JohnCh Posted September 15, 2014 Share Posted September 15, 2014 I need some help diagnosing a sporadic issue with my 2.0L Duratec running an Emerald ECU. The engine runs very lean at light throttle above idle, but is normal at idle and above 1/2 throttle. The issue presented itself after letting the car sit in a parking lot for 15 minutes after 20 miles of driving on a hot day. I initially suspected a heat soak related issue, but when starting the car up the next morning with temps in the mid 50's, the problem was still there. I drove the car for several days in an attempt to get a better handle on the issue, and it miraculously fixed itself. However, after repeating the initial circumstances (~20 miles on a hot day and then letting it sit in the same parking lot for 15 minutes) the problem reappeared. A new TPS did not fix the problem. Temp gun shows similar temps on each primary pipe, and after increasing idle speed to about 2300rpm where the lean condition is very obvious, pulling plug wires in turn had similar results for each cylinder. I haven't done a leak down or compression test yet, but it doesn't seem like I have a bad cylinder. I've also confirmed that the air and water temp sensors that feed the ECU are operating correctly, and have turned off closed loop running, and fuel cutoff on the overrun to no effect. Any ideas? Thanks, John Link to comment Share on other sites More sharing options...
xcarguy Posted September 15, 2014 Share Posted September 15, 2014 John, Long distance stabs here; O2 sensor? MAF sensor? Link to comment Share on other sites More sharing options...
xcarguy Posted September 15, 2014 Share Posted September 15, 2014 Sorry? Pushed the button too soon; loose connections at sensors/ECU? Link to comment Share on other sites More sharing options...
JohnCh Posted September 15, 2014 Author Share Posted September 15, 2014 I pulled the O2 sensor out of the loop when switching to open loop mode, so even if that's bad, it wouldn't affect how the engine runs, and there is no MAF (Direct to Head Throttle Bodies with a Throttle Position Sensor). I just heard back from Emerald and Karl suggested I also log the battery voltage and injection corrections to see if those are okay, as voltage issues can impact injector pulse width. He suspects that the issue could also be a faulty fuel pump or fuel pressure regulator. The latter is pretty easy to check, so I'll add that to the list when I log the items above. Thanks, John Link to comment Share on other sites More sharing options...
BusaNostra Posted September 15, 2014 Share Posted September 15, 2014 How about you AIT (air intake temperature sensor) working properly? If the intake air temperature sensor is not reading accurately, the PCM may think the air is warmer or colder than it actually is, causing it to miscalculate the air/fuel mixture. The result may be a lean or rich fuel mixture that causes driveability symptoms such as poor idle quality when cold, stumble on cold acceleration, and surging when the engine is warm. Link to comment Share on other sites More sharing options...
JohnCh Posted September 15, 2014 Author Share Posted September 15, 2014 That was my initial guess given how this started, but as indicated above, both the air and water temp sensors appear to be working correctly based on data logging (i.e. steady and showing temps that sound reasonable.) Karl at Emerald gave me a few specific channels to log to help trouble shoot. They should show if anything in the ECU chain, including battery voltage and sensors are acting up and causing the fuel injectors to turn down output, or if the problem lies elsewhere in the fueling system. Occam's Razor never seems to be correct when I'm troubleshooting, so I'm assuming this is a case of MADB (Murphy's Awkward Dull Blade: the most convoluted explanation is correct). Thanks, John Link to comment Share on other sites More sharing options...
NVP66S Posted September 15, 2014 Share Posted September 15, 2014 How are you seeing the lean operation? Wideband sensor or roughness followed by pulling a plug? I installed a wideband sensor with A/F readout and it's very interesting. Link to comment Share on other sites More sharing options...
JohnCh Posted September 15, 2014 Author Share Posted September 15, 2014 It certainly drives like a lean condition: roughness and small backfires as the unburnt mixture hits the hot exhaust system. The wideband O2 that feeds the ECU also shows things are lean. Preliminary logging showed around 13:1 at idle trending to 18:1 (where the WBO2's data logging range tops out) at part throttle and returning to 13-14:1 at heavier throttle. I suppose spark could also be the culprit. If the logging shows no timing or injection corrections when the problem occurs, and the fuel pressure regulator checks out, I might swap out the coil pack. -John Link to comment Share on other sites More sharing options...
BusaNostra Posted September 15, 2014 Share Posted September 15, 2014 Since you checked all areas that causes to lean, the only thing I could think now is the dirty injectors that can't deliver their normal dose of fuel. Dirty fuel injectors can't flow as much fuel as clean ones, and they can't deliver the correct spray pattern that is so essential for clean, efficient combustion and the horsepower desired. The fuel feedback control system will compensate for the leaning effect once it is in closed loop, but it can't correct the underlying condition that is causing the problem.... check fuel filter too. What is the fuel pressure @ slight throttle? Yeah 18 is playing fire. I know you have a turbo too....very dangerous. Link to comment Share on other sites More sharing options...
