[GeekyRob]

I'm at my wit's end on this one, and from reading some of these threads you guys really seem to know what you're talking about.

My 2000 plain-no-frills-out-of-the-box Focus went from a great car to a lemon in a very short time. Since the whole car is computer controlled, and trouble codes can only tell you so much, I got a PCM interface and software to talk to the PCM from my laptop. I can't find a button that says "Skip all this technical stuff and just tell me what's wrong with my car" so I'll have to do it the hard way and think a little. Bummer.

The problem I'm experiencing is very bad performance on acceleration. By bad performance I mean that power goes to almost zero, the car stutters, bogs down, jolts and starts throwing up cylinder misfire trouble codes. Unfortunately, it's not any single cylinder, and it's even tossed up a P0300 (Random misfire) a time or two.

I plugged in my computer and immediately saw the problem. Ok, I didn't see the problem, but I saw the manifestations of the problem. If I saw the problem, I'd have fixed it already. Sitting in the driveway, my TPS reads 20.4%. WOT reads 93.3%. This seems a little weird, so I plugged the computer into my wife's truck for comparison and her TPS shows 0% when you're not pushing the pedal and 100% at WOT. I immediately suspected by TPS, and being only $20 I replaced it. No change in performance.

I took it out on the road and recorded all parameters under various conditions. Gentle acceleration actually works, but when I start to press the pedal a little I immediately notice the timing advance drop from 20-30 down to around 5-10. This explains the very poor acceleration, but since the ignition timing is controlled by the PCM there's nothing to adjust. Either the PCM is bad or (more likely) it's getting a bad input from something and is trying to compensate.

The diagnostics also shows crazy numbers for my engine load and short term fuel trim. Engine load is supposedly calculated based on air flow divided by peak air flow, but there's a reasonable correlation between throttle position and air flow readings. If I press the accelerator, air flow increases. This is what I would expect since I'm opening the butterfly on the throttle body and letting more air in.

I tested the voltages on the TPS both at the sensor and at the PCM. At idle (ignition on, engine off) I get 1.00 volts and at WOT I get 4.77 volts. Reference voltage is 4.97 volts. Hmmm. 1.00 / 4.97 is 20.3%, which is what my TPS reads at idle. 4.77 . 4.97 = 93.3%, which is what my TPS reads at WOT. It looks like the PCM is showing absolute values instead of adjusting for actual range.

Finally, here's my question. Does anyone know what inputs the PCM uses to calculate timing advance? I'm suspecting my MAF sensor, but it has a reasonable correlation with RPM. Not as good as I'd expect, but nothing crazy that would indicate a really bad sensor, and certainly nothing that would cause a 50% drop in timing advance at 30% throttle (adjusted, so that's really 30% throttle).

Thanks in advance for your help, and sorry about the long post. There's just so much info to consider, it's hard to ask a question without giving background. I still think I'm leaving out a bunch of info. I'll check for replies after church.

Thanks!

 

[03OrangeSVT]

1 volt at idle at better than 4.5 at WOT is just about where they should be.

If I remember right timing is run off of load and rpm, the MAF is there to provide input on how much fuel the engine requires.

There is a sensor on the front of the motor, drivers side down close to the oil pan... its the only one on the front of the block... make sure its plugged in.. it "reads" the flywheel and sends the signal to the ECU so it can calculate when to fire off the coil.

 

[GeekyRob]

I'm slowly coming to accept the fact that the problem is NOT with my TPS. I was leaning that direction because it way up at the top of the engine and really easy to get to... I'll check the flywheel position sensor this evening when I get home, thanks for the advice. That's a good next place to check.

The problem with using engine load to calculate the timing advance is that engine load is itself a calculated value. I've read that it's calculated based on the MAF, but that doesn't sound right. I took it for an extended test drive with the monitors hooked up and recording, and there seems to be a fairly firm correlation between engine speed and air flow. My fuel trim tends to stay very rich (like, around +30%) and at about 15%-20% throttle the timing advance is abruptly cut in half. I disconnected the knock sensor for giggles since feedback from that will cause the PCM to adjust the timing. Bad idea, I plugged that back in before I even moved the car. Wow, I guess I kind of needed that.

So, considering that the timing advance was based on engine vacuum and RPM in the good-ol-days of vacuum advance and little weights that would swing out in the distributor to advance the timing, does anyone know what the PCM uses to calculate engine load now-a-days? My engine load is showing up really high (like ~30% at idle) and seems to correlate closely with my throttle position.

