TDCI Engine Cam & Crank sensor scope trace sanity check.

[ajlg]

Hello,

I have a Mk6 chassis cab RWD 2.4 TDCI 135 which is a non runner. Apparently it stopped one day, got towed in and has not run since. The previous owner had a garage replace two injectors under advice those were at fault and then cut his losses and bailed. It was well maintained in the past by the NHS with many injectors and new bits (it was / is an ambulance).

I have taken it on and set to work on it. I have Fordiag going via a modified interface (fantastic program) and it is repeatedly reporting a Cam sensor fault. I got a scope going on the Cam and Crank sensor in the correct manner (not shorting the crank sensor to ground etc) and got some steady and repeatable traces, though the Cam seems to have a very short duty cycle. Since other examples I have seen have seen have had a 50% duty on the cam I figured the PCM was correct and bought a (fairly budget) sensor, however this gives identical traces and the P0340 code again.

Is there anything obviously wrong with these traces, should the flywheel give 57 or so pulses on the crank sensor?
On what pulse of the crank sensor should the cam sensor trigger / angle?

Is it possible that damage on the Cam shaft would affect the Cam sensor output like this?

Engine turns over on the starter fine, fuel rail builds up pressure fine etc but no sign of firing...


Complete cycle, showing Cam pulses on top trace and Crank waveform on bottom trace.

TEK00000.PNG

Zoom in on phase delay of crank index to cam shaft sensor triggering.

TEK00003.PNG

Any advice is greatly appreciated!

Many Thanks,

- Alex

[paul2012]

P0340 Camshaft Position Sensor "A" Circuit

I suggest checking wiring

[ajlg]

Sensor gets a solid 12v and gnd, and with it unplugged the 5v pullup from the PCM control is present and correct. Sensor pulls down on signal line fine as per scope trace, so I am fairly sure wiring is ok...

[wojciech]

Is there anything obviously wrong with these traces, should the flywheel give 57 or so pulses on the crank sensor?

Crank sensor gives 60 - 2 = 58 pulses. This minus two pulses it is space without two teeth, localized 90 degrees before highest piston position on 3rd cylinder.
It looks like both sensors are ok, so pulses synchronization maybe is wrong. Check timing...

[paul2012]

Sensor gets a solid 12v and gnd, and with it unplugged the 5v pullup from the PCM control is present and correct. Sensor pulls down on signal line fine as per scope trace, so I am fairly sure wiring is ok...

You should get 2 pulses from crankshaft sensor to every 1 pulse from camshaft sensor.
If not then check timing chain. May have jumped a tooth.

[ajlg]

Crank sensor gives 60 - 2 = 58 pulses. This minus two pulses it is space without two teeth, localized 90 degrees before highest piston position on 3rd cylinder. It looks like both sensors are ok, but pulses synchronization is wrong. Check timing...

I am fairly certain, after finding a picture of a flywheel and counting teeth, that the crank does seem spot on, but the cam sensor you say also looks good, except for timing?

Should it not be high for one crank rotation and low for another, or am I reading the internet wrong? What is the best way to check the timing on these, chain cover off and look for markings?

Thanks for your assistance, its my first time working on a common rail diesel, but I am used to petrol beamers and their forest of sensors...

- Alex

[ajlg]

You should get 2 pulses from crankshaft sensor to every 1 pulse from camshaft sensor.
If not then check timing chain. May have jumped a tooth.

Are you sure about this, as this goes against everything I know?
I thought I was looking for (and seem to get close to) 2 complete 58 cycle crank waveforms to 1 pulse from cam sensor. My pulse just seems short. and possibly in the wrong place.

- Alex

[wojciech]

If both sensors give correct signal, and still you have P0340 error - first thought it is wrong timing. Sorry - looking for your diagrams, I can't say is it ok or wrong.

[wojciech]

My pulse just seems short. and possibly in the wrong place.

I think - somebody wise, looking for your diagram should say is it good or wrong timing. But it's not me... viewtopic.php?f=54&t=26501
First - be sure about diagram, not till than open timing cover.

[loot]

Looks like you have 4 options.

