[sci.electronics] Spark plug Inductive pickup.

wdperry@ducvax.auburn.edu (05/11/91)

I need to build a device, using an inductive pickup like a timing light, 
that will give a single clean pulse each time the spark plug fires. Pulse
needs to be at least 3.5 volts.  It must be used on the spark plug 
side of the ignition not on the coil side.

I have looked at the signal from an inductive pickup on the plug wire and
it is pretty messy.

Is anyone aware of acircuit that would cleanup this signal and output a
single pulse for each firing of the plug?  +5 volts is available to power
any chips if necessary.

Any and all suggestions would be appreciated.

Thanks,

W.D. Perry	wdperry@auducvax.bitnet
		wdperry@ducvax.auburn.edu

whit@milton.u.washington.edu (John Whitmore) (05/12/91)

In article <1991May10.164429.4146@ducvax.auburn.edu> wdperry@ducvax.auburn.edu writes:
>
>I need to build a device, using an inductive pickup like a timing light, 
>that will give a single clean pulse each time the spark plug fires. Pulse
>needs to be at least 3.5 volts.  It must be used on the spark plug 
>side of the ignition not on the coil side.

	The problem is not so much lack of signal as it is the 
waveform (lotsa jaggies).  So, use a one-shot (monostable multivibrator,
like a 74HC123).  The first transition will start the output pulse,
and until the pulse is finished, NO OTHER TRANSITIONS on the line
will have any effect.  For something like a spark plug, there is
a fairly slow firing rate (7000 RPM on a four cylinder engine
would only be 17 milliseconds, and more cylinders or slower
revs would be longer time).  Just set the monostable time constant
to 10 ms, and you'll get clean pulses from dirty input.
	Of course, you should be sure that the transformer
secondary is loaded with a low impedance (so the spark won't be slowed
by the transformer).  Best way to do that is with a diode-clamped
transistor inverter; when the transformer output is (+), the
base-emitter diode is forward biased; when it is (-), the clamp
diode (connected from base to emitter) is forward biased.  
	
	John Whitmore

tomb@hplsla.HP.COM (Tom Bruhns) (05/14/91)

whit@milton.u.washington.edu (John Whitmore) writes:

>will have any effect.  For something like a spark plug, there is
>a fairly slow firing rate (7000 RPM on a four cylinder engine
>would only be 17 milliseconds, and more cylinders or slower
                                    ^^^^
                                    John, I think you're slipping...

>revs would be longer time).  Just set the monostable time constant
>to 10 ms, and you'll get clean pulses from dirty input.