Servo Throttle controller for Petrol Engine (EMI interference issues)

Status
Not open for further replies.
K

kiwirider

Member
Hi All,

I'm trying to get a throttle control system working for a petrol engine ( a servo controls the throttle based on an rpm reading of a wheel - rpm is detected by a hall sensor using teensy interrupts).

I am having major issues with EMI caused by the ignition system of the small petrol engine I am dealing with.

I am using interrupts on a teensy 3.1 (which all worked fine when the engine was off) to count the number of rotations of a wheel.

A hall sensor is used to to send a digital signal to a teensy input pin on each wheel rotation.
search


I have screened the cable between the teensy and the hall sensor (about 500mm long) but the interrupt is still triggered when the hall sensor or the teensy gets close to the engine ignition. The engine runs at 1500-4000 rpm whereas the wheel that is measured by the hall sensor should be in the region of 100-400rpm.

Even when I unplug the sensors from the teensy and just put the teensy board near the EM field the interrupt is triggered (and I have tried adding a pull-up resistor with no luck).

I'm way out of my depths here so any help would be appreciated. I've looked into ferrite beads, low pass filters etc. and to be quite frank, I don't even know where to begin so any assistance would be appreciated.

My end goal is to have the teensy controlled via a wireless signal - and be able to send the rpm reading through a wireless signal, but I need to have the basics working first.

Cheers.
 
Designing circuits for the rather rough car environment is a separate engineering science... Important keywords are shielding, grounding and inductive/capacitive blocking, surge protection (diodes) and so on. Hammond mfg is a great manufacturer of liquid proof shielded aluminum boxes. That's from where I would start.

Using an optocoupler to galvanically isolate the hall sensor (and other inputs) from the microprocessor is highly recommended.
 
Indeed car ignition is a tough environment!

You tried a pullup resistor. What value? Maybe try a lower impedance, like 100 ohms?
 
I've shielded everything, and the board is in an aluminium box as Theremingenieur suggested - very little, if any change to the interrupt. (I haven't looked into octocouplers yet ).

Paul, I tried it initially with a 500 ohm pull-up but will give a 100ohm a go and let you know how that goes.

I'm also open to any other ideas (maybe I'm making things harder than they need to be). I basically just need to read the rpm of a mechanical wheel which is within 300mm of the petrol engine (while running).
 
Status
Not open for further replies.
Back
Top