Lack of Camshaft position sensor

[tolsen]

This problem has given me sleepless nights for years. The problem applies to both the petrol and diesel engines made by Mercedes Benz and used in the 450, 451 and 454.

How is cam position sensed by the ECU?
There is no sensor anywhere along the camshaft, cam drive or valve train. The injection is purely electrical and the ECU must have some means of sensing when each piston is in its compression stroke so diesel fuel can be injected at the right time. Crank shaft speed sensor alone is not sufficient for ECU to determine which of the two TDCs is start of firing cycle.

One possible answer is that, when starting engine, the ECU at first attempts firing the first cylinder at odd revolutions, and then tries firing at even revolutions if first attempt was unsuccessful.

Any thoughts and feedback would be appreciated.

[Thrumbleux]

Re: Lack of Camshaft position sensor

TK, I can answer this for you.
It is all done from the crank sensor. MB made it sensitive enough to discern the compression stroke from the exhaust stroke. Info gleaned from an official MB online document detailing the development of the three cylinder CI engine destined for the forfour.

Found it, here >> http://www.docstoc.com/docs/77633987...-Diesel-Engine

[tolsen]

Re: Lack of Camshaft position sensor

Many thanks Thrumbleux, that document explains how it is done:

5.2.2 Synchronization without a Camshaft Sensor
A special feature of the three-cylinder engine is the possibility to synchronize fuel injection and position of the crankshaft without using a camshaft sensor. As opposed to other engines, here the synchronization takes place here with the help of a “virtually” generated camshaft signal. This is detected using the irregularity of the crankshaft rotation as it develops through the compression processes in the cylinders, and it is made possible by a high-resolution rotation signal acquisition.

I assume there will be a noticeable angular de-acceleration at compression stroke which is detected by the ECU based on signal from crank shaft speed sensor.

[smartsparky]

Re: Lack of Camshaft position sensor

Have a look at the flywheel. You will see the extra large gap that is picked up by the TDC sensor. If the clutch assembly is fitted 120 degrees out the firing sequence will be 120 degrees out also. A locating pin should stop this from happening but I've seen garages that have still messed this up!

[tolsen]

Re: Lack of Camshaft position sensor

Have a look at the flywheel. You will see the extra large gap that is picked up by the TDC sensor. If the clutch assembly is fitted 120 degrees out the firing sequence will be 120 degrees out also. A locating pin should stop this from happening but I've seen garages that have still messed this up!

I am aware of that but the question has to do with how they can make engine run without having cam shaft sensor.

[tolsen]

Re: Lack of Camshaft position sensor

So the way synchronization is achieved is only possible on a single or three cylinder engine. On a two or four cylinder engine it will be impossible to detect any deceleration due to symmetry, i.e. one cylinder is on firing stroke whilst the other is on compression stroke etc.

[tolsen]

Re: Lack of Camshaft position sensor

Thanks again Thrumbleux, no more sleepless nights from now on. Think I might even go out and buy a bottle of Lidl Queen Mary Scotch whisky to celebrate.

[Thrumbleux]

Re: Lack of Camshaft position sensor

So the way synchronization is achieved is only possible on a single or three cylinder engine. On a two or four cylinder engine it will be impossible to detect any deceleration due to symmetry, i.e. one cylinder is on firing stroke whilst the other is on compression stroke etc.

Yep, that's my take also - not applicable 180 crank phasing. I think this technology came from MB's first V6 Diesel where 120 phasing would be the norm. Damned clever stuff!

Thanks again Thrumbleux, no more sleepless nights from now on. Think I might even go out and buy a bottle of Lidl Queen Mary Scotch whisky to celebrate.

You're welcome. Take it as a small thankyou for your help with my wheel bearing problem.

[smartsparky]

Re: Lack of Camshaft position sensor

I am aware of that but the question has to do with how they can make engine run without having cam shaft sensor.

By taking the reading from the crank sensor. Say that the long groove just to happens to coincide with cylinder 1 TDC, the ECU will be programmed to fire at that point. It will also know that cylinder 2 is compressing and cylinder 3 is on the exhaust stroke. It's all to do with the clever electronics.

[tolsen]

Re: Lack of Camshaft position sensor

By taking the reading from the crank sensor. Say that the long groove just to happens to coincide with cylinder 1 TDC, the ECU will be programmed to fire at that point. It will also know that cylinder 2 is compressing and cylinder 3 is on the exhaust stroke. It's all to do with the clever electronics.

It is not that simple. No 1 cylinder is at TDC twice for each rotation of cam shaft. Which of the two TDC's will ECU use for for firing cycle?

[smartsparky]

Re: Lack of Camshaft position sensor

Could it not just use it as a calibration check? After all, we already know that it doesn't matter if the cam shaft is installed 180degrees one way or the other. The camshaft is going to follow what the crankshaft is doing, albeit, at twice the rotational speed.

Upon starting, I would suspect that the crank sensor needs to 'see' TDC a couple of times before knowing when to start the firing sequence. Once the engine has successfully started, the ECU would just monitor that crank rotation is synchronised with the firing sequence using the feedback from the crank sensor. This may explain why the time between turning the key and the engine actually firing, varies.

Just my thoughts!

[tolsen]

Re: Lack of Camshaft position sensor

Yep, that's my take also - not applicable 180 crank phasing. I think this technology came from MB's first V6 Diesel where 120 phasing would be the norm. Damned clever stuff!

I think the way cam shaft position is sensed is only possible on single and three cylinder engines. A V6 engine with 120 degrees phasing will have 2 cylinders at TDC same time just like on a four cylinder engine so not possible to detect which cylinder is at which cycle.
Synchronization need only be checked or confirmed when engine is being cranked prior to firing. ECU does this by detecting changes in angular velocity and angular acceleration due to high rotational resistance at compression stroke. On a single cylinder engine, this resistance at compression stroke is quite pronounced - anyone owning a hard to start lawnmower engine will know. The resistance will be equally noticeable on the 3 cylinder Smart engine (as phased 120 degrees) and ECU will therefore "feel" when no 1 cylinder is at its compression stroke.

Here is an interesting thought:
What happens when compression on an engine is partly lost (remember member Foler's Cdi engine with intermittent start problem)?
Foler's engine had low compression on no 1 cylinder. The lower the compression the less chance of ECU detecting its compression stroke. Engine may fail to synchronize or synchronize on wrong cylinder.

Does this make sense Thrumbleux?

[Thrumbleux]

Re: Lack of Camshaft position sensor

Just read your post (elsewhere) re Foler's problem and was astounded at you drawing the conclusion you did from yesterdays bit of info. Good work! There is potentially a real inbuilt problem there re starting when #1 cylinder is low on compression. Presumably MB would respond with something akin to 'you shouldn't be trying to start such an engine but rebuilding it'!

Re V6. Admittedly I didn't think that right through. The angle of the V, the actual crank phasing, etc would need proper scrutiny before arriving at the correct answer. A closer comparison of the V6 with the OM639 (I'm near certain they were co-developed, one being a mirrored half of the other) would help - as would scouring the parts lists for a cam position semsor for the V6. If it has a 60deg bank angle and 120deg crank phasing, then almost certainly it must have a cam position semsor. Perhaps the cost saving on the triple was valued as destined for a lower cost car generating a smaller profit margn.