Cam position sensors

[Chris Wilson]

I am about to try and fabricate a cam position sensor set up on a Nissan RB26DETT Skyline engine, as I believe that the stock photo electronic "CAS" sensor is not always read cleanly by none stock ecu's? I am new to this, so pardon me if this is a daft question, please. I have vernier adjustable camshaft pulley wheels. Should I read the cam position from the hub, or the belt pulley part of the wheel? If you imagine looking at the front of the wheel / engine, will a sensor in front of the wheel reading a tab face on, read as well as from a tab on the periphery of the wheel? I believe I can mod the stock, sturdy and dowelled cast alloy CAS bracket to hold a sensor firmly, but it would need to read a tab from in front of the wheel.

Is there any good reason to favour a crank position sensor on the front or back end, (reading tabs on the flywheel, say), of an in-line six cylinder engine?

Thanks.

[Chris Wilson]

Just as an addition, someone sent me some photos of a crank position sensor based on a ring fastened to the OUTER part of the stock, two piece, bonded rubber crank harmonic damper. So the teeth were on the outer ring, mounted, effectively, to the crank on the rubber ring. That doesn't seem a good idea to me, is it in reality an issue? Thanks.

[RossB]

Ideally you would mount the cam trigger tooth on the cam pulley. The positioning of this will be more critical if you are using a multi-tooth ref/sync mode as apposed to a missing tooth crank trigger which gives more scope for movement in the sync sensor trigger position. It doesn't matter which end of the crankshaft you trigger from but I would not recommend using flywheel teeth as a trigger, you will probably find it easier to mount a trigger wheel and sensor on the front of the engine. Mounting the trigger on the harmonic balancer is an absolute no no! This will lead to in-acutrate timing, it must be mounted on the part of the pulley that is fixed to the crankshaft (not the rubber mounted part).

[PQatPIT]

Mounting the trigger on the harmonic balancer is an absolute no no! This will lead to in-acutrate timing, it must be mounted on the part of the pulley that is fixed to the crankshaft (not the rubber mounted part).

Hmm, is there any measured data to back up this? On the same engine, on the same wheel?

Few months back I tested an '91 RB26 engine stock pulley wheel. I attached it to very heavy steel table, with wheel's big center screw and also those smaller ones to keep it from turning. Then I made a tool with two meter shaft to try to turn rubber bonded piece against the one that is fixed on the crank. I put a friend to turn and while he was pushing his eyes off it turned almost not at all. I would estimate that it turned less than one degree.

I have done a couple RB26 installs that have had material removed from stock wheel outer ring that is rubber bonded on crank fixed part, to form a 12 tooth sensor wheel. Compared to OEM cam mounted position sensor system there has been noticeable step up in accuracy when measured with good ignition lamp, in any rpm and in any load site.

So, I disagree but I haven't done any serious digging into this so really I do not know. And disagreeing MoTeC guys might be a losing fight already...

[RossB]

The only way to acurately measure this would be to set up a phase difference test and log the results throughout the whole working rev range of the engine. Then there are other factors such as detreriation. Now lets assume we are using a 12 tooth trigger and you have 1 degree of movement, the balancer can move in both directions so your tuning can only be considered acurate to within 2 degrees (the M800 is acurate to 0.1 degree). Two degrees may not sound like a lot but it can make a big difference in a boosted engine, and. Now if you have 1 degree of movement between 2 of the trigger teeth your RPM calculation will be out by 3.3% so at 6000 RPM this would make an inacuracy of about 200 RPM (this would be worse if you have more teeth or more movement). I would not recommend this method for mounting a trigger on any engine.

[PQatPIT]

The only way to acurately measure this would be to set up a phase difference test and log the results throughout the whole working rev range of the engine.

We could also take a long oscilloscope log and analyze if significant edges have fast changing interval value differences. If there are inaccuracies between teeth they should show up here, right? I mean, significant edges should occur at slow changing intervals, and interval relating to engine speed. If fast interval changes then it is an inaccuracy. When I checked the last RB26 project's crank sensor signal with high resolution scope I didn't see anything like this but I didn't take a log and analyze it afterwards.

