I'm getting the BOB hooked up this weekend. One word of warning for other users of this service tool, in line with hanky's words of wisdom about checking connections: out of curiosity, I opened the BOB's junction box (housing the 104 pin male connector and from where the two wire looms originate), and found one internal wire ribbon connector unplugged from its pin header on the PCB ... If I were to assume that the BOB was guaranteed good I would have probably missed a few (otherwise good) signals and drove myself crazy. Bottom line: don't take stuff for granted ... ever.

Cheers,
cougar_fan

 

Well, I connected the BOB ... it was an eye opener, but not the way I expected it. Details below ...

ALL O2 sensors are working fine, their output voltages swing between 0 and about 0.95V as expected. This made me wonder why the ScanGauge was showing one sensor so much lower than the other. Before blaming the PCM for misinterpreting the correct signals from O2 sensors, I thought I'd check the ScanGauge programming.

It turned out that I had a misplaced "1" in the MTH definition of the XGauge for the left O2 sensor. MTH is the scaling factor used by ScanGauge2 to format the value read from the PCM ... I had it 10 times smaller than the other, so, obviously, the O2 reading was always under 10...

Having said that, I swapped in an exact replacement PCM (identical model and flash codes) and the 95 PCM. Both yielded the exact behavior as the original PCM - no improvement of the rough running. The injector signals are identical (as averaged by the digital meter, at least), no wiring problems (I pulled every single wire loom in the engine compartment), the ECT and IAT work fine, the TPS works fine - all verified at the BOB.

At this point I feel like I should push the Cougar off a cliff ... The only thing left - I can think of, at least - is a vacuum leak, although this doesn't explain why I'm getting that great fuel economy. The way I understand it, if I step on the gas, the PCM should increase fuel dispensing according to what the MAF and TPS see. Residual air will make the mix leaner, but without cutting on fuel instant consumption ...

Please let me know what you think, guys. No idea is too wild at this point...

Thank you!

Cheers,
cougar_fan

 

Oh, and one more thing: two days ago I configured the ScanGauge to display the fuel trims (1 and 2, both long and short). It turns out that all four are negative (long trims go around -8 .. -9, short trims are around -2 .. -3) when the engine is acting up. If I step on the gas, the short trims turn positive for a little while, but then go back to negative as above.

While running in open loop (cold engine), the short trims are around + 50 ... +60, and the engine purrs (although even then it still hiccups every two or three seconds, randomly)

I know greasemark advised me against messing up with sensor signals ... but I wonder, what if I were to simulate the MAF output with a potentiometer, and tell the PCM that the engine gets more air than it actually does? This should make it dispense out more fuel, which would take away the lean condition that apparently is causing the superb fuel economy I'm getting without me ever asking for it ?

Dunno ... please let me know what you think about this. Thanks again!

Cheers,
cougar_fan

 

Here's where I am now ... maybe someone would pick up a problem with my logic:

- the engine runs lean - the actual fuel consumption is a lot less than it used to be. One tank now lasts for over 200 miles, where it used to last about 140-150 when things were running fine. Once the engine enters closed loop operation, all fuel trims are negative in steady mode. Short trims turn slightly positive (less than 10) during acceleration, but then go back to negative.

- the engine hesitates at any speed: idle, full throttle, anywhere in between. If it were a vacuum leak, it should have only hesitate when the throttle was closed or just slightly open - please confirm ...

- air and coolant temperatures are reported fine. I disconnected the ECT and, two seconds later, the fan went straight on high speed, confirming that the PCM wasn't happy with what it saw as a grossly overheated engine. With ECT connected, the radiator fan kicks in on low speed when PCM reports 98 degrees Celsius, and turns itself off when the coolant reached about 93 degrees Celsius.

- no codes are put out by the PCM

- the engine runs better (but still with hesitations) after a KAM clear or PCM swap, but it gets back to rough running after about 10-15 minutes (when, I assume, the PCM learned the engine).

- the MAF output is about 1.25 V when idle, and goes up as I step on the gas. I can post actual values (MAF, TPS and RPM readings at different engine speeds) - I wish I had a cross-reference chart for this ...

