LTFT should be +/- 5% ideally and certainly under +/- 10%, and should be similar for both banks under all driving conditions. LTFT and STFT on each bank at any instant should add up to roughly zero. LTFT values change slowly and STFT values jump around pretty fast. Depending on PCM calibration, LTFT over +20% or so will set a 171/174 code and over -20% or so will set a 172/175 code. Over +/-25 to 30% will set O2 stuck lean or stuck rich codes.

 

Ok. So should I press record for the data, then turn the car on, let it idle for a minute, then drive around a bit?

 

Correct. Then playback the data, as long as you know when it's idling and when it's not idling.

 

I'll tell it to record speed and rpm as well.

 

There ya go! You can almost tell when you are idling based on the trims.

 

You have to have the key on for it to connect and start recording. So you can turn the key to on without starting, hit record, then start it.

The short term fuel trims will be constantly moving up and down to generate the O2 sensor lean/rich switching. The long term trims will be more stable, but will change with different operating conditions, like stepping on the gas.

Like Jeff said, ideally you want to see ±5% trims. To get the total fuel trim, you have to add the short term number to the long term number. So if the long term is +7% and the short term is -3%, the total fuel trim is +4%. If the short trim numbers are moving around a lot try to average them.

 

I already told you this in chat.... but I'll write it here! