Diagnosing a Crank Sensor Problem

[SmithSwede]

I'd be interested in any thoughts about how to track down what I think is a fault somewhere in the circuit of the crank position sensor on a 2013 V7 Stone small block with 92,000 miles.  I am not getting any spark or fuel at either cylinder, so I'm pretty sure this is something to do with the CPS.   I've had the exact same no spark/no fuel problem several times before, and it was always solved by replacing the CPS or fixing the air gap. 

It would be nice to know if (a) I am getting a signal from the CPS that is somehow not reaching the ECU, versus (b) no signal is being generated by the CPS in the first place.   

For example, is there some way I can use a multimeter to see if the CPS is indeed generating a signal like it should?  I know I can ohm out the sensor, but that's a pretty crude test, and both my new CPS units ohm out ok.  So, could I hook up a voltmeter to the connector of the CPS, crank the engine, and make a diagnosis based on the readings of the voltmeter?  (I don't have an oscilloscope)

Similarly, suppose I removed the CPS from its hole in the front of the crankcase, but left it connected electrically, and put the ignition switch into the RUN position.  Then I rapidly move a piece of metal close to the end of the sensor.  Would that action "trick" the engine into thinking it was getting a signal, and thus I would expect to see a spark at some point, and a fuel injector pulse at some point?  What about spinning some kind of gear in an electric drill close to the sensor to make the signal faster--would that trick the ECU into generating a spark and a fuel pulse?  Or would the engine not be "fooled" but such crude antics?

Is there some other connector I should be looking at besides the one the CPS plugs into?  I really don't want to condemn the ECU, and I know it is a major PITA to get to the ECU and its connector anyway. 

Another possibly relevant fact.  When I remove the CPS from its hole in front of the crankcase, and I look down through that hole at the gear ring, the gear ring is not centered in the hole as I would have expected.  The tunnel can only "see" about half of the gear ring.  Half that space, towards the front wheel, has a gear present.  The other half, towards the rear wheel, if just empty space.  Of course, I don't know if this configuration is perfectly normal. I just would have assumed the CPS would be positioned in such a way as to see 100% of the edge of the gear.  Is it possible that somehow the gear ring has moved forward in such a way that the CPS can no longer reliably sense it and generate a signal?   (My clutch is beginning to slip too, which makes me even more curious about a possibly moving gear ring).

Background info:

I'm getting no spark and no fuel. I've had this problem four times before, and it was always fixed by replacing the CPS, or minimizing the air gap between the end of the CPS and the gear wheel (I once tried to use an O-ring that was too fat and had too much sealant on the sensor, causing the air gap to be too big).

The last time this no spark/no fuel problem occurred was 60 miles ago.  I replaced that CPS with a brand new one, whereupon the bike started and ran perfectly for my 60 mile trip back home.  So I figured it was another failed CPS.   I let the bike sit for three weeks while I ordered yet another CPS to have on hand in case this problem occurred again.  After three weeks, I went to start the bike, and I'm right back to no spark, no fuel, no start.   

I then swapped out the 60 mile old CPS with my brand new CPS, and that made no difference.

Every fuse and connector I'm aware of that is possibly relevant has been treated repeatedly with De-Oxit.  The CPS connector looks good and snug.  The relays have been cleaned and swapped around.  I've tried wiggling the wiring, and the CPS, with no difference.  There is no crud on the end of either of the two CPS units.  I've cleaned all gasket sealer off the top of the hole in the crankcase for the CPS, and have tried fitting the CPS with the absolute minimum air gap, and also tried wiggling it up and down a bit to vary the gap.  I never get a spakr. 

The bike is under a porch, and it has been Texas hot for days, so there is no water or humidity present.

The battery is fully charged, and spins the starter motor vigorously.   Fuses are good. 

[fotoguzzi]

Just for kicks I would disconnect the battery for a good 5 minutes and see if it can clear something.

[beetle]

It's s Hall Effect Sensor. You should be able to test it with a meter set to AC volts. Connect across the two pins and spin the motor over. Output will be probably less than 1 Volt.

