[alfa] RE: High output coil and plug gaps

["John Fielding" <johnf@xxxxxxxxxxxxxxx>]

Hi Anthony,

I saw your enquiry on the AD about the coil and plugs.

If I install a high output coil, good leads, and NGK plugs on my car, I'd
usually increase the plug gap to get a stronger spark.  If, however, I'm
running Golden Lodge multiple electrode plugs, should I do anything?  Is
there as much of an advantage to the high output coil with the Lodge plugs,
or is some of the advantage lost?

The question you first need to answer is "How is my new high-output coil going to develop
more output with a standard coil driver module?"  The answer is that it cannot!

To develop extra energy, as spark, the primary inductance of the coil needs to be lowered,
because the equation for energy transfer in an inductor has a 1/2 x current squared factor
in the formula.  By lowering the primary inductance the peak primary current needs to go
up to maintain the same energy level.  Most coil driver modules use a peak current
clamping circuit to keep the peak current at some specified value.  Hence, if you swap the
coil for a lower inductance primary you need to increase the peak primary current to allow
the extra energy to be developed.  (This seems to be going the wrong way but as you will
see later when high rpm is required it is the only option).

The equation for energy stored in an inductor is:    E = 1/2 Lprim x I2  gives the answer
in Joules or milli-Joules (mJ).

Assuming the original coil had a primary inductance of 10mH, which is about the average
for an electronic ignition coil fitted to todays vehicles. A coil of this type will have a
primary resistance of about 1.5 ohms.  A high energy coil will have a primary inductance
of about 4mH, it hence has less primary turns and the primary resistance is consequently
less, about 0.7 ohms.  Working out the energy transferred for each, with a standard coil
driver module with a peak current of 5A, gives the following results:

For a  10mH coil with 5A current E = 125mJ of energy
For a 4mH coil with 5A current E = 50mJ of energy.

Hence, for the new coil, with the same ignition module with 5A peak primary current,
the energy actually goes down and not up!

Most coil driver modules have an internally limited maximum peak current of 5 to 6A,
anything more than this requires a change in the coil driver module to a higher rating
type. To obtain more energy than the standard coil requires the primary current to go
up to about 9A.

With a 9A coil driver module and the 4mH "high-energy-coil" the energy avaiable to be
transferred is now 162mJ, not a significant improvement over the original 125mJ with the
standard coil.  To double the energy you would need to run a peak primary current of about
12A (E = 288mJ)

Today there are available coil driver modules with up to 30A current capabilities, but
these are not drop-in replacements and  a lot of system changes need to be done to
make them work.

You might question why the high energy coil has a lower inductance?  The answer is simple.
For high rpm applications the primary inductance has a profound effect on the maximum
sparks per second the coil can develop.  This is because the inductance when the coil is
initially energised tries to prevent the current flowing in the primary, it eventually
looses the fight but in the interim it generates an equal and opposite voltage (the back
EMF) and slows the rate of change of current.  The primary of the coil is an inductance
in series with a resistance, this forms a time constant.  If the L and R are both
large the coil will never build up sufficient current before the spark is required.
Hence, as the rpm goes up you need to reduce the inductance and resistance to ensure
enough energy is stored in the short time available between sparks.  In a 4 cylinder
engine turning at 6000 rpm a spark occurs every 5ms.  An 8 cylinder at the same speed
has a spark every 2.5ms, so an 8 cylinder engine requires twice as many sparks for the
same rpm. A 6 cylinder falls in between the 4 and the 8 cylinder types.  This is why
modern engines have dispensed with a single coil and  use either a wasted spark
system, where one coil supplies two cylinders, or the later system where each cylinder
has its own coil.

