problems with EFI car idle and pressure sensors

[Bernd Felsche]

David Sackett tapped away at the keyboard with:
> hi, i'm developing a simple EFI computer based on a Atmel AVR
> microcontroller running at 8Mhz for a Uni project. I have
> installed it on my old 1977 Saab 99 and merged it with a Bosch
> D-Jetronics fuel injection system. the car originally had carbies
> and i have only used the inlet manifold, injectors and fuel system
> from the bosch system.

You're using D-jet. injectors?

What are you using as a time base? Camshaft or distributor sensor?
Is the distributor _static_ or does it have variable advance?

> just to get started I decided to only use throttle position and
> rpm as inputs to the computer. a lookup table is used to determine
> the correct injector open time from these two inputs. we have
> successfully performed test runs in the car and it accellerates
> fairly well.

Do you have idle and WOT switches as well?

Given that you have only those two inputs, you can only do n-alpha
control of the engine. Your mapping must incorporate a knowledge of
actual airflow to engine speed and throttle position.

Closed-loop control by means of lambda (O2) sensor can help to
determine fuel injection amounts and to maintain appropriate
injection amounts; but the sensor needs to warm up first - which can
take a minute even with a built-in heater. 

A simple closed-loop control can utilise a narrow-band (switching)
lambda sensor. These have a voltage output of 0.45V when the mixture
is burning at stoichiometric. Either side of that, and the output
coltage swings rapidly to 1.1 or 0.1V. The 8515's analog comparator
can be used to ensure that the sensor's output is oscillating about
stoich. constantly - unless you're at WOT in which case you
deliberately want to run quite rich - and on over-run when the
throttle is closed rapidly with the engine used for braking,
allowing you to cut off fuel completely until you approach idle.

> the problem we are having is that we cannot get the car to idle
> smoothly. the car refuses to run on all cylinders until it is
> above about 2000rpm. the car runs VERY rich and the plugs have to
> be cleaned every 1/2 hour or so of running. if we reduce the
> mixture at idle it just stalls. also the system fires all
> injectors at once every revolution of the crankshaft. i have
> replaced much of the ignition system and am fairly sure it is
> working correctly.

Sounds like an ignition timing problem, in combination with
incorrect mixture to start.

Are you controlling ignition timing as well?

> does anyone have any ideas as to how to fix the idle?

Hard to say exactly what's wrong without more information.

You need to consider the effect of changing ignition timing with
speed.

-- 
 /"\  Bernd Felsche - Innovative Reckoning, Perth, Western Australia
 \ /      ASCII ribbon campaign | I'm a .signature virus!       |
  X       against HTML mail     | Copy me into your ~/.signature|
 / \      and postings          | to help me spread!            |
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org