RT

Am running a JDM STI7 RA, Aquamist 1s kit.

Since I had the DD up and running (approx 1 week ago), I have logged the car running approx 28-32deg BTDC at 4500-5000rpm, 1.03bar boost. 98RON fuel and WI as above. I also monitor knock as best I can, on a KL.

Just this morning, I thought the car felt a bit more sluggish than usual. I did not have my laptop, but I do have a PSI3 plugged in. Flooring it again, instead of the usual 28-32deg ign advance I am used to seeing, the PSI3 displayed approx 22-24deg. It is the same tank of gas, and I have not seen any activity on the KL - Not even a single green (mounted on block set to max sensitivity). I tried again later in the drive with the same low-20s ign result. WI is working fine, I do have the DDS2 flow sensor to check that.

Any ideas behind what the ECU is doing? In the past, I might have thought the car felt sluggish on occassion, but this is the first time I've actually eyeballed lower ign adv figures. I will log it as soon as I can.

In the meantime, does anyone know what the ECU is doing? Does it regularly retard/advance timing to perhaps "check" something?

Cheers.


EDIT:
No change in ambient temperature either. If anything, it was even colder this morning, if 28degC can be considered cold!


[Edited by RT - 12/22/2003 3:38:49 AM]

RT

An added twist to the tale.

On very next drive, the timing remained "retarded". About 15mins into the drive with the car well warmed up, boost dropped from the usual 1.1-1.2bar to 0.5bar. To my understanding, this is the "safe mode" boost when the ECU does not pulse the boost solenoid.

This has happened about 4 times in the past in the 2yrs+ I've had the car. I've always previously thought it was a sticky boost solenoid, and changed it the last time it happened about 6mths ago.

The 0.5bar of boost lasted about 3mins, after which the ECU recovered (as suddenly as it began) and gave me back max boost at advanced timing (30+ deg @ 4500rpm, 1bar).

This seems to suggest that the ECU had done 2 things - retarded timing and reduced boost - in an attempt to do something or run some test. Is anyone aware of any such "test" that it may be performing? I have read it runs catalyst efficiency tests and the like. I must reiterate that the KL was quiet through all this, not even a single flashing green in the last 2 days, so I'm pretty certain it is not real det.

I've pluged in both the PSI3 and the DD and both showed no current or historic fault codes.
On the DD, my advance multiplier is currently showing 15.

The "retarded" timing felt like the car was running basic base ign map values, with no advance map added, although since I have not done a full log with it running in this condition, I can't be 100% certain of this.


Is anyone aware of any such test that the ECU may do that will bring the car momentarily into "safe" mode?

Bob Rawle

There is no test ... the ecu "thought" it detected knock, it would have seen a higher knock sensor output and so reacted, it dropped boost cos it saw it for too long and played safe. Now it probably wasn't as you saw nothing so maybe you have some sort of resonant point that sets if off.

bob

RT

Thanks Bob. Would the ECU react to what it thinks is "knock" by retarding timing / reducing boost for 20mins in a drive and then *suddenly* give all timing / boost back?

Wouldn't it give back say boost, then timing in gradual steps, as opposed to all at once?

Also would my advance multiplier be at 15? Wouldn't it be lower at say 5 to 10 if the ECU thought it saw knock?

Cheers.

Bob Rawle

The ecu can go to limp home for a variety of reasons and knock sensor output is just one, for example there could be a reaction to maf output if you exceeded the maximum of that for too long and also there is a boost solenoid duty limit which could also trigger it. I'm speculating here a bit, whatever the multiplier the ecu is learning, if it saw excessive noise then maybe it decides to drop boost, its limp home mode.

Bob Rawle

Also the PSi 3 displays ign and knock correction, the ign is the base, the correction will vary with the multiplier, in addition is the learning which is then applied above and beyond, learning can be positive or negative.

bob

RT

Thanks for the reply Bob.

Boost solenoid duty limit? Thats interesting, I did increase my restrictor size to drop the 1.4bar spikes. Do you think that the boost solenoid could be hitting its limit while trying to reach the target boost? (The higher the solenoid duty cycle, the higher the boost right?) Previous DD run showed max duty cycle to be approx 60% only tho at 1.03bar boost, gear 3 WOT.

Also, this solenoid duty cycle may not account for the ign timing being dropped, would it?

thanks.

dowser

I've also had a crank position sensor cause this (on a WRX though) - first time there was no CEL, just the boost drop. Second time it lit the CEL.

Basically, the ECU is seeing something it doesn't like, and dropping boost (as Bob says, many things will put it into limp-home mode). Strange for me is that it retarded timing first - suggesting your ecu thought it was hearing knock, couldn't fix it with retardation so dropped the boost too.

Have you confirmed your KL output with det cans to be sure?

Richard

RT

Not confirmed the KL output with det cans. But I've always felt the KL was more sensitive than the OEM sensor. This is because I have seen 2 greens or 2 greens + 1 amber with no reduction in boost or timing, in the past.

Also, sometimes 2 greens is accompanied by an audible det-trinkle. Most times it is not.

dowser

If you can hear det, and the KL isn't giving big red, it ain't working! Is it set to max sensitivity, and where is the sensor located?

Be very careful and *never* rely on the ECU's knock circuitry - simple modifications to exhaust systems can render it next to useless! Although it is interesting that it seems to have registered det at some point....

Are you sure about the quality of your fuel? I'd make some det cans and do some careful observations....

What boost are you running, and what's your max airflow voltage?

Richard

RT

No mods, intake standard, exhaust standard, downpipe cat standard...

Fuel quality: I used the same pump from the same station. I have been doing this for the past 2 yrs.

Max boost: approx 1.1bar in gear 2, 1.2bar in gear 4, WOT.

Max MAF voltage: I only logged in mass air flow (grams / sec), value was approx 200g/sec at WOT, 5800rpm, 0.93bar boost.


EDIT:
Knock Correct values were still in the 7-10 degrees *positive* region on WOT. Wonder why the overall timing dropped. Seemed like altho the KC was positive, a different base ign map was used?

[Edited by RT - 12/23/2003 12:11:30 PM]