Aftermarket EFI, speed density etc.

[robert dingli]

Peter,
> 
> I think we are getting a little mixed up here!  
> 
> >> 
> >> What is the mapping for?
> >
> >The mapping is required since the user will not usually desire a constant
> >air-fuel ratio.........
> 
> The user *will* require a constant air fuel ratio! Stoichiometric to be
> exact. This allows the catalytic converters to work correctly by providing a
> uniform gas mixture in the exhaust. ...

Thankyou for your *amazing* insight.

Let me explain a couple of things.  When I stated *aftermarket efi*, I was
referring to non OEM efi systems.  Such systems are designed and operated
with performance, drivability, efficiency and very occassionally emissions
in mind.  I have never encountered an aftermarket efi system fitted to a 
car with a catalytic converter.

Also note that the aftermarket market that I am dealing with here in Australia
has much less strict emissions laws.  Cats became manditory in 1986 and pre
June 76 there was almost no restrictions (apart from charcoal canisters).
Given the average age of local vehicles (10-15 years)  you may start to
understand the above priorities.

> 
> The EPA defined a test schedule which each manufacturer has to go through
> and the total emissions allowed during this test are defined. The test is
> usually very low throttle and accelerations, I suppose typical of a cruise
> through rush hour traffic. Thus the manufacturer has the option of staying
> stoichiometric just in this schedule, or as most have done, all of the time
> the throttle is less than fully open. At fully open throttle the oxygen
> sensor is out of the loop and the system is said to be running open loop
> which means it is running on the map data only. This is preprogrammed in on
> dyno testing and usually gives best power and performance. 

I agree here, but note that some systems (eg Ford Falcon EA26-EA77 EECIV
and EECV) also run open loop during idle (lambda = 0.9) and while cruising
(lambda ~ 1.22).  WOT can be as low as lambda = 0.85 for peak performance
and engine longevity.  Engine warm up is another matter entirely.  Current
laws state that the closed loop control has to kick in after 20-30 seconds
(depending on the country and state) and before this time generally run
quite rich.  The aim is to reduce the high concentrations of pollutants
which are produced in concentrated city environments.

> >> | 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.........  
> 
> Again Robert is getting mixed up. The reason racers use throttle angle and
> engine speed is that it is very easy to set that up on a  dyno and then know
> which row and column in the map to change. If you use a MAF sensor of any
> type then you have to define the load and this is less easy to reproduce on
> the dyno. Mapping is still fairly easy if you use the voltage output as a
> reference. Getting a stable airflow reading at a given RPM depends on
> throttle opening and dyno load. It is much easier to define the throttle
> opening and use the load to control the engine speed to where you want it.

?????  

I'm not sure who is getting mixed up here.  Map based system are much easier
to set up (on the road or dyno) than throttle position systems.  There is
a much simpler relationship between MAP readings and engine load during steady
state operation.  Basically, assuming constant volumetric efficiency (and
remember, I'm talking about the initial stages of open loop tuning), the
air flow into an engine per cycle is closely proportional to the manifold 
pressure.  Map based systems also automatically compensate for different
ambient pressures.  The main correction is for inlet air temp.

For fine tuning (dyno work), it's just as easy to set the engine speed where
required (assuming you have a decent dyno) and adjust the throttle position
to give the required manifold pressure (vacuum).  I don't understand how
Peter finds this any more complicated than setting a throttle position.
MAP readings are easily reproducable and actually mean something as
opposed to a voltage output from a throttle positon pot.

The maps generated using MAP based load are better in that each point
is relatively representative of a speed/load combination.  This is a
bit messy to explain, but as an example, consider the table values
for the 2000 rpm operation condition.  On a throttle opening scale of
1 to 10, the map values at throttle openings 1-3 will mean something 
while those above will all be the same, ie the opening the throttle 
past a certain point makes very little difference to the air flow at 
this speed.  Throttle based maps tend to have a lot of detail squashed
into small regions of the map and thus suffer from resolution problems.

Throttle position is used as a measurement of load where there isn't a
stable or wide ranging manifold pressure signal to use.  (Refer to the 
examples I posted yesterday.)  The feed forward nature of the measured
variable also allows faster control system response.  Throttle position
is useless for turbo systems as the assumption of constant upstream
inlet pressure is no longer true.

> Lastly, flap air sensors are not notoriously unreliable. If they were there
> would be ten trillion cars breaking down every minute because of them. They
> are very reliable in service and if used correctly in race engines they work
> adequately there as well. 

I disagree here.  Ask any Jaguar mechanic about the reliability of the Lucas
(Bosch) moving vane MAF meters fitted to XK engined XJ6s.  The greatest
improvement I could have made to the Jag FI system I fitted to my previously
carbed XK4.2 was to ditch the air flow meter and use a MAP sensed load 
system.  It performs much better as well.  I have also encountered faulty
Bosch moving vane meters on other brands.

Backfires through the inlet can destroy such sensors very easily.

> There is a school of thought which says they
> should be replaced because the mechanical flap across the air stream
> restricts the air flow and causes loss of power, but most racers are at full
> throttle anyway so it is kind of academic.  

The mechanical flap is not the restriction but the overall cross sectional
area.  

> Peter Wales

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
----------------------------------------------------------------------