If inaccuracy happens every time at the same rpm and the same load then this inaccuracy wouldn't affect ignition timing the wrong way, would it? Incoming signal represents reality and even if we knew it is inaccurate we could live with it if it is everytime the same. Right? Is there something I do not see here? Maybe dynamic behaviour of crankshaft/balancer wheel combo relating engine speeds, loads and/or crank position?

In this last RB26 project there are timing marks at pulley wheel, they are on the part that is fixed to crankshaft and there is no rubber between. When checked with high quality timing lamp there was no ignition timing drift I could see. Of course human eye is not the fastest instrument here but it was a huge step up from oem system. Of course timing lamp has its own errors but I would say they are same all the time and do not come in to play here.

I understand your view here, let's not give errors any chances to creep in. I'm wondering if the error really is significant.

Thanks for your time!

[RossB]

We may be talking about small in-acuracies here but you are also potentially introducing a variable in to your tuning that you cannot control. You say that the in-acuracy is the same at any given RPM but is it? The harmonic balancer is bonded together, what happens to the bonding when it gets hot, contaminated with oil or deteriates with age? How is it affected by RPM rate of change or variations in engine loads? You wouldn't mount your crankshaft sensor on a flexible mounting so why do it with the trigger. The M800 is a very acurate ECU and will allow you to tune the engine very precisely. This means that you can, for example put in a timing figure of X because that is where you will get the most torque without detonating the engine and you can be confident that you will always have a timing figure of X. Ultimately the M800 is only as good as the information it receives so as soon as there is a question about the acuracy of the crankshaft position information it compromises your tuning because you might need a timing figure of X but you might only be able to have a timing value of X +/- 1 0r 2 degrees so you have to back your timing off to ensure that the engine is "safe". This can cost you significant power, particularly on a turbo engine. If you look at it in "horsepower dollars" think about how much has been spent on the motor to achieve the power you are getting and how much you have saved by skimping on the mounting of the sensor/trigger. I know it might mean a bit more work or buying a special pulley but as a percentage of what you might have spent on the motor/engine management system it isn't much.

[SprinterTRD]

I know of a harmonic balancer where the outer ring had rotated around 30 degree. The owner wondered why the engine had no power and the exhaust glowed red hot. It wasent until we told him to check the actual TDC of the piston using the old screwdriver method down the sparkplug hole, he realised that the timing mark had retarded heaps.

[PQatPIT]

You say that the in-acuracy is the same at any given RPM but is it? The harmonic balancer is bonded together, what happens to the bonding when it gets hot, contaminated with oil or deteriates with age? How is it affected by RPM rate of change or variations in engine loads?

Hmm, exactly I'm not saying it is, I'm asking if it is? These are the questions I would like to have an answer, too. Because I do not know and everbody I have talked to do not really know. Maybe you MoTeC guys know? You must have dealt with these questions sometimes in the past, some of otur engineers, maybe?. Engine type/pulley type/good or bad condition/etc. also affects this.

I'll do some digging if I find a suitable test rig for this, and will report my findings.

PQ

[Chris Wilson]

If I turn a 8mm thick dural ring to mount in front of the cam pulley, and then screw a single steel tab on the periphery as the single tooth, how big do I need to make the tooth? At half engine speed i doubt such a tooth will seriously affect balance, but being anal, would having a bigger dural "bump" opposite it, at 180 degrees, as a balance mass be seen by the sensor?

Now, hiding in shame, as i should know this, but got to ask I'm afraid. If I buy a pre laser cut crank trigger wheel for the crank nose, (solid to the hub of the damper... ), from one of the vendors of generic timing wheels, they all appear to have been pre cut with a missing tooth. Is that what I need, or does sequential injection, coil on plug set up need all crank teeth present? Thanks, I need to be clear as I read crank and cam triggering issues are one of the commonest problems people new to installing EFI suffer.