Unless anyone else has a potential next step, I think the next thing to check is for one or more bad injectors. I'll set up a test rig to apply a known pressure to the injectors and measure the volume of gas delivered over a known interval. It will also show me if one is leaking, or if a few are clogged. That would account for my rediculously rich short term fuel trim numbers, and I'm not sure how the PCM would try to compensate for that.

This is really difficult with the PCM being a 'black box'. It would help to know how the thing is programmed, but I'm sure that's a closely guarded secret.

Thanks!

 

[djmontara]

well your rich fuel trim to me points to either vacuum leaks, leaky fuel injectors, duff O2 sensor or poor fuel delivery.. have u checked for these things?

 

[Grumpy]

I doubt that the ECM's basic ignition timing program has been toasted. The retarding of the ignition timing that you saw when you opened the throttle (higher engine load), I believe, is normal. Did it advance again when you backed off the throttle, or depending on the RPM, when the RPM increased? I'm old enough to have driven cars with (and dealt with) the old centrifugal flyweight/vacuum distributors. The advance curve on the old cars was that by the time the engine RPM was at some mid-range RPM the flyweights had put the ignition timing to max. advance, but that was with high vacuum (low load).......as soon as you opened the throttle and the vacuum decreased (higher load) the spring in the vacuum pot pulled back (retarded) the timing. High engine loads (larges throttle openings = low vacuum = high manifold pressure = high compression pressures) makes the engine prone to detonation if the ignition timing is too far advanced. Conversely it is true that late timing can cause poor performance like you stated. I can't speak directly to the Focus but my Nissan which has an ignition system similar to most cars of this era inc. your Focus, uses the following sensors for ignition control: MAF, crankshaft pos. sensor(2) camshaft pos. sensor, TPS, coolant temp., vehicle speed sensor and a knock sensor (not normally needed). The TDC ref. for no.1 is the camshaft pos. sensor......not sure about your Focus. Remember that the real "biggy" sensor that is really driving the fuel delivery and control system (which inc. the ignition), is the MAF.

 

[GeekyRob]

That's good insight, Grumpy, Thanks! I hadn't considered that decreased timing advance might actually be normal under high loads, since the low advance corresponds closely to very poor engine performance. I'll have to consider that maybe the major symptom I'm looking at is just incidental to another problem. I agree with djmontara that my next step is to look at the fuel delivery system. I've checked everything I can find for a vacuum leak and can't find one. I put on a vacuum gauge and the numbers look really good. I pulled and cleaned each of the injectors by sending carb cleaner (mostly just acetone, MEK and methylene chloride) through under low pressure, and all the brown fuel residue is now gone. There was no change after doing that, so I'm thinking that might not be the problem. Still, a volumetric throughput measurement under normal operating pressure will provide positive assurance that the injectors are operating normally.

The MAF was my next target, but everything I've seen tells me it's ok. That doesn't mean it is, though. Functioning and functioning properly are two different things. The PCM is seeing a good flow profile from the MAF, based on driving the car and recording the reported mass air flow from the computer.

Here's a tidbit I haven't shared yet. About a month prior to this happening, I had a voltage leak on my coil pack. My Door Ajar light would come on for no reason driving down the road. The car would periodically 'spit' and at the same time my speedometer would jump. It turns out that the coil pack was leaking voltage, and the surge of current into the engine block when cylinder #2 fired caused weird issues all over the place. A new coil pack and new wires fixed the problem. I've been suspecting a ground fault ever since, since Ford uses ground switching all over the place. Any disruption in the ground, or a floating current through one section of the ground would cause a ground loop just like in a nice car stereo system. I've been going all over and cleaning every ground connection I can find. Any time a wire is screwed to the frame it gets sanded, cleaned and bolted down nice and solid. I don't think I've found them all yet. Does anyone think this could be contributing to the problem?

 

[malim]

How are your plugs ??? When was the last time you changed them ?
Start with the basics. Don't jump straight on the sensors.

Check your wires, spark plugs, coil pack. See if the wires are loose on the coil pack.

 

[Grumpy]

Yeah! If one were to look at only your third paragraph one might think that it was a case of simple fuel starvation causing a lean miixture during acceleration. It usually takes a really rich mixture to cause an engine to "bog". For acceleration the system should go rich. Some engines when you go with WOT (high loads) the system goes into open loop and goes rich to protect the engine (detonation again). I'm assuming the fuel trim figures were at steady state?

 

[Dan-O]

+30% fuel trim means the O2 sensor is reading WICKED lean exhaust and it's adding fuel.
What's your fuel pressure while driving steady and under acceleration?
I like to look at long term and not short term when diagnosing driveability issues but +30% is still a good indicator that something is up.
Short term is too rapidly changing.