Could it be a MAF gone bad but still alive (the engine ran better with the MAF disconnected - I will try that again to confirm), combined with a vacuum leak that's causing the lighter, but still present roughness during limp-mode operation (no MAF)?

Thanks again!

Cheers,
Dan

 

If I can, may I ask you for the latest fuel trims? Maybe we can see something that is being overlooked. Thanks

 

Sure, hanky:

- cold engine - the first minute or so in the morning: LF: +1..5, SF: +50..65 (almost smooth running). I can check to see if that high SF coincides with open loop operation, although I'm pretty sure it does.

- after that: LF: -4..+2, SF: -5(steady 55 mph drive on horizontal road)..+4(hard acceleration). When both trims are on zero or slightly above, the engine is running pretty smoothly, however this only happens for 10-15 seconds, then the roughness returns and both trims go on or below zero for anywhere between a few minutes and half hour. Most of the times, the engine wold not return to smooth running unless I let it cool for an hour or more, but even when it comes back to its senses, it is only for a few seconds. As a rule, the more I drove the car that day, the worse the hesitations.

This afternoon, for the first time in the past two months or so, the engine ran fine (for about 10 minutes) after just a couple of hours break.

I also noticed that the roughness is a bit less acute since the weather turned colder here n Thunder Bay. Dunno ...

The reported fuel economy (as per Scangauge) varies wildly: for steady 50 mph highway driving, when the engine runs well I get 8-9 l/100km, but when it is acting up badly, the scanner reports 3-4 l/100 km. What I'm used to see from the golden days is 9 l / 100 km ...

Thanks again!

Cheers,
Dan

 

Past advice from greasemark might go well here. It might be a good idea to not get too involved with other miscellaneous numbers and just concern yourself with the fuel trim values and concentrate on what is going on there. I will try to shed some light on the possible reasons for unusual fuel trim values .Have to get ready for work. Will try to help.

 

Thank you kindly, hanky!

When I started the car this morning (IAT = zero degrees Celsius), it ran slightly rough from the get-go. I watched it for a minute or two.. LF1 stayed at +5 while the car was idling. The short trim (SF1) started at +55, and went down to +8 when the water temperature reached 33 degrees Celsius, open loop operation.

I took the video below today on my way to work, after driving the car for about 20 minutes (hot engine). It was chilly outside (2-3 degrees Celsius), and the engine ran significantly better than what I've been seeing for the past two (summer) months:

http://www.youtube.com/watch?v=tKJZ0...der&playnext=1

Please let me know if you'd like to see other inputs while I drive the car, I can configure the scanner to show O2 outputs, all four fuel trims, TPS, RPM ... any combination of four gauges. I can also put a meter on the BOB's MAF signal line, and record that side-by-side with the fuel trims ... just let me know what would you like to see.

Again, thank you for your help with this!

Cheers,
Dan

 

Your video is cool.
I do have some questions though,,,,,,,,,
your video is showing bank 1 which is the RH bank and the fuel trim appears normal.
what about bank 2's fuel trim that is the LH bank , cyls 5-8. Do you have any trim readings for that bank?
Also , when you disconnected the ECT the engine temp reading should show (minus)- 40 degrees not an overheated engine. Can you check that again?

 

Here are three videos I took this morning.

Cold engine (outside temperature was -2 degrees Celsius, there were frozen droplets of water on the car):
Cold start - YouTube

Fully warmed up engine (after about 25 minutes engine-on time):
Warmed up engine - YouTube

Second part of the previous movie. Car parked, scanner showing SF trims along with exhaust manifold O2 sensors (before the cats):
Fuel trims anf oxygen sensors - YouTube


There are two sensors ( I know of, at least ...), both located on the front side of the intake manifold, behind and on either side of the alternator. The RH one feeds into the temperature gauge in the cluster (single wire sensor). I unplugged it and the cluster water temperature indication went to zero (cold). The other one, on the driver side, looks similar but has two wires. When I disconnected it, the radiator fan kicked in on high speed (after a 2-3 seconds lag) .. so I ASSUMED that the PCM thought the engine was overheated. I didn't actually read the CWT parameter with the scanner, but will do that today and post tonight.

Thanks again!

Cheers,
Dan