If the engine requires a spark every 5ms it follows then that the maximum time for the
coil primary current to rise to its maximum is less than 5ms or the spark energy will
start dropping off dramatically at the higher rpm.  In truth the manufacturers do this
deliberately, it is a "hidden-rv-limiter" to protect the engine.  If the spark energy
falls off at the point just before the actual rev-limiter is activated the engine is
starved of power and will not over-speed as it hits the rev-limiter proper.  The old
Kettering points contact system has a very pronounced fall off in spark energy at high rpm
because the coil inductance and resistance are both high, because the coil primary current
is only controlled by the resistance to around 4A or so by using a 3 ohm primary.

The basic problem with ignition coils is that the manufacturers do not tell you what
the primary inductance is, all they tell you is the resistance, which is pretty
meaningless.  I am fortunate in that I have test equipment to measure the inductance.
From tests I have done with a huge variety of coils I know which are the good coils
and which are rubbish.  There is a certain well known US manufacturer of after-market
coils who claim their coils have higher output than the standard types.  They come in
fancy chromed canisters with impressive names tempting you to believe their claims, but
when you measure them they actually have lower energy capabilities than a standard coil,
much lower in some cases!  The very best coil I have come across is made by Lucas
and is a low cost part used on Ford vehicles.  This sells new for 1/3rd of the price of
a chromed canister coil and you can pick them up from scrap yards for next to nothing.

Some manufacturers waffle on about the ht voltage the coil can develop.  This is actually
BS!  Provided the spark voltage exceeds about 10kV the coil will be perfectly OK.  It is
very rare to require more than about 15kV even with a high compression racing engine.  The
more important factor is the energy stored and available to form the plasma arc.
Volkswagen some years ago demonstated a lean-burn petrol engine that only required about
15mJ of spark energy and a spark voltage of 5kV, but the combustion chamber shape made it
impractical to manufacture in serious volumes and the plugs were very special and costly.

The plug gap is also an interesting thing.  With more energy available the ht voltage can
peak at higher levels, this means that the plug gap can be opened up some.  But the main
reason for using a wider gap is to promote a wider flame front to get the mixture burning
faster.  The spark event contains two phases.  The initial high voltage breakdown phase
and the later lower voltage maintaining phase.  When you see the spark jump across the
plug gap all you see is the later lower voltage phase, called the plasma phase.  The
initial first stroke of high voltage is not visible and only occupies about 1us, the
second phase lasts up to about 500us, depending on the energy available from the coil.
This initial strike gets the electrons ionised and able to pass current with a lower
maintaining voltage.  It is the plasma phase that absorbs about 99% of the total spark
energy.  It is like an arc welder where you need a high voltage to strike the arc but a
much lower voltage to maintain the arc.  The plasma phase of the arc sees the spark
voltage falling rapidly to about 500V and this voltage maintains the arc and uses up all
the energy stored in the coil primary.  The more energy stored the longer the plasma arc
lasts and the better combustion occurs in the cylinder.

The multiple spark gap types are interesting to observe on a bench test rig.  The
electrode with the smallest gap will naturally make the spark jump across.  As the gap
erodes slightly the next gap with the smaller gap will then take the spark.  And so on.
The erosion is evened out between 3 or 4 gaps and the rate of wear is slowed down.  On my
Fiat Uno Turbo, which uses NGK 3 electrode plugs, the centre electrode ends up being a
tri-lobular shape with equal gaps all the way around.  The 4 electrode types end up with a
4 sided central electrode with rounded corners.  Swapping in a new set of plugs makes not
one scrap of difference to the power output on the dyno, because the energy available
allows big plug gaps which naturally occur as the plugs erode away.  The downside of very
big gaps is that the intial high voltage gradually rises throughout the life of the plugs
and eventually something other than the plug gaps start to track over and cause a
mis-fire.  As an experiment I have run up to 80,000kms on one set of plugs, when the gaps
were over 3mm, before a mis-fire forced a change of plugs.  The power output with the
new plugs did not differ more than 0.5kW in 100kW from the original plugs measured a month
earlier. This is well within the tolerance of the dyno of 2%.

John
Durban
South Africa
--
to be removed from alfa, see /bin/digest-subs.cgi
or email "unsubscribe alfa" to majordomo@domain.elided