15

The only thing I can come up with is timing belt / chain flex which can cause a very small amount of timing difference. I guess they could detect if the belt was put on wrong, skipped a tooth, stretched, or broke. Everything but stretched you would notice the problem and timing belts don't stretch much last I heard.

Other than those failure states, you should be able to program everything using just a cam position sensor. So why do cars have a crank sensor in addition to a cam sensor? This seems redundant and unnecessary. Or is its job just to tell you something major is now wrong with your car after the belt snaps and valve mashing commences?

Here is an example of a Honda's parts list as you can see it has both sensors.

  • 4
    Adding details to your question is a good thing, but adding comments to your question in response to each answer is making it a mess. That is why there is a comment section for each answer. Adding your comments to the question will likely make it long enough people won't both to read it. This could easily be considered a commentary rather than a specific Q&A. – CharlieRB Nov 22 '16 at 12:59
  • 3
    Please note that we're a question and answer site, not a discussion forum. If you're looking to have a discussion with people about their answers, Motor Vehicle Maintenance & Repair Chat is the place. – David Richerby Nov 22 '16 at 13:23
  • @CharlieRB good point I will move those to the comments section! – Cc Dd Nov 22 '16 at 19:22
12
  1. The biggest reason is the need for misfire detection.

The best way to tell if the crank shaft is speeding up and slowing down more than it is supposed to is to directly measure its speed. This is how misfires are detected. It is because the timing chain or timing belt always has a little give and this can obscure the movement of the crankshaft from the camshaft position sensor. Further because the crank shaft is spinning at twice the speed of the camshaft the signal is of a higher resolution.

Until Honda went to coil on plug. The ignition and fuel injections systems were run from sensors inside the distributor. Since the distributor is driven by the camshaft it is effectively a camshaft position sensor.

  1. A secondary reason is legacy.

When the switch over form distributor to distributorless occurred the crank shaft position sensor was used as the pick up for timing. Manufacturers that dumped distributors in the late 80s didn't always pick up a camshaft position sensor right away. It was when they went to sequential fuel injection that a camshaft position sensor was actually required.

  1. A third reason is variable valve timing.

Any variable valve timing system that uses a cam phasor requires both sensors. By using a PWM solenoid the position of the phasor is infinitely variable through its travel. To keep track of where the phasor is in relation to the crank shaft both sensors are needed.

  • updated my question. Excellent answers! – Cc Dd Nov 22 '16 at 6:01
  • 2
    Please stop updating your question. – Selali Adobor Nov 22 '16 at 13:13
12

Both are needed to know which stroke each piston is in

Variable valve timing or not, on four-stroke engines the camshaft spins at half the speed of the crankshaft.

Four-stroke animation

This means that for any given crankshaft angle there are two possible camshaft positions, which means that knowing the crank angle alone is not enough to eliminate aliasing, therefore not enough to know when to ignite the spark plug/fire the injector.

It is therefore mandatory in distributor-less setups to have input from both sensors to determine which stroke each piston is in.

  • Updated my question and oops your right I need to re update it. – Cc Dd Nov 22 '16 at 6:04
  • While true for injected system, many motorcycles ran wasted spark ignition systems running off a single pickup on the crank. The ignition is triggered when the crank is in the appropriate position, but whether the cams are is irrelevant. – Kickstart Nov 22 '16 at 11:04
  • @Kickstart you are right. It's not easy to accommodate all possible variants in one answer – Zaid Nov 22 '16 at 14:39
  • 1
    I think the point of the question though is, why have a crank sensor? Use just the cam position one and extrapolate the crank position from that. – Steve Matthews Nov 23 '16 at 10:12
11

These two sensors are not always used for the same purpose. Obviously, reading each cam and crank shaft position gives more granular information to the computer for tuning of the engine.

According to Standard (manufacture of automotive engine management components);

The Camshaft sensor determines which cylinder is firing to establish injector synchronization and coil firing sequence in DIS systems. Crankshaft sensors set ignition timing, supply the RPM signal, and determine engine speed.

  • updated my question Good link and info! – Cc Dd Nov 22 '16 at 6:00
  • 1
    Don't update your question with comments on answerers. Put it in comments underneath their post. – anonymous2 Nov 22 '16 at 13:46
5

I know from working on Mazdas that several reasons are:

  • the signal from one can be used to verify the signal from the other so an engine code can be thrown if the belt slips a tooth. For example, 96-97 BP blocks have a 4 tooth crank sensor with a magnetic pickup- it just provides the ECU a verification that everything is aligned correctly but can't be used to run the engine on its own.
  • a car manufacturer can switch from crank to cam or vice versa over the years and it's often possible to combine heads/blocks from different years, so you can end up with a car that has both sensors. For example, putting the head from cam angle sensor engine on a crank angle sensor block and then hooking up whichever sensor the car's ECU uses
  • Sensors can often be retrofitted on blocks and heads that don't normally have them. This comes down to expediency/cost/laziness on the part of the manufacturer. For example, a 99 head doesn't normally have a cam angle sensor, it's just a block-off plate from the factory. But you can remove the plate and put in the old sensor and it works fine, allowing you to run a 99 head with an ECU that likes the hall effect sensor.
  • it could be used to provide feedback from a variable valve timing system- VTEC is a poor example because traditional VTEC is just a simple lobe-switching system with a single RPM switchover point, but a system that works around advancing and retarding the cams would definitely need a cam angle sensor that provides feedback to how far advanced or retarded the cams are.
  • updated my question good set of reasons. – Cc Dd Nov 22 '16 at 6:06
3

The clue is probably in the manufacturer you picked. Honda is famous for V-TEC which is a method whereby the camshaft advance profile can be changed depending on engine RPM. I assume this would mean a crank and cam position sensor would both be required?

  • edited my question with an answer to your answer. – Cc Dd Nov 21 '16 at 12:12
  • 1
    Bear in mind it's not just Honda that use variable valve timing, Alfa Romeo use cam "variators", MG Rover use VVC and Porsche have their own system (can't remember the name at the mo). – Steve Matthews Nov 21 '16 at 12:31
  • 1
    Some manufacturers rotate the camshaft relative to its drive gear to achieve variable valve timing. Honda doesn't do this: they include an extra cam and a system to switch between cams. – Hobbes Nov 21 '16 at 13:32
  • @SteveMatthews but a specific point is you can calculate the crank position using the cam angle so vtech will not be a reason why. – Cc Dd Nov 22 '16 at 4:24
  • 1
    @CcDd: The original VTEC system didn't use camshaft rotation, the more recent i-VTEC system does have camshaft rotation in addition to lobe switching. – Hobbes Nov 22 '16 at 12:07
2

The sensors feed the computer with which numerous other sensor inputs control cylinder firing and fuel to air mixes. A loss of signal or an incorrect one from a cam/crank sensor causes immediate engine shutdown - hopefully before damage is done. There are two types of engines I know of. 1- Interference type: If timing belt breaks, valves may hit pistons and do considerable damage. 2- Non-interference type: Belt breakage will likely do no harm. Replacing the belt on the recommended schedule is a must do situation unless you like to walk.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.