On Wed, 3 Nov 2004, Sascha wrote:

> Hi guys,
>
> I am designing a multispark ignition for my car and I encountered a problem.
> I whant to know how much time the ignition coil needs to be driven so it
> stores the maximum magnetic field in it's core. Unfortunately I can not find
> the specs of the induction coil so I can compute it. Is there any way I can
> find this time? The induction coil has a primary winding of about 200 turns
> and a secondary winding of 20.000 turns. That's all I could find about it.
> If I keep the driving transistor saturated too much I loose power and if I
> turn it off too soon I will loose energy in the spark. None of this is
> acceptable.
> First I thought of adding a emitter resistor and measure the voltage drop
> across it. But this again means power loss 1.5V across the transistor, 0.5
> across the resistor and if the battery is not very strong when I try to
> start the car it will drop to 9-10V so the remaining voltage across the
> induction coil will be as low as 7V. And this will translate as a weak spark
> right at the moment where I need the strongest.
> I want to drive the coil with a BUT12 transistor which is driven by a PNP or
> a PMOS.

You can turn the transistor off at a set current. But you want to do this 
when it's time to make the spark. So you turn the transistor on a set time 
before it's time to spark. Since the associated pll cannot be perfect you 
arrange for the spark to occur when the external ignition signal comes, 
and this will be before the maximum possible energy, as timed by L and 
Isat_coil is defined. Again: you make a pll that syncs on the ignition 
pulse, and have it trigger the transistor Tx before the next expected 
ignition pulse is due, where Tx is the maximum time you can have the 
transistor on before the coil saturates, for a given supply voltage, 
Lcoil, Rcoil, Vcesat. Then you need to compensate for coil temperature, 
supply voltage, and Vcesat for the transistor as it heats up if you want 
to squeeze out top performance. And at high rpm you probably cannot put in 
multiple sparks, as there is no time to charge the coil properly.

Then there are magnet-compensated spark coils which can take twice the 
current of a normal one (and 2 times the energy probably), or twice the 
speed for the same energy (on account of L/2 for the same energy due to 
the compensating magnet in the magnetic circuit). Then there are 
twin-spark coils which are made to be driven push-pull drive and give a 
spark at each transition. Then there are high energy spark systems where 
it is easier to do the timing (multiple short discharge of the storage 
capacitor into the ignition coil). Then there are multiple-plug systems 
where the cylinder head has several plugs (typically two). This can double 
the apparent flame speed and is probably better than any multiple spark 
ignition used alone.

Peter
_______________________________________________
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist