efi A Cam Position Sensor design - please critique

[AustrianVespaGuy]

@veronatii If that's an 8mm rod you're using, would not one of the M8-bodied sensors also work?

[veronatii]

Just now, AustrianVespaGuy said:

@veronatii If that's an 8mm rod you're using, would not one of the M8-bodied sensors also work?

 

I actually ordered a threaded M8 sensor as well but the length wont work. Its ~75mm from the outside of the cylinder head to the long part of the cam lobe. The threaded portion of the sensor would need to be at least 45mm in order to have threads on either side of the cylinder head bolt support. Another issue could be having exposed cable inside the head. Here's a pic of the "45mm" body sensor which has more like 40mm of threads.

 

https://www.automationdirect.com/adc/shopping/catalog/sensors_-z-_encoders/inductive_proximity_sensors/8mm_tubular/ae1-ap-3a

 

 

[TobyB]

So, as Jim noted above, it's going to be really hard

to get a precise, repeatable signal switch

off of the fuel pump lobe.  It's too smooth.  It really needs a step.

Even if that step's just a chunk of something epoxied onto the lobe.

Or a milled ring clamped over the lobe.  Or SOMETHING.

 

t 

still firing off the crank, over here.

 

[veronatii]

1 minute ago, TobyB said:

So, as Jim noted above, it's going to be really hard

to get a precise, repeatable signal switch

off of the fuel pump lobe.  It's too smooth.  It really needs a step.

Even if that step's just a chunk of something epoxied onto the lobe.

Or a milled ring clamped over the lobe.  Or SOMETHING.

 

t 

still firing off the crank, over here.

 

 

My primary trigger is on the crank as well. I plan to use the cam sensor for sequential ignition, so the ECU only needs to know if we're on the first or second crank rotation for the cycle. 

[jimk]

2 hours ago, Son of Marty said:

That's not a particularly hot area of the engine but 158' is going to be close. 

Try holding your and on a hot engine valve cover.  Human can touch momentarily 140F but not for long.  It's a lot hotter than 158F in there.

[jimk]

40 minutes ago, veronatii said:

 

My primary trigger is on the crank as well. I plan to use the cam sensor for sequential ignition, so the ECU only needs to know if we're on the first or second crank rotation for the cycle. 

You are correct.  The cam signal can bounce around so long as it doesn't run into the period where the crank position sensor is over the missing teeth section.

[popovm]

10 hours ago, veronatii said:

Operating temp is definitely my main concern for this design. 

 

I know next to nothing about any of this but maybe use a vertical hall effect sensor mounted outside the head? I am wondering if you can gut a mechanical fuel pump and put the sensor inside there to get it away from the heat. Maybe something like this TI board mount sensor reading a magnet epoxied to the end of the fuel pump shaft? https://www.mouser.com/ProductDetail/Texas-Instruments/TMAG5124F1CQDBZR?qs=zW32dvEIR3uQShIKU0DCgQ%3D%3D&mgh=1&utm_id=17222215321&gad_source=1&gclid=CjwKCAiAxaCvBhBaEiwAvsLmWP52KLRhGY8Dxwi5TY4Zb-1AOKL5aFkClO9mhf-6ZMltDX4QsAwwGhoC92wQAvD_BwE

 

I have no idea what type of signal you need out of this so maybe this is completely incompatible. 

 

[TobyB]

3 hours ago, veronatii said:

 

My primary trigger is on the crank as well. I plan to use the cam sensor for sequential ignition, so the ECU only needs to know if we're on the first or second crank rotation for the cycle. 

Da, then it should be fine, for sure.

Popov's idea of using the pushrod to remote it a bit's good, too.

t

 

[Son of Marty]

I was thinking about the push rod but got stalled when it was dinner and how to work a stiff enough spring to keep the rod in contact with the cam, but some garage engineering  might make it work, maybe a few more beers will help.

[Tommy]

1 hour ago, Son of Marty said:

I was thinking about the push rod but got stalled when it was dinner and how to work a stiff enough spring to keep the rod in contact with the cam, but some garage engineering  might make it work, maybe a few more beers will help.

Gutting a fuel pump for the sensor could make a really stealth setup. Fuel pump spring should do the job - right?

[Son of Marty]

Yeah, but I was just thinking about using the rod not the pump body for some reason my mind stuck on that.

[veronatii]

7 minutes ago, Tommy said:

Gutting a fuel pump for the sensor could make a really stealth setup. Fuel pump spring should do the job - right?

 

Not sure that a fuel pump body would fit around many EFI or ITB manifolds... But I do like the idea of using the rod/spring to move the sensor out of the cylinder head!

[jimk]

10 hours ago, popovm said:

 

I know next to nothing about any of this but maybe use a vertical hall effect sensor mounted outside the head? I am wondering if you can gut a mechanical fuel pump and put the sensor inside there to get it away from the heat. Maybe something like this TI board mount sensor reading a magnet epoxied to the end of the fuel pump shaft? https://www.mouser.com/ProductDetail/Texas-Instruments/TMAG5124F1CQDBZR?qs=zW32dvEIR3uQShIKU0DCgQ%3D%3D&mgh=1&utm_id=17222215321&gad_source=1&gclid=CjwKCAiAxaCvBhBaEiwAvsLmWP52KLRhGY8Dxwi5TY4Zb-1AOKL5aFkClO9mhf-6ZMltDX4QsAwwGhoC92wQAvD_BwE

 

I have no idea what type of signal you need out of this so maybe this is completely incompatible. 

 

Board mount means soldered to a circuit board, otherwise the switch would work from an electrical perspective.

[jimk]

8 hours ago, veronatii said:

 

Not sure that a fuel pump body would fit around many EFI or ITB manifolds... But I do like the idea of using the rod/spring to move the sensor out of the cylinder head!

If it can be held against the cam at redline.

[AustrianVespaGuy]

4 hours ago, jimk said:

If it can be held against the cam at redline.

Even that *kinda* wouldn't matter.  At least with almost all ECUs I know of, once the engine starts and you get good sync with the cam signal, even if you then loose the cam signal while the engine is still running it stays sequential as it can just count crank rotations.  At least I know MS will throw a 'cam fault' error, but it won't drop back to wasted/batch if the crank signal stays good. Needs to be able to RE-sync on the next startup of course, but just loosing cam signal at high RPM from shaft float should be kinda meh-no-big-deal.