Testing sensor on R1

[Hooli]

My only comment to that is our suzukis have cam and crank sensors without any of that vtec nonsense. I think it's so they only fire one pair of plugs at a time.

[bass gt3]

Mr OCD said:

If you had told me that over the weekend I would have tested it ... ... it was a proper tt to get at as well despite only being held in with one bolt

Edited by Mr OCD on Monday 28th July 16:00

It's not important. What matters is it's running 100% now. But out of interest, where is the sensor located? In the cylinder head or bottom end? And if the bottom end, by the ends of the crankshafty or by the bottom tming chain gear?

[Mr2Mike]

bass gt3 said:

Indeed.
There's no condition at start up that would require a dedicated sensor for that initial 3-4 seconds and then switch timing toaseperate sensor. Everything the ECU needs can be derived from either a cam OR crank sensor.

That's not quite correct. The cam sensor only tells the ECU which phase cylinder #1 is when it's approaching TDC, usualy this sensor provides only a single pulse for every two crank revolutions so the accuracy would be extremely poor if the ignition timing was run from this. The crank sensor provides all the high resolution timing requirements i.e. ignition and end of injection events for sequential injection.

A crank sensor by itself can be used for for a wasted spark ignition and batch fired injection, but if you want to run sequential injection and COP (coil on plug) ignition then you need the extra information that the cam sensor gives you. However, in theory you only need this information at start up, once the ECU knows the cam/crank phase relationship it can track this purely through the crank sensor. I'm not saying this is how the R1 does it, but it's certainly possible and there are definitely some cars that do this.

[bass gt3]

Mr2Mike said:

bass gt3 said:

Indeed.
There's no condition at start up that would require a dedicated sensor for that initial 3-4 seconds and then switch timing toaseperate sensor. Everything the ECU needs can be derived from either a cam OR crank sensor.

Mr2Mike said:

A crank sensor by itself can be used for for a wasted spark ignition and batch fired injection, but if you want to run sequential injection and COP (coil on plug) ignition then you need the extra information that the cam sensor gives you. However, in theory you only need this information at start up, once the ECU knows the cam/crank phase relationship it can track this purely through the crank sensor. I'm not saying this is how the R1 does it, but it's certainly possible and there are definitely some cars that do this.

Not strictly true. If you're going to jump in and start discussing ECU functionality, please be accurate. COP doesn't require a cam sensor per se, as it'sperfectly possible to use output drivers from an ECU to fire COP's rather than the typical multi way coil blocks often seen in cars in a wasted spark mode. COP is a device, not a strategy
As for only needing the crank signal at start, again, not so. If you want to talk about cars, then you'll understand that when a VTEC system changes the cam timing,the ECU needs to know the crank position andthe relative camposition to provide mapping as the cam timing changes.
However, where both a Crank and Cam sensor exist in a non VTEC motor, it's often done for running sequential, bu tthe crank signal is retained as a datum. Disconnect it and the motor won't run despite ostensibly being used to sync the camsignal and tell the ecu where the engine is at. Thats why EFI bikes take 3 or 4 seconds to fire as the ECU works out the synchronisation and then swings into action

Mr2Mike said:

That's not quite correct. The cam sensor only tells the ECU which phase cylinder #1 is when it's approaching TDC, usualy this sensor provides only a single pulse for every two crank revolutions so the accuracy would be extremely poor if the ignition timing was run from this. The crank sensor provides all the high resolution timing requirements i.e. ignition and end of injection events for sequential injection.
.

Nope, not even close.
By timing off of the cam signal, as long as it isn't VTEC,, you have a constant against the crank. So there's no need for a crank sensor at all in fact. Go look how Ducati run. They use a 48-2 pickup off the bottom cam belt wheel.Hence your arguement of the ignition timing being inaccurate is in itself, inaccurate. And consider that acam wheel turns at a ratio of 1:2 to the crank. So 48-2 on the camequates to 94-4 on the crank, which initself far exceeds the normal 60-2 configuration. And consider that ECU's use the X minus Y as the count the teeth, so they know/predict when the next timing event is coming, even as the crank is accelerationg. Far superior to the old single pick up systems of old.
As for end of injection and ignition timing how does the crank do this?? Surely you're getting confused?? The injection timing when run as a sequential system is taken from the CAM signal as it's relative to the cam cycle, whereas the crank will read the same position for power and exhaust TDC's hence you now have two possible crank angles, 360 degrees apart at which to end injection. How would the ECU derive this from the cranK?????. You CANNOT run sequential without a cam sensor, but you can run batch or sequential without a crank sensor.