I agree with Grumpy, sounds like a fuel starvation issue.

A misfire will also read lean because of the extra oxygen being read by the O2 sensor which has an end result of high fuel trims like you're getting.

Also like what's already been said, Let's keep this simple first, then later start blaming the electronics.
As far as the PCM is concerned Garbage in = Garbage out
Your engine needs air fuel and spark.
Basics first, that's the way I start.

 

[Grumpy]

A real basic question......when was the last time the fuel filter was replaced? Plus, wasn't the fuel pump suspect & didn't it have an associated SB for the 2000 Focus? As Dan-O suggested, it could be helpful to carry out a fuel pressure check, which will also prove out the fuel pressure sensor operation. Maybe this poor thing has been trying to "add" fuel that it can't get.

 

[03OrangeSVT]

Misfires cause RICH conditions not lean since your sending unburnt, raw fuel down the exhaust pipe.

Asfor the camshaft sensor... its really not needed, you can unplug it and the car will keep on running fine... not even a hiccup. We did that to my car as an experiment... ran it for over a month without being hooked up... didnt even effect the power output on the dyno.

If your datalog is showing its adding 30% fuel then I have to agree that you have either a dirty MAF sensor, plugged fuel filter, vaccuum leak or something along those lines.

 

[Dan-O]

Misfires cause RICH conditions not lean since your sending unburnt, raw fuel down the exhaust pipe.

Well....
You'd think that, but the oxygen sensor measures oxygen not fuel.
If you don't burn the oxygen, it's still left in the exhaust and the sensor will read that there's extra oxygen left over because it didn't inject enough fuel to burn it all.
You get high (+) fuel trims because the PCM is reading lean and fattening up the mixture.

 

[GeekyRob]

Sorry, all, I was trying to keep the post short so I left out a ton of details. It's always been kind of cool the way you guys can come up with good suggestions based on very limited information. I'll back up a bit, but some of this may repeat some info I've posted above. Sorry about that, but hopefully I can put it into context.

Several months ago my Focus started to misbehave on the open road. First the Door Ajar light started coming on, then the engine would occasionally miss and the speedometer would jump. All I had to work with was a simple OBD II code scanner, and since the car wouldn't set a trouble code I was kind of out of luck. I discovered that I could cause the problem by placing the engine under load and allowing it to miss. It missed enough that it finally set the MIL. The trouble code came back as P0302, which my trusty Chilton's told me was a cylinder #2 misfire.

The plugs looked good to me but I changed them anyway. I drove it some more and the next time the MIL came on it was cylinder 4 that was misfiring. After that it went back to cylinder 2 again. The car would keep setting trouble codes and I'd keep clearing them. I spent hours pouring over the wiring schematics trying to find a connection between the dashboard lights and the engine, and it started to look more and more like a ground fault. A ground fault on one piece of the car would cause the current from that section to flow to an adjacent section seeking a ground. In doing so, it would alter all the associated reference voltages and may even activate sensors from time to time. I tore into the car and very carefully cleaned and secured any ground connections I could find. Nothing worked, and although I'm pretty familiar now with how my car is wired it was otherwise a wasted effort. I ran across a forum somewhere that had someone with a similar problem and the issue turned out to be the coil pack and plug wires. Each time a spark was generated, some of it would 'leak' directly to ground instead of going through the plug (and across the gap). The result of an unrestricted voltage pulse into the engine block was a subsequent voltage spike throughout the entire car. The problem manifested itself in the instrument cluster. The misfire was caused by insufficient voltage at the plug. I replaced the coil pack and plug wires and my car was happy once again. I cleared the settings on the PCM by disconnecting the battery for a long time, and the car worked great for a few months.

