Technical Information OBD1 Part Throttle Tuning
Part Throttle Tuning
This article outlines the steps taken to fix a part throttle tuning issue. The vehicle used was a EG Civic with a turbo-charged GSR engine, running about 9 lbs boost. Performance under boost was good, but at part throttle the vehicle had a hesitation at around 3000 rpm while cursing. Fuel consumption was high, either in open loop or closed loop. After approximately 90 minutes part throttle tuning using a wide band o2 sensor, the vehicle drove much better.
A common problem is too little ignition advance at part throttle, especially at about 10" vacuum and 1000 - 3000 rpm. This will cause difficulty tuning the fuel tables, as the engine becomes very sensitive to small changes in fuelling if there is not enough ignition advance. Driving the car we could feel a lot of surging under light throttle, so ignition was advanced at part throttle.
The original low speed ignition table.
The final low speed ignition table. Care was taken not to increase overall advance, nor add advance where the engine might ping. The peak increase was from 20 to 29 degrees at 1750 rpm and 5 inches vacuum. This change made a huge difference to the way the car ran, with a low more engine output below 3000 rpm.
The 2D view is a good way of evaluating how well tuned a table is. Lines should not cross each other, and should be evenly spaced. We had to alter the low speed fuel tables only.
The original low speed fuel table. The vehicle actually ran quite well from this table once ignition advance was increased, but you can clearly see that the table is not good, with inconsistent fuelling .
The final low speed cam fuel table, are tuning using a wide band O2 sensor. The pattern of lines in the 2d view is very typical of a well tuned engine. The lines representing the bottom four columns should be parallel and evenly spaced, the middle two columns are spaced further apart but are also parallel, and the top four columns are evenly spaced. The lowest column should be about half way between the second column and the bottom of the graph.
It is important not to make the fuel tables too rich in the bottom two columns, otherwise in closed loop the ECU will lean out the mixture 20-30%, which results in hesitation when the throttle is applied after decelerating briefly.
Idle speed was increased from 750 rpm to 850 rpm.
VTEC was increased from 4000 rpm to 5200 rpm.
The minimum engine load from VTEC was increased to 950 mbar.
The vehicle had an autometer air/fuel gauge connected to the stock O2 sensor. The combination of an old O2 sensor and the air/fuel gauge resulted in the O2 sensor feedback voltage being lower than normal. This made the O2 sensor response slow, and causes the engine to run rich (as it is getting false lean feedback via the O2 sensor).
The O2 sensor feedback voltage was incorrect, which was traced to a dirty wiring connector on the right hand strut tower. It is recommended to clean the main wiring connectors with contact cleaner from time to time.