Vitara J20A ECU Analysis

Suzuki Vitara 1999 J20A – Australian version D. Harvie 2011

Test Condition Explanation •

All tests conducted with ambient air tem p between 15-20 degrees Celsius, and a corresponding IAT within 2 degrees of this t his (measured in the air box.)

•

A 1-bar MAP sensor (standard GM unit) was used to record engine load. 0% load indicating total vacuum (~0V sensor output) and 100% load indicating atmospheric pressure (wide open throttle).

•

The objective was to build a base set of data that can be used as a starting point for an aftermarket ECU (probably a Megasquirt.) As I also wish to use a MAP sensor and dispense dispense of the MAF sensor, all measurements are judged judged against MAP vs RPM. This never was going to give a perfect relationship since since that’s not what the computer uses. However quite consistent results were observed regardless.

•

If you don’t understand the typical EFI computer, you won’t understand what I’ve written below. Megasquirt has some good reading material for starting out. out.

Idling: •

Warm engine idle speeds of 800-820RPM,

•

Typical load 20%,

•

Stable pulse widths between 2.6 – 2.7ms.

Start up: •

Cranking pulse width not observed.

•

Startup enrichment of approximately +1ms, tapering off linearly over t he first 5 seconds.

•

Fast idle of 1200rpm with a pulse width of around 3.2ms for the first minute or so of idling (given a reasonably warm engine to start with).

•

Average fast idle load was 16-17%.

Wide Open Throttle •

Test run at 1500-2000rpm 1500-2000r pm to gauge 100% volumetric efficiency pulse width.

•

Pulse widths 11-12ms reasonably consistent across 1500-2000rpm range.

•

High speed runs to red line (6500rpm) (650 0rpm) showed a reasonably consistent pulse width gradually increasing to just over 12ms. Interestingly no taper off of the pulse width was observed up to redline. Though this data is only rough because acceleration acceleration enrichment could not be separated from the base map, as I have not been able to hold the vehicle at high RPM under full load. Regardless this is a good starting point and I don’t think it will be too far off.

Mid-throttle •

Not much interesting to talk about in this area. It basically follows a linear path between idle load to wide open throttle for pulse width. However for completeness here are some generalized ranges of results from the data sets: o

< 10% – closed throttle – fuel cut or 1.7-2ms

o

20% – 2.5 to 2.7ms

o

30% – 3.5 to 3.7ms

o

40% – 4.5 to 4.8ms

o

50% – 5.5 to 5.6ms

o

60% – 6.5 to 6.6ms

o

70% – 7.3 to 7.8ms

o

80% – 8 to 9ms (this is based on much more limited data set).

o

90% – 9.8 to 10.8ms

o

WOT – 11.5 to 12.5ms

Conclusions •

Given the 2.5ms idle condition, this follows “standard” practice of sizing injec tors to ensure a pulse width >2ms at idle to ensure stability (i.e. injectors able to fully open and close.)

•

At 12.2ms @ 6500rpm, this represents a duty cycle of 66% (sequential injection, one pulse per two revs.) Even with keeping duty cycles below 90%, if RPM is not increased beyond the current red line, this should be able to support an increase of 30-40% in MAP without ex tra fuel pressure or extra injectors (maybe 6ish PSI boost).

•

From the fairly large amount of data collected, I would conclude this should be quite an e asy engine to get up and running on an aftermarket computer.