The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support RTM Racing
Please Support Rix Racing

ECMlink Cylinder 4 Knock?

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

shinzon

15+ Year Contributor
350
72
Sep 11, 2006
St. Petersburg, Florida
1). No boost, vac, or exhaust leaks.
System holds up to 20psi before my tester begins to leak. Car is running stock boost.

I didn't have a boost gauge on the car during this log, but I had it on the car within the last month or so to verify that it makes 9-10psi at full boost, but it does creep to 12-14psi above 5500rpm.

2). Verify mechanical timing.

OEM timing equipment was replaced about 4 years and 8k miles ago. Replaced crank sensor back in September and verified all timing marks were lined up.

3). Verify base timing

I don't have a timing light, so I have not done this yet. I'll rent one soon and verify. I did notice that when I grounded the timing lead to do a BISS adjustment that ECMLink reported the engine was running at 6deg advance, not 5deg, but 2Gas have non-adjustable timing, right?

4). Ignition system

NGK BPR6ES gapped at 0.28. Newer aftermarket coils, newer OEM PTU, and newer NGK wires--maybe 6mths old.

5). Motor health

Compression test done roughly 1k miles ago. Compression results were 1) 195 2) 180 3) 180 4) 195 dry iirc.

6). Basic throttle body adjustments.

Cleaned and rebuilt the throttle body 6 months ago. Throttle position sensor was readjusted per the service manual. I do notice that my TPS no longer has it's full range: it's missing the top few percent of it's range, topping out at 98% and under 5v as reported by ECMLink. Since it was close, I re-scaled it. BISS adjustment done per service manual within the last few months.

7). Compression ratio

Factory built engine with 155k miles.

8). Wiring and sensors

I've inspected the majority of the sensor wiring in the engine bay in the last few months, both visually and testing per the service manual, and notice no abnormalities.

9). No DTC/CEL codes

Car was throwing an EGR flow code (if I'm remembering correctly). Around the same time, I found that my vacuum canister was leaking significant amounts of boost out of one of it's fittings, so I deleted all the emissions stuff by capping off the nipples on the throttle body. EGR CELs are currently disabled, but the phantom knock problems predate capping the nipples and the flow code by many years.

10). Electrical system

New aftermarket 90amp Canadian alternator, and the battery is less than a year old. ECMLink reports an average of around 14.0v at idle/light cruise. Voltage drop tests have been done on the alternator wiring per the manual and they are within spec.

11). Base fuel pressure and injector values


Fuel pressure has not been checked. Factory pump, factory injectors.

12). Properly calibrated and configured wideband sensor

No wideband. NTK narrowband o2 sensor in stock location.

13). Type of fuel

93 octane.

14). ECMlink how-to videos
Watched.

Background: I've had what I suspected was phantom knock on my '96 for many years. The car had insanely bad lifter tick when I bought it, so about 12 years ago I upgraded to 3rd revision lifters. Even after, the car would randomly seem way down on power, and I suspected that the ECU was pulling timing. As confirmation, I've recently purchased ECMLink v3, and have noticed the CEL indicating the ECU is retarding timing at light cruise conditions, particularly after pulling away if the car has been sitting at a light or something for a few minutes. I've frequently noticed it after accelerating after a turn, and I've noticed it come on before going over rough roads/train tracks.

Car has a 14b intercooled by the stock side-mount, boost controlled by the factory BCS. Car has a significant exhaust leak in the stock piping from where the flex section meets the downpipe. I do hear a metal on metal tapping, almost like a stuck lifter when the car is revved to about 2k coming from the head. It isn't noticeable at idle. Maybe one of the lifters is failing?

Conditions of the log: I took a short drive with the car to run an errand. The car sat in the parking lot for less than five minutes while I was inside. I started the log, drove through the parking lot, and took a turn out onto the road. As I accelerated mildly (I doubt I was making much boost, if any), I noticed the CEL come on indicating the ECU was pulling more than 5deg of timing. I tried to keep steady throttle as I accelerated; the knock continued until I let off when I was at speed.

Ambient air temperature is around 75deg this morning; and the car is at sea level.

Can you guys offer me some direction on this? I have racing gas available and I was thinking about running a few gallons to rule out lean knock once and for all, but it seems unlikely at light cruise.

Any ideas?
 

Attachments

  • PhantomKnock.elg
    80 KB · Views: 37
Last edited:
Here's something I noticed from the log: the highest counts are RawKnock4--about 3x the nearest count at it's peak. So whatever is happening, it's happening near the Cylinder 4 firing event, right? Wouldn't phantom knock be more consistent over the entire RawKnock range?

Doesn't the EGR flow into cylinder 4?
 
Yes it does if it works.

If it was stuck open, that would cause cylinder 4 to run lean and knock, wouldn't it? Looks like I got some direction on how to attack this problem...

edit: and another semi-related question: how does a narrowband o2 sensor respond to WOT? Presumably, it shouldn't continue to cycle when the computer is ignoring its input at WOT, because fuel trims shouldn't be moving around stoichiometric in open loop. I know a narrowband is woefully inadequate to tune the engine, but if I'm seeing it peg at 0.90v, does that suggest I'm running rich at WOT open loop, or is this normal since the factory fuel maps are so much richer than stoich at WOT?
 
Last edited:
The narrowband will show rich, because the engine does go that rich after 75% throttle, and the ecu ignores the O2 at that point, thats why you smell the rotten egg smell when someone floors it when passing you, you know they went 75% or beyond on the throttle and the ecu is going full rich just dumping fuel in with no trim to it and its burning raw fuel in the converter which causes the converter to overheat and stink like rotten eggs.
 
Here's another question, one that I'm probably not going to like the answer to: if I ground the timing connector via ECMLink the engine bounces between 6.0deg and 7.0deg. That should be 5.0deg, shouldn't it?

Since 2gs have non-adjustable timing, that means my crank timing mark is off--perhaps 1 tooth retarded? I checked the cam timing marks recently and they're correct.
 
It would be off by more than 1 degree if the cams were out of sync with the crank, I wouldnt even worry about 1 degree, your light could be off by that much or your not physically able to look straight on at the crank pulley vs timing marks and its actually on.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top