Questions (mostly ignition)

[robert dingli]

Hi Guys,

Bill writes,
> 
> 1) Just how sensitive is the ignition timing?  Can you tell a difference
>    between 1 or 2 degrees?  What about 1/2 a degree?
> 

Bosch (in ther M1.7 Motronic, as used on our current Lancia research engine)
have a 1 deg timing resolution which is clearly visible when using a timing 
light on a calibrated flywheel scale.

Ford (EECIV and EECV, as used on the Falcon 4.0) have a 0.125 (that's right
one eighth of a degree) timing resolution.  Obviously this is significantly
less than the +/- 5 deg combustion variability that conventional engines have.
The important point is to optimise the average timing.

Personally, I've resorted to 45/128 deg (~0.35 deg) resolution as a suitable 
comprimise between accuraccy and eight bit implementation.

As Peter explained, there is little noticeable difference on the dyno for 
changes less than one or two degrees and even less on the road.

> 2) I've always thought the whole idea of spark advance was to give the mixture
>    time to get really burning.  And that the idea of advance in degrees BTDC
>    was an artifact of the distributor - what you really wanted was time BTDC.
>    
>    Why do centrifugal advance curves typically max out well below the redline?
>    Do we really want the ignition point in terms of crankshaft degrees, or
>    would it be better in terms of time?
> 

There are two primary factors which govern the ignition advance.  Firstly,
the a constant time to burn results in a linear increase in advance as a
function of rpm.  Secondly, higher rpms and loads lead to higher turbulence 
and faster burn times.  Thus the ignition advance should initially increase
with rpm and then taper off and even decrease as the engine speed increases
further.  As a function of load (manifold pressure for example) the timing
should decrease, ultimately being limited by engine knock and exhaust
temperature.

> 3) Many modern micro controlled engines have discarded the distributor entirely
>    in favor of multiple dual-output coils.  In a dual-coil system, aren't
>    the extra sparks in the exhaust-intake overlap period dangerous on a engine
>    with radical cam timing?  Why does the current Porsche 911 use still use
>    a distributor?
> 

Radical cams have been known to cause problems with dual fire ignition systems.
One thing to note is that the mixture (combustible or not) at the point of
ignition during the exhaust stroke is not under compression and will have an
extremely high concentration of exhaust in the mixture diluting it.  It's
hard enough to get a stable ignition at the best of times.

> 4) Do Bosch Motronic systems that take pulses off the flywheel ring gear count
>    these pulses and trigger events at a particular tooth number?  Are these
>    pulses used to get a better idea of the engines acceleration?  A 911 has a
>    130 tooth ring gear.  That's a little less than 3 degrees per tooth.
>    Do they use a software PLL to synthesize 'extra' teeth?
> 

Our test engine uses a 60-2 tooth wheel at the front of the crank.  The Bosch
system would have to interpolate between them to achieve 1 deg timing 
resolution.  I personally use a three tooth crank wheel to minimise the 
processing overhead on my six cylinder. (2 for a four cyl and 4 for an eight)
This is similar to the six vane hall effect distributor system used on pre
94 Ford Falcons (EECIV) and works very well.

> 5) When the knock sensor kicks in, how much do factory systems typically
>    back off the timing?
> 

Bosch systems - 3 deg backoff and then 0.5 degree advance per cycle until the
map value is reached or knock is detected again.

> 6) What is a good minimum RPM for a system?  i.e. what is cranking speed?
> 

I use a 60 rpm minimum on my system.  60 - 300 rpm is cranking and the first
map point is 500 rpm.  With EFI and electronic ignition, my Jaguar (as
unreliable as some people think it may be) will happily idle below 400
rpm and pull from 500 in fifth gear.

> 7) After you decide you need a particular size injector, say 21 lbs., how do
>    you go about finding a part number or source for it?
> 
> 7) What size injectors are used in the Chevy 305 and 350 Tuned Port systems?
> 

I generally tend to use which ever injectors I can find that come off a 
similarily sized OEM engine.  Luckily, in a lab that does a lot of alternative
fuel research, brand new petrol injectors are easy to find.  :-)

Robert 'if only the rest of the car was as reliable as the EFI' Dingli

-- 
----------------------------------------------------------------------
             Robert Dingli           r.dingli at ee.mu.oz.au

Power and Control Systems                 Thermodynamics Research Lab
Electrical Engineering                    Mechanical Engineering
   (+613) 344 7966                           (+613) 344 6728
  University of Melbourne, Parkville, 3052, Victoria, AUSTRALIA
----------------------------------------------------------------------