1. Scope another mk6 tdci same model to compare with your scope.
2. Scan it with first ids which will be better to tell you if it is correct.
3. Whip the cover off and time it up/confirm it is timed.

4. Wait for someone to come on with exactly the info you need.

Personally I'm not upto date with your findings
But if I'm reading it right 200ms doesn't look that far out considering the size of the sensors/notches and the speed of rotation, I'd expect too done a tolerance and you'd have problems all the time wouldn't you? Just my thoughts tho, maybe 100ms, maybe that is massive?

I do think the quickest thing would be to visually inspect the timing and meter the sensor to pcm cabling etc.

[knobby1]

Looks like you have 4 options.

1. Scope another mk6 tdci same model to compare with your scope.
2. Scan it with first ids which will be better to tell you if it is correct.
3. Whip the cover off and time it up/confirm it is timed.

4. Wait for someone to come on with exactly the info you need.

Personally I'm not upto date with your findings
But if I'm reading it right 200ms doesn't look that far out considering the size of the sensors/notches and the speed of rotation, I'd expect too done a tolerance and you'd have problems all the time wouldn't you? Just my thoughts tho, maybe 100ms, maybe that is massive?

I do think the quickest thing would be to visually inspect the timing and meter the sensor to pcm cabling etc.

I'm sorry Loot.....you're talking sense again...we can't have that ....you better have a lie down and come back to your normal "nonsensical" self..

Lord Knobrot

[v8dave]

I know some engines, rover K series for example, have a special lobe for the cam sensor which is normally shaped to give the 50% duty cycle.
But I thought the cam position sensor on the transit looked for one of the inlet cam lobes.
If so then the pulse length would be 45 to 60 degrees of cam rotation or 90 to 120 degrees of crank rotation. The precise value would depend on cam ramp rates and sensor range.

Personally, I think the cam trace looks Ok.

If the cam and crank on a mk7 are not timed correctly the ecu will not inject. It is likely the mk6 TDCi operated in a similar way.

If you can access a running tdci you could check the angle between the centre of the crank wide pulse and the centre of the cam pulse and compare it to yours.

Otherwise I'd get the front cover off and check the chain guides and tensioner for wear.
If they're not worn then time it.
If they're worn, replace them and time it.

[ajlg]

Thanks for the replies guys, its really appreciated!

It is starting to sound like I need to get covers off and check the timing on this thing, but I was still hoping someone might be able to shed light on what the correct duty cycle is of the Cam sensor (that is, what the thing it looks at looks like - is is a 180 degree semi-circle or just a small index tooth?)

Attached are some more scope shots for completeness, the first shows the complete width of one cycle of the crank as 277ms (so 1/0.277 * 60 = 216 RPM on the starter) and the second shows the width of the CAM signal pulse (which occurs every other cycle of the crank) The width of this is only 56ms, giving a overall duty of around 10%...

TEK00002.PNG

TEK00005.PNG

Thanks again,

- Alex

[ajlg]

I know some engines, rover K series for example, have a special lobe for the cam sensor which is normally shaped to give the 50% duty cycle.
But I thought the cam position sensor on the transit looked for one of the inlet cam lobes.
If so then the pulse length would be 45 to 60 degrees of cam rotation or 90 to 120 degrees of crank rotation. The precise value would depend on cam ramp rates and sensor range.

Personally, I think the cam trace looks Ok.

If the cam and crank on a mk7 are not timed correctly the ecu will not inject. It is likely the mk6 TDCi operated in a similar way.

If you can access a running tdci you could check the angle between the centre of the crank wide pulse and the centre of the cam pulse and compare it to yours.

Otherwise I'd get the front cover off and check the chain guides and tensioner for wear.
If they're not worn then time it.
If they're worn, replace them and time it.

If this is the case, then I would agree that it is starting to look a lot like rubber glove time.

[ajlg]

Just noticed something that is probably relevant on my scope trace, the first cam pulse goes negative 112ms after the rising edge of the Crank index pulse. The second cam pulse in the trace goes negative 108ms after the rising edge. So the Cam pulse triggered 4ms earlier the next time around (half a crank tooth earlier)

This would point to a sloppy chain too then I guess?