JohnCh Posted September 15, 2014 Author Share Posted September 15, 2014 No turbo. It's a Euro 2.0L Duratec with Kent DTEC10 cams, a tiny bit of head work, Jenvey DTHTBs, lightweight flywheel, 4-2-1 header, Raceco silencer, etc. Wouldn't clogged injectors and/or filter impact running at larger throttle openings/higher rpms where there is a bigger demand for fuel flow? My car runs fine in those situations and at idle, it's just the inbetween that's problematic. Also, there was no change to the running in either closed loop or open loop mode. It really feels like TPS is bad -- but that's been replaced -- or the ECU is applying a correction in those situations. Hopefully more logging will add clarity. Thanks, John Link to comment Share on other sites More sharing options...
xcarguy Posted September 15, 2014 Share Posted September 15, 2014 John, Another long range stab; fuel map? You're going from 13:1 to 18:1 and then back down to 13ish:1 on the big end. Link to comment Share on other sites More sharing options...
JohnCh Posted September 15, 2014 Author Share Posted September 15, 2014 Nope, ruled that out too. I uploaded the original version of that map to my ECU yesterday and the results were the same. -John Link to comment Share on other sites More sharing options...
NVP66S Posted September 16, 2014 Share Posted September 16, 2014 Different ECU manufacturer, but the original default map on my Megasquirt was very rich compared to the Mazda factory ECU. I'm guessing they do that because extra rich doesn't create warranty claims compared to extra lean. Link to comment Share on other sites More sharing options...
JohnCh Posted September 16, 2014 Author Share Posted September 16, 2014 The map I'm referring to isn't the default map that came with the ECU. It's the map I've been running for 4 years that was done on a dyno. I simply uploaded the version stored on my laptop in case the map in the ECU had somehow become corrupted (long shot, but only took a couple of minutes to test). -John Link to comment Share on other sites More sharing options...
xcarguy Posted September 16, 2014 Share Posted September 16, 2014 This one's turning into a head scratcher. :rofl: Do you think it's possible for a sensor (those discussed, or other) to partially fail within the range of 'between idle' and 'WOT'? . . . . Or the ECU for that matter? Sorry . . . best way I know to ask this. Link to comment Share on other sites More sharing options...
JohnCh Posted September 17, 2014 Author Share Posted September 17, 2014 Yep, definitely a head scratcher. But then again, I never do things the easy way If a sensor is failing, it should cause the injector to "correct" which is a loggable channel with the Emerald. I hope to have time this weekend to try that. -John Link to comment Share on other sites More sharing options...
Klasik-69 Posted September 18, 2014 Share Posted September 18, 2014 Have you installed a fuel pressure gauge on the fuel supply side ? It sounds to me like the fuel pressure is drifting down thus causing a diminished fuel supply. On my car (Zetec 2.0L), I installed a fuel pressure gauge as well as a AEM AFR gauge so I could monitor the AFR as I drive since my Pectel ECU doesn't do wide and data logging. Fuel pressure will definitely affect the AFR. Your fuel pump may get weak once warm or hot. Usually fuel pressure regulators work or don't work, independent of temperatures. You should be able to connect your laptop to the ECU while this is occurring and adjust the pulse width of the injectors at the problem rpm range to see if it has any effect on the rough running. One thing that puzzled me is you said the engine is backfiring which mean unburnt fuel and that doesn't translate to a lean condition, unless you meant pre-ignition from an ultra lean condition making the valve glow and ignite fuel ahead of full compression. Could your ignition coils be breaking down with heat ? My prior experience with bad coils, however, is that once they go bad, they stay bad so your situation is weird. A slightly loose crank sensor ??? I feel your pain. Link to comment Share on other sites More sharing options...
pi7ot Posted September 18, 2014 Share Posted September 18, 2014 How about a smoke test to rule out an intake air leak ? m Link to comment Share on other sites More sharing options...
coffee break Posted September 18, 2014 Share Posted September 18, 2014 How about a smoke test to rule out an intake air leak ? m I've seen guys use spray GUMOUT to look for vacuum leaks, but this problem doesn't sound like it is. :svengo: Link to comment Share on other sites More sharing options...
NVP66S Posted September 18, 2014 Share Posted September 18, 2014 How about a smoke test to rule out an intake air leak ? m Hmmm... there would be a different response to a manifold vacuum leak if the ECU used the Mass Air Flow sensor, or if it only used throttle position and manifold pressure. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now