[Mr OCD]

bass gt3 said:

It's not important. What matters is it's running 100% now. But out of interest, where is the sensor located? In the cylinder head or bottom end? And if the bottom end, by the ends of the crankshafty or by the bottom tming chain gear?

It's on top of the cam cover, right in the middle ... impossible to reach without dropping the radiator ... which of course means removing the síde fairings too...

It runs like a dream now... it seems keener on the throttle from low down as well - ironic really as I changed the gearing to improve low end which now feels rather instant!

[Mr OCD]

Rather annoyed ... The code is back ... Flashed code 11 when trying to start - turned off and on and started on button.

Code is logged in diagnostics... Bike feels fine...

What next? Surely can't be another duff sensor?!

[Hooli]

Dirty connector, or maybe the 'stickyoutbit' (technical term) on the cam has broken off so there is nothing for the sensor to read.

/Optimist

[bass gt3]

1. Clean the mounting surface where the sensor is.
2. Make sure that the sensor cable and especially the connection between the sensor and the loom is away from high current / high voltage cabling like HT cables or the coils. At the connection there's no shielding so can be a point of interference.
3. Make sure that the sensor cable is only grounded at one end, usually at the ECU

Sounds like it picked up a bit of interference while cranking. Not too much to worry about but clear the log and keep an eye on it.
Number 2 is the most likely, especially if the sensor is in the middle of the cam cover close to spark plugs. Just check where the connection is and make sure its away from any source of electronic interference

Edited by bass gt3 on Friday 8th August 19:34

[Yazza54]

Mr OCD said:

Rather annoyed ... The code is back ... Flashed code 11 when trying to start - turned off and on and started on button.

Code is logged in diagnostics... Bike feels fine...

What next? Surely can't be another duff sensor?!

Polish it

[siamblue]

Can i ask where is the CIS?i have been having problems with mine for awhile,but mine will stall at low revs intermittantly.

[Mr OCD]

siamblue said:

Can i ask where is the CIS?i have been having problems with mine for awhile,but mine will stall at low revs intermittantly.

At front of engine, between No2 and No3 spark plugs on rocker cover

[Mr OCD]

bass gt3 said:

Sounds like it picked up a bit of interference while cranking. Not too much to worry about but clear the log and keep an eye on it.
Number 2 is the most likely, especially if the sensor is in the middle of the cam cover close to spark plugs. Just check where the connection is and make sure its away from any source of electronic interference

Thank you sir ...

Will check all that ... My worry is this is exactly what happened last time with logging code but starting ... A few days later it would not start at all with code 11.

[siamblue]

Mr OCD said:

At front of engine, between No2 and No3 spark plugs on rocker cover

Thanks for the reply,so it's the camshaft sensor then?

[Mr OCD]

siamblue said:

Thanks for the reply,so it's the camshaft sensor then?

Correct name for it is Cylinder Indicator Sensor (CIS) but yes basically the same thing.

[Mr OCD]

bass gt3 said:

1. Clean the mounting surface where the sensor is.
2. Make sure that the sensor cable and especially the connection between the sensor and the loom is away from high current / high voltage cabling like HT cables or the coils. At the connection there's no shielding so can be a point of interference.
3. Make sure that the sensor cable is only grounded at one end, usually at the ECU

Sounds like it picked up a bit of interference while cranking. Not too much to worry about but clear the log and keep an eye on it.
Number 2 is the most likely, especially if the sensor is in the middle of the cam cover close to spark plugs. Just check where the connection is and make sure its away from any source of electronic interference

Edited by bass gt3 on Friday 8th August 19:34

1. Done (was done when fitted it)

2. I've moved the cable away from the centre coil and up and over the top of the loom.

3. ECU is under the airbox so that's not getting checked lol

It started this morning fine ... although I waited till the system checks were complete before prodding the starter button (just in case they cause interference) ... so far so good ...