[Kiwi_Roy]

On my V7 I have a light to tell me when the CPS relay is picked up.
I did this partially because of your experience.
The first thing that has to happen, the relay has to liven up the fuel pump.
I doubt you would be able to simulate the pulse train from the sensor, too many pulses between each spark or shot of fuel.

I see a short pulse when the key turns On
Then the instant it starts to crank the light turns on again and stays on.
http://www.thisoldtractor.com/guzzi007/schematics/2013_V7_Series.gif  The red wire coming out of relay 28

[Chuck in Indiana]

It's s Hall Effect Sensor. You should be able to test it with a meter set to AC volts. Connect across the two pins and spin the motor over. Output will be probably less than 1 Volt.

I *really* hate to say anything, but I thought it was nothing more than a magnet and coil of wire.. (?) That wouldn't be a Hall effect sensor as far as I know. (practically nothing)

[Kiwi_Roy]

Yes Chuck is right, it's the wheel sensors are Hall Effect
You should be able to measure the resistance, around 680 Ohms

[beetle]

I *really* hate to say anything, but I thought it was nothing more than a magnet and coil of wire.. (?) That wouldn't be a Hall effect sensor as far as I know. (practically nothing)

Happy to be corrected. Whatever it is, it still produces an AC voltage across the output. My test is still valid. 

[Wayne Orwig]

It is a variable reluctance sensor.

Basically, it is a coil of wire, and a magnet.
If the resistance is correct, it is not opened or shorted.
If it is still magnetic, it will generate a signal when the ferrous timing wheel moves by it, if close enough.
And if the gap is right, without metal bits on it, done, it will work.

With the right meter and the motor spinning rapidly, it should show something on AC volts. It won't be much. A digital meter may be hard to read the fast pulses, and an analog meter may have too low of an ohms per volt value, to see it. And if you do see a voltage, how much is enough?

Have you actually measured the hole depth and compared it to the sensor length. That is, measured the gap? I have to admit, I never looked closely to see if the timing wheel is centered in the hole. I would have thought so, but................ ..  I have measured the gap on a couple of them, and recall turning over the motor slightly to get a tooth centered in the hole, left and right. It never occurred to me that it might be off front to rear, so it must not have been noticeable. Maybe your cam thrust plate, whatever it is called, is worn.

As for waving a metal bit in front to get it to trigger, that is unlikely. The pulse timing has to be very exact, so it can look for a missing pulse. That is what lets it know where TDC is.

[SmithSwede]

Yeah, this crank position sensor is not a Hall effects sensor.  I cut an old one apart with a Dremel and it is just a long piece of metal oriented longitudinally with a lot of fine wires wrapped around it.   It confused me for quite a while because this sensor has 3 wires (like a Hall effect), whereas most passive inductive sensors have just two wires.   The older Guzzi inductive sensors only had two wires.   So I erroneously though it was a Hall. 

Turns out the third wire on this sensor is attached to a metal sheath that runs the entire length of the cable. So I’m assuming this third wire is some kind of anti-static or RF thing that is grounded.   The other two wires are attached to either end of the coil around the metal bit.

[SmithSwede]

Kiwi Roy:

I should have made that clearer.   When I first turn the ignition switch from OFF to RUN, the fuel pump cycles normally for about two seconds and then stops.   So I think that portion of the relay circuit is working correctly.   

But after this 2 second priming thing is done, I think the ECU never receives the pulse signal from the crank phase sensor, even though the starter is turning over the engine vigorously.    So the ECU thinks no cranking or engine rotation is occurring, and thus never commands a spark or fuel injection pulse. 

I’m wishing there was some shade tree way to test this stuff.  Like in the Olden Times when we could just open up the points to at least see if you get a spark. 

