[ROVERNET - UK] Ignition systems

defender110 at ozemail.com.au defender110 at ozemail.com.au
Wed Jan 17 22:57:07 GMT 2007

 From the searching I have been doing, it seems that you can decrease 
the time constant of the coil primary by reducing the coil primary 
inductance. This allows the coil to recover faster at high revs.
This comes at a price ... reducing the inductance reduces the stored 
coil energy and the "fatness" (not voltage) of the resultant spark.

Another problem seems to be that most ECU controlled ignitions are 
designed around a particular coil inductance value. Changing to another 
coil could have serious ramifications for the rest of the vehicle's 

We fortunately do not have this problem with our basic Kettering 
systems. <grin>
They seem to be very tolerant of our "mix n match" component approach.

I have not heard of increasing the permeability of the iron core ...
Virtually all iron transformer cores are made from the same laminated 4%
silicon steel.

I too am a CDI fan, having built a few in the past, but ...
Anyone fitting a CDI unit to a "standard" coil must reverse the coil 
primary connections as the output polarity is reversed to the standard 
The faster CDI spark rise time also can create severe plug crossfiring 
if attention is not paid to lead layout.
I found these out the hard way <gloom>

Maybe these are some of the reasons that CDI's have fallen out of 
favour?  Paul?

South Oz

Paul Smith wrote:
> There are 2 main types of electronic ignition, TAI and Capacitor Discharge Ignition.
> Personally I like CDI but they are out of favour currently.
> I think Pertronix are a Hall Effect or inductive sensor triggering a TAI.  I have points triggering a TAI.
> On a normal (Kettering) ignition, the points take 3 to 4 amps (more on Jags, don't kow why) which runs through the coil primary.  The coil is a step up transformer.  When the points open breaking the current (fairly quickly, there is arcing) the spike change induces a high voltage in the secondary - the HT side which goes to the spark plugs.   If you get 15kv it will produce an ok spark at the plug.
> However it gets trickier...
> As the revs rise, the time for the primary current to establish itself (recover) once the points close can mean it never reaches the 3A, so the secondary voltage falls.  This happens with V8s easily, since there is half the time of a 4 for each cylinder fire.  
> Soooo the solution is to extend the dwell time - the time that the points are closed establishing that current.  This can be achieved mechanically by 2 sets of points, so that one close shortly after the other opens.  A TAI does it electronically, by switching a transistor to reestablish primary coil current.  The other solution is a better coil - they have higher permeability ferrite in them, I don't know more.
> This also removes arcing (and thus lots of wear) from the points since they now handle 0.3A or so.
> The next step (a la Pertronix) is to replace the points with a non contact trigger, a magnet of some form.  This elimminates the need to adjust at regular intervals.
> -----Original Message-----
> From: rovernet-bounces at lyris.ccdata.com
> [mailto:rovernet-bounces at lyris.ccdata.com]On Behalf Of geffandjulie
> Sent: Thursday, 18 January 2007 4:50 am
> To: rovernet at lyris.ccdata.com
> Subject: Re: [ROVERNET - UK] P6 -SD1 spark plug gap?
> What is a transistor-assisted ignition?  That's a new term. I think I 
> understand electronic ignition, and the Pertronix system, but...

More information about the rovernet mailing list