hall and magnetoresistive signals are scoped as digital (levels and duty against the vehicle's requirements); variable-reluctance signals are scoped as AC (amplitude and frequency, checked at a known speed). Judging a VR signal's amplitude at cranking speed against a running-speed requirement is a misread, not a failure.

The scan-PID clue is not proof

A cam PID on the scan tool shows what the module thinks about the cam signal - an interpreted value, sometimes inferred from sync logic, sometimes a substituted value when the real one is missing. It does not show the wire. A PID reading "cam present" proves only that the module is reporting cam present; it cannot prove the signal wire between the sensor and the module is intact, and it cannot prove the module's input circuit actually received the signal.

This is the clue-versus-proof discipline: the PID is a clue that opens the question. Proof is a capture - the actual signal, on a scope, at the sensor's test point and again at the controller's input, checked against the vehicle's requirements. If the PID says good and the wire is broken, the PID is wrong. The wire always wins the argument.

The related explainer on evidence discipline covers why a later normal reading never erases a reproduced failure (A later normal reading never erases a reproduced failure), and the crank-signal explainer covers the same clue discipline on the crank side (Crank/RPM Signal as a Master Enable Clue).

The signal's path: source, wire, controller input

The cam signal travels three legs, and each leg answers exactly one question:

No leg's answer settles another leg's question. A perfect capture at the source proves nothing about the wire. A good capture at the input proves the whole path - but a failed capture at the input says nothing about where it broke until the source is captured separately. Work the legs in order: source first, then input, and let the difference between them point at the wire.

What the cam-signal check is not

How SpeedNeeds uses it

The cam signal is proved leg by leg, in the problem condition, against the vehicle's own requirements. The design comes first - Hall, VR, or magnetoresistive, identified from service information - because the design decides what the capture must show. Then the source signal is captured and checked (pattern, duty, levels for digital; amplitude and frequency for VR), then the same signal is captured at the controller input and the two are compared. The scan PID starts the question; captures answer it.

Where the evidence cannot be captured honestly - no access to the test point, a VR signal captured below its usable speed, a capture taken in the wrong condition - the test stops and says so instead of guessing. A leg that cannot be proved stays open; it is never marked good by default.

Caution and edge cases

Closing clarity

The cam signal is the computer's "which cylinder" answer - one sync mark per cycle that lets sequential fuel, coil order, and VVT find their place. Prove it the way it travels: design identified, source captured, input captured, each leg judged on its own evidence. And never let a scan-tool number stand in for a capture - the PID is the clue, the wire is the truth.

Copyright 2026 SpeedNeeds.net. All rights reserved.