Suddenly Running Lean

[Tailwind]

Well, I've got an interesting problem. The car has been running great until this afternoon. It suddenly lost power and while it continued to run, it ran very roughly. When I was able to look at things in TS, it showed that the AFR was over 19 at idle and really didn't change much when the engine was reved. While, I hadn't changed the AFR or VE tables recently, reloaded them, which had no effect.

I did find that the inlet fitting to the fuel rail is leaking, the fuel pressure was a bit over 30 psi, instead of the 45 it should be. When I removed the fitting, the o-ring kind of disintegrated. At least that would explain the leak.

I've ordered parts to replace the fitting, but was wondering if this is likely the source of the problem, or something unrelated.

Thanks for any thoughts.

Bill
Ford EDIS
MS-II, version 3

[kjones6039]

Yikes! Gotta watch out for those fuel leaks!!

but was wondering if this is likely the source of the problem, or something unrelated.

More than likely..... but an msg and log would also be in order here, I think.

Ken

[Tailwind]

Unfortunately, I did not get a log file, and now the fuel line is apart. I'll certainly get a log if the problem persists when I get the line repaired.

Bill

[Tailwind]

The story continues: I got the fuel leak at the rail corrected, but the engine is still running lean. The fuel pressure is actually supposed to be 30 - 40 psi with the engine running and it is. However, the engine still is not running correctly, with a lack of power and AFR of 19 or more.

I did get a log file which is attached, along with the msq file. Still looking for ideas where to look. Thanks, again, for the suggestions.

Bill

[kjones6039]

The fuel pressure is actually supposed to be 30 - 40 psi with the engine running and it is.

It is, what? 30, or 40, or something else? I'm not trying to be a smart a$$, but a 10# differential in fuel pressure can make a profound difference in injector flow.
In your first post you said the fuel pressure was supposed to be 45 psi, which sounds very reasonable to me.

Bottom line.... are you able to accurately measure your present fuel pressure?

Ken

[kjones6039]

What are the specs (ie. rated flow at what pressure) for the injectors you are running?

Ken

[Tailwind]

According to the fuel pressure gauge installed in the fuel line, the pressure is 35 psi. The 30-40 range came from the manual for the engine ( Ford Escort 1.9 L) which provided the parts for the installation. As near as I can tell, this is the way things have always been.

Just to add a bit more, I have discovered that the O2 sensor controller is showing a fault during the warmup phase. The documentation says this is either shorted wiring or shorted sensor. (Like what is left?) I plan to temporarily disable the O2 sensor in the software and see what effect that gas.

Bill

[Tailwind]

This is just to give some "closure" to this thread. I (think I) FINALLY figured out what was causing the engine problems. It turned out to be a loose wire in the plug connecting the coil pack to the EDIS unit. This wire controlled the firing of the plugs in cylinders 1 and 4 and basically they were not working. This explained the rough running and loss of power. I still don't understand why the AFR showed 19+ but when I reseated the wire in it's connector ( and replaced the coil pack) the car ran beautifully again.

Bill

[MonzaRacer]

Because when a cylinder or two or more fail too fire it drives the AFR LEAN due to having NOT used the air for combustion. O2 sensors rear O2 after reaching over 600F DEG when they go active. NOT fuel. See an engine management system is used to READ the air used by the engine, if an engine gets 19lbs /hr per injector then it "KNOWS" X amount of fuel is going in at set vacuum rating, and it "should" be using X amount of air to BURN said fuel, basically if an injector outputs 19lbs of fuel in an hour at set pressure then 14.7 to 1 would mean it needs 279.3 lbs/hr of air.
IF it is idling at 700rpm, has set PSI at 43.5 fuel pressure and opens said injector 1.5 MS then X amount of fuel is injected and requires Y amount of fuel.
1 hour has 3,600,000 milliseconds, SO divide 19(lb/hr) by 3,600,000 and that is amount injected EVERY injector opening.
Now since engine control unit "READS" amount of air ingested, namely by computing from baro sensor reading, temp reading and manifold vacuum reading. It "computes" the amount of air that SHOULD be present in exhaust, thus it is in constant control. Namely by varying the injector pulse width.
First thing in diagnosis, is whats engine doing:
running rough
Why:(things that can be wrong)
spark(none,low,going wrong way like bad wire)
fuel(plugged injector, low fuel pressure)
mechanical(broken valve spring, broken ring, hole in piston, blown head gasket, burnt valve, flat cam lobe)
What to do for each variable. So you checked computer and AFR was off, ding ding ding. There are several simple inductive spark testers, lay up side wire to see if its firing, injector fireing can be checked with noid light, mechanical issues usually can be found by analog vacuum gauge or compression test or leak down test, and fuel pressure by gauge and/or lab scope/amps clamp.

[Tailwind]

Thanks for the explanation. Unfortunately for the diagnostic process, I did everything you suggested EXCEPT using an inductive timing light. That is how I ultimately found the problem. Should have done that first.

Thanks, again

[Bernard Fife]

Now since engine control unit "READS" amount of air ingested, namely by computing from baro sensor reading, temp reading and manifold vacuum reading. It "computes" the amount of air that SHOULD be present in exhaust, thus it is in constant control. Namely by varying the injector pulse width.

This is wrong.

The EGO sensor gives a indication of the incoming air/fuel ratio without an calculations by the ECU whatsoever. EGO sensors can (and are) used on carbureted race cars, for example. What the narrow band EGO sensors does is measure the oxygen level in the exhaust, period. If there is substantial of oxygen in the exhaust, then there wasn't enough fuel to burn the oxygen (in a properly running engine), so the intake mixture was lean. If there is no oxygen in the exhaust, then there was more than enough fuel in the exhaust to use up all the oxygen, so the mixture was rich. See: http://www.megamanual.com/v22manual/mwire.htm#ego

Wide band sensors are more sophisticated, but generally work on the same principal (except the measurement of the oxygen is how much current is required to drive the sensor to a stoichiometric state). See: http://www.megamanual.com/PWC/LSU4.htm for the details.

Luckily, that means an EGO sensor is an *independent* measure of the air/fuel ratio, so it helps keep the engine tuning on-track by not making any assumptions (or calculations) about the engine itself (such as its air flow).

However EGO sensors are not infallible. It is very important that there be no exhaust gas leaks upstream of the oxygen sensor. Certain conditions can draw ambient air *into* the exhaust if there is a leak, causing the MegaSquirt® EFI controller to compensate for an apparently lean condition. A misfire can have a similar effect (while this pumps unburned fuel into the exhaust, it also pumps unburned oxygen into the exhaust, and remember it it the exhaust gas OXYGEN that we are measuring with the sensor). This will falsely create a rich mixture in your system that can be difficult to diagnose. Furthermore, oxygen sensors do not tell you what the optimum AFR is for any engine operating condition; you have to tune to find that. EGO sensors have a finite response time, so they reading isn't instantaneous, and the sensor and the associated input circuit and firmware may introduce a degree of averaging to the reading. And finally, EGO sensors do not work while cold (for cold starts). So they are a useful tool, but they aren't the 'be all and end all' of tuning.

Lance.