Using rocker arm as cam position sensor?

[Kitto]

*EDIT* rewritten it as i had made a confusing error.

Hello everyone,

This may sound like a bit of an oddball question, one that is more so sensor based than Link based.

Using a rocker arm as a cam position sensor, can it be done? We all know the conventional way is to take a reading from the rotating cam or cam sprocket as a home position reference for the ECU however i'm wondering if the same thing can be done by reading the rocker movement?

My idea is as follows;

Weld in a fitting to the rocker cover for number 6 inlet rocker, use a hall effect sensor such as the GT101 to have around 1-2mm of air gap to the rocker when the rocker is at idle (valve closed, sensor state High). Obviously if the cam is changed then clearance will need to be rechecked.

My thoughts of what might be a problem;

- The sensor will be in a high state for most of the time and the change point will be when the rocker moves down causing a low state on the sensor, will the ECU accept a low signal as the trigger point?

- Hydraulic lifter, so the rocker height may vary

- At high rev's the rocker may tend to float, heavy valve springs should solve that however

- What point to use as the trigger point that the ECU will reference as home? When the rocker moves downwards away from the sensor causing a low state or when the rocker moves back up as the valve is closing causing a high state but noting that the sensor will sit in a high state far longer than in a low state.

- Heat was of some concern, but the GT101 is rated to 150C.

You must be thinking why i one would want to do that when you can just reference the cam gear..? Sure using the cam gear is easy, there are even cam position kits available for the engine in question (RB30E) however they are pretty ugly, and in your face at the front of the engine - you can even use the factory distributor for home position but i have deleted it already. I'm trying to keep the factory OEM appearance and well i like a challenge.

I have almost ruled out fitting a discreet sensor behind the cam gear (there just is no room), hence my train of thought took me to look at sensing the camshaft itself or a rocker.

Thanks for your time,

Regards, Dan.

[Adamw]

This is a little difficult to explain clearly in words, but most hall sensors wont work predicably if oriented so that the target moves longitudinally/axially towards the sensor, they are designed to have the target "swipe" past the end of the senor traversely. 

So if the sensor was inserted in through the top of the rocker cover I would say it would likely be un-reliable.  If however you could poke the sensor in through the side of the rocker cover so that the tip of the rocker wipes past the end of the sensor then it should work ok.

Another option is to go for a missing tooth wheel on the crank and dont bother with the cam sensor (no sequential).

Another option might be to poke a sensor down in the old distributor hole and grind the old distributor gear into a single tooth lug shape?

 

[Kitto]

1 hour ago, Adamw said:

This is a little difficult to explain clearly in words, but most hall sensors wont work predicably if oriented so that the target moves longitudinally/axially towards the sensor, they are designed to have the target "swipe" past the end of the senor traversely. 

I understand completely what you are saying. That might pose an issue.

1 hour ago, Adamw said:

So if the sensor was inserted in through the top of the rocker cover I would say it would likely be un-reliable.  If however you could poke the sensor in through the side of the rocker cover so that the tip of the rocker wipes past the end of the sensor then it should work ok.

That is an option i'll look at, the RB30 rocker cover is just perfect to have a boss welded on directly above the inlet rockers.

1 hour ago, Adamw said:

Another option is to go for a missing tooth wheel on the crank and dont bother with the cam sensor (no sequential).

Nah, going sequential.

 

1 hour ago, Adamw said:

Another option might be to poke a sensor down in the old distributor hole and grind the old distributor gear into a single tooth lug shape?

I did think about this initially weeks ago, however due to the design of the head and the distributor hole it simply won't work - well not even remotely easily. Certainly not using the distributor gear, if i could get the sensor to look at the cam's teeth then machine off all the teeth bar one that would work but not an easy feat.

[Stevieturbo]

I did this a few years ago with another ecu brand although I used a generic threaded hall sensor pointed at the steel roller tip of a rocker.

 

Whilst it did work to a degree, it proved a little unreliable.

 

However that ecu insisted on the cam sensor being validated full time. If the ecu allows you to ignore the cam signal once in 720sync then that would not be a problem but mine didnt at that time..

Some ecu's allow that,  some dont, and some crank triggers lend themselves a little better to this than others ( ie those with an index/missing tooth etc )

 

Although if there is room...screwing say a small cap screw into the end of the camshaft..or somewhere along its length ( but the dead end is most sensible ) would give you a valid tooth to use.

 

Does the engine you are using not already have some sort of phase trigger ?

[Kitto]

13 hours ago, Stevieturbo said:

I did this a few years ago with another ecu brand although I used a generic threaded hall sensor pointed at the steel roller tip of a rocker.

 

Whilst it did work to a degree, it proved a little unreliable.

How did you find it unreliable? Hit and miss on the the trigger point?

 

13 hours ago, Stevieturbo said:

However that ecu insisted on the cam sensor being validated full time. If the ecu allows you to ignore the cam signal once in 720sync then that would not be a problem but mine didnt at that time..

Some ecu's allow that,  some dont, and some crank triggers lend themselves a little better to this than others ( ie those with an index/missing tooth etc )

The G4+ is ok with receiving a cam sync once every 720 crank rotation.

The current crank trigger i have bought is a 12 tooth flat, but i can change it to a 12-1.

 

13 hours ago, Stevieturbo said:

Although if there is room...screwing say a small cap screw into the end of the camshaft..or somewhere along its length ( but the dead end is most sensible ) would give you a valid tooth to use.

Yeah i know this would work perfectly, there is no room for this near the cam gear and it's pretty busy inside the head as well. (Rockers, lifters and lifter blocks)

 

13 hours ago, Stevieturbo said:

Does the engine you are using not already have some sort of phase trigger ?

From factory, The RB30 uses a distributor with an internal cam trigger with two trigger points. One trigger is every 1 degree of cam rotation and the other is every 60 degrees of cam rotation.

They are very unreliable for accurate trigger measurement. The timing belt and tooth drive on the distributor introduces A LOT of timing scatter.

Whilst a lot of people remove the factory wheel out of the distributor and replace it with a single slot wheel for cam sync on a crank trigger setup and it works quite well, i personally wanted to delete the distributor altogether. There are aftermarket kits that look at the cam gear itself for cam sync but they are large and bulky. Hence my idea of looking at a rocker arm (extremely discreet and quite hidden).  

[Stevieturbo]

10 hours ago, Kitto said:

How did you find it unreliable? Hit and miss on the the trigger point?

 

The G4+ is ok with receiving a cam sync once every 720 crank rotation.

The current crank trigger i have bought is a 12 tooth flat, but i can change it to a 12-1.

 

Because the ecu was always validating both crank and cam...the signal was just erratic, so any loss caused the engine to cut out. So I wouldnt say the actual trigger point was the issue....just that it was unreliable. The trigger point should have a wide window of being ok anyway given all it is doing.

But of course all the ecu needs in order to start and run would be a single reference to ID cyl 1 and thereafter as long as crank signal is good with all the teeth etc it expect and no errors it can very happily ignore the cam trigger. ( But again some ecu's allow you to do this, some dont. Not sure if Link does )

So whilst it didnt give me a perfect signal, it would still have been a good enough signal to do the above....except my ecu at the time wouldnt allow it. I did change to a different ecu later on and ran with that same trigger for a few months, until I was able to make use of the OEM trigger setup but that required a lot of work as I'd rendered it inaccessible with other work.

 

But if you had a little room on the end of the camshaft to screw in a small M6 capscrew you could very easily use that and it would give you a good signal directly via the camshaft rather than trying to use a lobe or rocker.