[Wayne Orwig]

Another possibly relevant fact.  When I remove the CPS from its hole in front of the crankcase, and I look down through that hole at the gear ring, the gear ring is not centered in the hole as I would have expected.  The tunnel can only "see" about half of the gear ring.  Half that space, towards the front wheel, has a gear present.  The other half, towards the rear wheel, if just empty space.  Of course, I don't know if this configuration is perfectly normal. I just would have assumed the CPS would be positioned in such a way as to see 100% of the edge of the gear.  Is it possible that somehow the gear ring has moved forward in such a way that the CPS can no longer reliably sense it and generate a signal?   (My clutch is beginning to slip too, which makes me even more curious about a possibly moving gear ring).

Here is a thought. Take out the spark plugs. Put it in gear. Bump the motor forward a bit. Then bump it back a bit. The timing wheel will of course spin. But also see if the timing wheel moves forward and back any, like something is loose or worn.
 

[SmithSwede]

I don’t know if this is useful data, but here goes.

Original sensor, # 0, that came with the bike, lasted about 50,000 miles.  Then did the no spark/no fuel thing.   I replaced it with a brand new phase sensor #1, and bike ran great for 40,000 miles, then we’re back to no spark/no fuel.

So I replaced it with brand new sensor #2.  Bike ran great for about 1,800 miles.   Then I’m back to no spark/no fuel (which happened at 1:00 am while parked at DFW airport; not a happy camper). 

Replaced with brand new sensor # 3.  Bike started immediately, ran great for the 60 miles from airport to home.  Bike sat for 3 weeks.   Then would not start.  No spark/no fuel. 

Replaced with brand new sensor #4. This time the replacement trick didn’t work.   No spark/no fuel. 

I have a hard time believing that a brand new sensor, and another one with only a single 60 mile trip, and a third one with only 1,800 miles, are all defective.   Since this is not a Hall sensor, just a passive inductive, I don’t see how the vehicle’s electrical circuit could fry them.

All of them ohm out at 670 ohms across the coil; no discernible physical damage to the body or the wires. 

[RinkRat II]

         Almost  sounds as if the cabling from the ecu to the sensor has an intermittent disconnect or possible grounding issue. More time spent with your ohm meter is in your future. Could you remove the alternator cover and try to push the crank back and see if the ring not being centered is suspect. That's all I have.

  Paul B

[Wayne Orwig]

All of them ohm out at 670 ohms across the coil; no discernible physical damage to the body or the wires.

The problem is very unlikely to be the sensor.

What is the gap?

[John A]

Of course your sposed to pull the front cover and use a feeler guage but if you are not sure of your measurement down the hole,  stick a bit of clay on the end of the probe.  If you clean the probe well, the clay will stick to it and the residual oil on the phonic wheel will act as a mold release

[PJPR01]

Prescott...remind me if you have ever replaced the spark plug wires in this whole process?   Just for grins, have you tried a new set of plugs lately as well?

[Kiwi_Roy]

Of course your sposed to pull the front cover and use a feeler guage but if you are not sure of your measurement down the hole,  stick a bit of clay on the end of the probe.  If you clean the probe well, the clay will stick to it and the residual oil on the phonic wheel will act as a mold release

I do that with JBQuick or even JBWeld if I don't have the former, leave it to set up then remove it and measure the thickness, don't turn the engine of course while waiting for it to set.

The gap for the early ones used to be 0.6 - 1.2 mm (according to Efiman)
I couldn't find the gap in the Workshop Manual  http://www.thisoldtractor.com/mg_manuals/workshop_manual_v7-iii-abs_en.pdf  but I'm sure it's in there somewhere
On page 86/87 they go to great pains to show the layout of the cable.

[Kiwi_Roy]

Kiwi Roy:

I should have made that clearer.   When I first turn the ignition switch from OFF to RUN, the fuel pump cycles normally for about two seconds and then stops.   So I think that portion of the relay circuit is working correctly.   

But after this 2 second priming thing is done, I think the ECU never receives the pulse signal from the crank phase sensor, even though the starter is turning over the engine vigorously.    So the ECU thinks no cranking or engine rotation is occurring, and thus never commands a spark or fuel injection pulse. 

I’m wishing there was some shade tree way to test this stuff.  Like in the Olden Times when we could just open up the points to at least see if you get a spark.

