How does the ECU meter the incoming air to determine the amount of fuel the injectors deliver to the motor? In the older FI units I used to work with it was done by determining the flow of air through the throttle body, but I don't see a metering device on the Accords to do that.

 

It uses the map sensor input based on vacuum as well as for other things like intake temp.

 

Thanks. I have been playing around with throttle bodies and was wondering if I were to replace the one I have with a larger one, how the ECU would compensate for the reduced vacuum. I suppose it will "learn" the new flow based on input from other sensors.

 

With speed density systems there is a volumetric efficiency(VE) table that is programmed into the ECU. Since there is no MAF sensor to directly measure actual air flow, the ECU bases air flow on pre-determined/calculated flow of the engine. Changes in air density are monitored via the IAT and MAP sensors with operator demands(expected airflow through the TB) monitored via the TPS. Any changes to the VE of the engine (intake manifold, TB, camshaft, etc) should have the VE table adjusted or scaled for the new airflow a component impacts on the system, as on s SD system this is the main function of how the ECU knows how much fuel the engine will need.

IMO no point in adding a larger TB unless you're running the car at WOT or need airflow greater than the factory TB can provide. Otherwise the scaling of the VE tables at small throttle openings will be incorrect as the ECU will have lower air flow(VE) rates. Resolution of throttle response will drop off, if the TB is large enough this can cause some drivability issues.

ECU can compensate to a degree, but this will be a constant adjustment and not ideal for power or drivability. With an SD system there is an added delay in adjustment as the ECU will have to calculate air density, if the VE of the engine has changed the ECU will have to recalculate again as it cannot directly measure air flow. Your fueling tables will be incorrect for the added airflow. Readjustment of the VE tables would be best, as well as rescaling of the TB(airflow past the blade) to TPS output.

Will the engine run, yes.
Will the engine run efficiently, no.

If the TB is quite a bit larger it may set a code as the ECU will be expecting a given VE of the engine, but the sensors would indicate different calculated airflow under a given load.

 

Thanks. That's good info. I pulled a throttle body off either an Integra or a Prelude (don't remember which) and it is about 60mm as opposed to the Accord's of about 56 mm. I was thinking about swapping the two. I guess I would need, at the minimum, need a new map, or even a tune. Most of my experience has been with motors which did not utilize an ECU. I once had 3 Weber 40DCOE carbs on a 2.5 liter motor and got some good improvement in the 5000 and up rpm range, but also some drivability issues below 3000. Tuning was a matter of re-jetting the carbs. I see this will be more difficult on an Accord.