[Mr2Mike]

bass gt3 said:

Nope, not even close.
By timing off of the cam signal, as long as it isn't VTEC,, you have a constant against the crank. So there's no need for a crank sensor at all in fact. Go look how Ducati run. They use a 48-2 pickup off the bottom cam belt wheel.Hence your arguement of the ignition timing being inaccurate is in itself, inaccurate.

You aren't reading what I wrote. I mentioned that cam sensors, when used on conjunction with a crank sensor typically provide only a single pulse to identify engine phase in order to implement sequential injection and COP ignition. Timing from this single pulse (with interpolation) would provide poor accuracy. Obviously it's possible to have a cam trigger wheel with multiple teeth, but see below.

bass gt3 said:

And consider that acam wheel turns at a ratio of 1:2 to the crank. So 48-2 on the camequates to 94-4 on the crank, which initself far exceeds the normal 60-2 configuration.

Sorry but that is completely backwards! To gain the same positional resolution from a cam sensor, you'd need twice as many teeth as you would on the crankshaft, simply because it's turning twice as slowly. The more teeth that pass under the sensor for a given crank rotation, the better the positional accuracy that can be achieved.

bass gt3 said:

And consider that ECU's use the X minus Y as the count the teeth, so they know/predict when the next timing event is coming, even as the crank is accelerationg. Far superior to the old single pick up systems of old.

Also consider that the crank speed varies significantly through the 720 degrees of a four stroke engines cycle, so there is always some residual error when you use time to interpolate between tooth positions. The more teeth you have, the better the final accuracy. This is why I mentioned trying to timing from a single tooth sensor on a cam shaft would be a disaster.

bass gt3 said:

As for end of injection and ignition timing how does the crank do this?? Surely you're getting confused?? The injection timing when run as a sequential system is taken from the CAM signal as it's relative to the cam cycle, whereas the crank will read the same position for power and exhaust TDC's hence you now have two possible crank angles, 360 degrees apart at which to end injection. How would the ECU derive this from the cranK?????. You CANNOT run sequential without a cam sensor, but you can run batch or sequential without a crank sensor.

Yes, a sequential system requires cam information. There are two possibilities:

1) Time everything from the cam, in which case a crank sensor is not needed, but you have the downsides of lower resolution for any given trigger wheel tooth count, and you have any slop or resonances introduced by the cam drive system to deal with. For instance Honda used this system on the older VTEC engines.

2) Time everything from the crank sensor, but use a simple cam sensor to determine in which of the two engine phases the crank is currently at. The cam sensor in these systems needs only a single "tooth" to for the ECU to determine engine phase. This is how the majority of sequential systems work.

In the second system, once you have determined engine phase correctly, it's possible to simply track it through the crank sensor, i.e. at every complete revolution of the crank you know the phase has changed. In practice you'd usually keep checking the cam phase sensor to ensure you can quickly re-synchronise if e.g. you miss a tooth count, but certainly there are cars that will continue to run after being started if the cam sensor is disconnected (e.g. Fiat Coupe 20V)

[Mr OCD]

This old problem reared it's head again today... So bumping this thread up as not sure where to go from here.

Bike just cranked over on starter for ages ... Left it on charge whilst went to work... Come back and tried again and code 11 flashes straight up.

Turned off and tried to start and after turning over a few times it fired up.

[Mr OCD]

Well it started this morning on the button as it normally should...

I'm being to suspect interference of the sensor from somewhere. Given the spark plugs are overdue for a change I'm wondering if it is possible this could be the issue.

[bgunn]

Not sure if the R1 uses Coil on Plug - if not, check HT leads too? Might be a coil breaking down also?

[Mr OCD]

bgunn said:

Not sure if the R1 uses Coil on Plug - if not, check HT leads too? Might be a coil breaking down also?

It does use coil on plug ... it is possible and I should be able to test the coils.