Aftermarket EFI, speed density etc.

[robert dingli]

Hi everyone,

> | > 1)  Speed density vs mass flow:
> 
> robert dingli <r.dingli at ee.mu.OZ.AU> replied:
> | 
> | Mass flow system still require a mapping between engine speed and mass
> | air flow and injector output (which can exist in many forms such as
> | desired AFR maps and/or volumetric efficiency maps).  
> 
> What is the mapping for?

The mapping is required since the user will not usually desire a constant
air-fuel ratio.  The desired air-fuel ratio is usually a function of speed 
and load.  Most of the system I've encountered have an open loop map which
the system would switch to under certain conditions such as high speed 
cruising, WOT and other speed/load combinations which aren't encountered
in drive cycle testing for emissions certification.   Such maps are 
determined empirically as are all the aftermarket efi systems available here.

> 
> | In fully tuned form
> | they perform as well as speed-density systems and are more suitable for
> | many production systems.  Aftermarket EFI is a totally different matter. 
> | 
> | MAP and MAF sensors are both useless for much serious performance work.
> | Throttle position is the only reliable method in such cases.
> 
> Why is this?
> 


I suppose I should define the 'serious performance work' that I've dealt
with.  Most of the racing work I've encountered utilize individual throttle
bodies for each cylinder or rotor.  They generally don't use a plenumn
and the reverberations in the inlet tracts result in useless pulsating
manifold pressure signals.  The inlet is typically fed through a filter
and air box which are considerably larger than OEM MAF sensor inlets.
Certainly, the moving vane MAF sensors are notoriously unreliable and 
fragile.

I only recommend MAP sensors for turbo applications and other systems
with stable manifold pressure signals (ie OEM plenumns).  Rotaries
(especially extended ports, bridge ports and peripheral ports) don't
tend to produce suitable MAP signals, even if they do have plenumns.
Throttle position based load sensing has proven to be the best otherwise.

Robert
-- 
----------------------------------------------------------------------
             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
----------------------------------------------------------------------