Yes the first run of the pump is nothing to do with the sensor, it's just a simple timer in the ECU, the timer proves the relay is ok because it's able to run the pump.

When you crank the motor over the pulses from the sensor are seen by the ECU so it starts the pump again (same output) (see on the drawing how it also powers up the coils and injectors)
So if the pump doesn't start again chances are the ECU is not receiving pulses from the sensor.
Measure the sensors you have for resistance, if they measure Ohms chances are there is nothing wrong with them.

Perhaps there is a little corrosion in the sensor socket, it's a very small signal, needs a good contact I doubt any spray on would fix that

Too much gap.

The third wire is the shield, very common to have a shield around a low signal sensor

My LED under the seat tests it every time I start, of course I'm not looking at it and I hope I never have to, Basically it just monitors the common red wire

[Chuck in Indiana]

The gap should be .030"
Too much, and it won't work. Don't ask me how I know..

[pat80flh]

     Anything I know about crank sensors relates to cars. You might see .5 volt AC while cranking but has little to no value during diagnosis. An oscilloscope is the only real way to diagnose one, and few carmakers believe field technicians are smart enough to use one.  In the automotive field, a quick test for a no start is to observe the tach, or observe RPM reading on a scan tool.

     Most systems need to see 200 RPMs before they will "turn on " the ignition system. I have seen worn crank thrust bearings which let the crank movde enough to lose the CPS signal, especially on a manual transmission with clutch disengaged. Maybe rough check crank end play , or observe your tone  wheel while moving crank back and forth through timing hole.  Most engines I've ever worked on crank end play should be around .010".

   I agree that after 3 new sensors the problem probably lies elsewhere.

[malik]

Bloody frustrating. Can't help much with the testing/measurement bit - my skills with a meter are rudimentary & I'm never really sure what im looking at. I have, in the past, identified a bad oil pressure sensor (tho there were intermittent hints from the dash) and a dodgy TPS (on the 2TB).  I do know I wasn't happy with the stock connection between spark plug & cap on the 2014 Special & and changed over the caps & plugs early on. I have also found corrosion at the plug end of the HT wire that affected running. But nothing to the extent of your symptoms.

Your clutch slip? It may be a separate problem. I've now had 3 clutches slipping, all at around 100,000km, due to oil on the clutch - seemingly gear oil travelling up the clutch push rod and on to the cup & clutch. Other V7s & Brevas have reported similar symptoms at that mileage. You are well into that zone. First test is to remove the rubber bung covering the inspection hole on the RHS of the bell housing, stick your finger in there (engine not going) and see if you find it damp - traces of oil mixed with friction pad powder. If there is oil in the bell housing, and there shouldn't be, it's likely the clutch slip is due to oil on the clutch pads and you've a leak in there. Possibilities are the clutch push rod seal, the seal on the clutch shaft, or the main engine seal behind the flywheel. So far, the ones I worked on seem to be the clutch push rod seal. Anyway, the fix is a new clutch & new seals, which means engine out, I'm afraid.

Good luck,

Mal

[Kiwi_Roy]

Reading through the posts again Wayne asked if the phonic wheel could be moving sideways easy to checkisn't it fixed to the camshaft? It would be fairly obvious looking in the hole if it were off centre.

Do you still have the shim/s under the sensor?
     I don't think shims come with a replacement, without a shim I think it would be too close to the wheel, some owners have used an "O" ring in an attempt to make it oil tight moving it too far away. I have tried to measure the depth with a caliper, doesn't work for me, something setting up on the tip is easier.
 

[Idontwantapickle]

I just read through all the posts.
Have you checked the continuity of the entire circuit to the sensor? Looking at the diagram KiwiRoy posted it would be pins 20 and 29 at the ECU. Find those pins and check for the 680 ohms there with the sensor connected.
Just my .02
Hunter

[SmithSwede]

Thanks Hunter.   I’ll add that to my list.   

Only problem is that it is a royal Pain to gain access to the ECU and its connectors.