engine time controlled actuator

[dieseltekkie]

I'm trying to design a part for an engine that will control the exhaust valves on an engine much like that of a semi truck engine brake but I'm running into a bit of an issue with figuring out how to make the timing module. I need a trigger wheel, pickup sensor, and the stuff that will make it work to operate a small solenoid at just before top dead center of each piston. it is a 4 stroke 6 cylinder engine so every 120deg of crank rotation when engaged it needs to close the circuit on the solenoid and keep it open for a period of 25deg. any ideas? Don't worry too much about the mechanical aspects as I'm working that from another angle other than an idea for a tone ring. i thought about the flywheel but there is not a missing tooth or anything to designate as a timing point

 

[Harald Kapp]

You could use magnets spaced 120 ° apart on a suitable rotating part of the engine plus a hall sensor fixed such that the magnets pass near enough to the sensor so every 120 ° of rotation the sensor will register the passing magnet. You will have to add some amplification and filtering to remove noise and poduce a signal suitable for your controller.
A second sensor, spaced 25 ° away from the first sensor can generate the signal for closing the valve.

Like a bike odometer, but with more magnets and a sturdy hall sensor (bike odometers often use reed contacts, but I wouldn's recommend them in the environment of an engine since they are much less robust).

Regards,
Harald

 

[dieseltekkie]

yeah i was thinking of a hall system but a way of having 1 hall sensor with a toothed wheel would be much better and run it similar to a modern car ignition system so that you don't have a group of magnets on the harsh environment of the front of an engine crankshaft or in the confined space of the camshaft gear.

it would also be easier to balance for spinning at 2000-4000 rpm

 

[Harald Kapp]

With my setup it's 2 sensors, 3 magnets

Alternatively, but more complicated, you could forgo the second sensor and compute the required 25 ° angle after activating the first sensor from the RPM of the engine.
As a first step, youd determine the RPM by counting the number of pulses (3 per revolution) within a defined intervall of time, say 20 seconds.
30 impulses /20 seconds = 90 impulses /60 seconds = 90 RPM
From the RPM you calculate the time for 25 ° as follows:
T=25 /360*RPM in the example this is 0.000772 minutes or 0.0463 s or 426.29 ms.
(I hope I didn't miss anything here).
So you could close the valve 426 ms after opening it. Of course, you'd have to continually update the RPM and add some failsafes like e.g. a max. time the valve is allowed to stay open even at low RPM.


Harald

 

[dieseltekkie]

I was thinking of running a tone wheel with 23 teeth (the 24th tooth missing to denote #1 cyl at tdc) then there would be a 7.5deg gap between the teeth if the teeth were to encompass 50% of the time. therefore cyl #1 would close the circuit on the gap immediately following tooth #23 and then open immediately after tooth #1 passes the hall sensor thus closing the solenoid. now the trick is finding a circuit to program this into

 

[Harald Kapp]

For starters you could use an arduino, for example.

Harald

 

[dieseltekkie]

hmmm that might be very helpful

 

[MeAndSteve]

That's an aggressive project you've undertaken! I know Audi, Gm, Toyota, and others have that system in place, VCT that is. Some of them use oil pressure controlled via a computer to control oil flow to the cam actuator also called a phaser which is constantly being updated depending on engine load and speed.

If you're working on a modern computer controlled vehicle, there will undoubtly already be camshaft and crankshaft sensors in place that you may be able to draw crank and cam position information from without too much trouble.

I'd be interested in seeing your design when it's finished. Being an auto tech, I always find easier ways to make things happen in cars, but never bothered to implement any of them.

 

[dieseltekkie]

yeah, im looking for the actual control system i would need, i've already got the hall system figured out