Recently, the misfire has returned. This time, the trouble codes point to no cylinder in particular. Since I just finished replacing most of the ignition system, my attention turned to the fuel. I replaced the fuel filter, which was dirty, but it didn't help. I figured the only way to actually know what was happening in a computer controlled car was to hook it up to a computer. I ordered an interface from obddiagnostics.com. It works great, and I have continuous logging of some very good information while driving and while the problem is occurring. The data pointed me towards the TPS, because it was reading absolute position instead of relative position like it does on my wife's Chevrolet. I've recently run a baseline on another properly functioning Ford (not a focus) and it seems that the absolute TPS reading is a feature of the Ford PCM. I relpaced the TPS anyway, because it's cheap and easy to reach, and nothing changed. I've also cleaned the injectors by running carb cleaner through them under low pressure to disolve the gasoline residue, cleaned the MAF sensor and I've checked all the vacuum hoses I could find. I hooked up a vacuum gauge to the intake manifold, and the pressure ranges from about 10" Hg to about 28" Hg absolute under acceleration, which is consistent with what I expected to see. The only really weird reading is the timing advance. It's weird because there's a step change when the accelerator pedal is depressed and you begin to accelerate. This is true for leaving a red light as well as accelerating from 50 to 55 on the freeway. If you press the pedal just a bit, you can get fairly good performance. However, just a touch too much and the timing advance cuts in half and you start missing. The diagnostic interface I have also measures (calculates?) fuel economy, and if I run that portion of the program when the missing occurs the fuel economy plummets. This is implying that there is WAY too much fuel in the cylinder(s) for some reason. The PCM is doing the best it can with the information it can, so I guess I should be impressed that the car even runs, but somewhere it's getting a bad data feed. Unfortunately, I can only work on it on the weekends because I've grown rather fond of eating and living indoors, and that requires that I actually attend my regular job. As it stands, my next target is the sensors that tell the PCM where the engine is in it's combustion cycle (camshaft and crankshaft position sensors). Aside from that, I'm facing the very real possibility that I can't fix this and I'm going to have to take it to an (ugh) dealership to get it looked at. The problem is made very difficult because the PCM is a black box, and I can't tell what's actually going on inside. I'd like to be able to see the actual data the PCM sees, but there are 60 wires on that thing (I counted) and they're none too easy to get at while driving.

So, to sum up:

malim - plugs, wires and coil pack have all been replaced within the past few months. All seem to be connected firmly, and the random nature of the misfire leads me towards the fuel system instead of the ignition system. If it were a wire or plug it would always fail on the same cylinder.

Grumpy - the data logger tracks open/closed loop status and I haven't seen it go back to open loop unless it does it quickly between polling. Unfortunately, I don't know what the long term trim figures were, since I haven't been at steady state since I got the data logger and the mixtures have been skewed for quite some time now. The fuel filter was replaces within the last month. I hadn't considered testing the fuel pressure apart from the fuel pressure sensors to see if they're reading fuel when there isn't any.

Dan-O - fuel pressure varies from 34.8 psig to 47 psig, with the average being 39.46 psig.

03OrangeSVT - I cleaned the MAF sensor and it appears to be operating properly but it could certainly be the culprit if it's not reading accurately.

Thanks for all your help. I can't wait to find out what the problem really is. I'll post the answer when I get it. In the mean time, please send any other suggestions you have my way, I can use all the help I can get. If anyone is interested, I'll post the OBD log files somewhere so you can look at the actual data I'm getting.

 

[Dan-O]

What's the pressure under load or when you're experiencing the misfire issue?

I'm not liking the 34 PSI if that's under load.
PM me your VIN, I'll check Oasis to see if you fall under the fuel pump service issue.

Oh yeah, are we talking about a Zetec or SPI motor?
SPI's had sticking exhaust valves.

 

[aironjack]

I have read the whole thread. I would encourage you to real test the fuel pressure in the system, I think somebody already said it and I agree this could give you some clue on what's going on.

Apart from that, I think we all would like to see what the ECU is doing at all time, this would help a lot troubleshooting problems.

Good luck!

 

[Grumpy]

34 psi is the idle/no load pressure? How good is the spark? A big, fat one that makes a good "crack" or a thin, weak one. A spark that is weak when you test it, can cause a miss, particularly under high loads.......ie. like you've got under acceleration. High compression pressures can kill a weak spark. Your "vacuum" numbers threw me until I realized they had to be MP figures. Do you, or does anyone know if there is a condensor in the coil primary circuit of the Focus?

 

[allanzx3]

One thing i didnt see anyone suggest is some of the easiest things to do. Do you have a friend with a focus as well? Swap the MAF, or the ecu and go for a rip. Keep swaping parts one by one until you narrow down the issue.

Having said that it does sound like a fuel issue. When was the fuel filter replaced last?

 

[djmontara]

okay heres a question... what does your data logger say ur ignition advance is at idle? and at WOT?

 

[Dan-O]

Well, did you dig into anything?

 

[1turbofocus]

Everyone is making this harder then it needs to be , If he is at 30% on the o2 there is a bad o2 or excessive fueling going on and that I doupt

I have said this a thousand times and no one listens the knock sensor does more then knock it also does inefficiant combustion and if he is dumping 30% more fuel in then needed then this will cause the knock sensor to pull timing as well , hence the poor power , low timing

Check your o2 sensor it may be bad , how old is it , how long has it been sence you cleaned the MAF ETC

My cell is 704-648-6944 if you would like to go over things

Tom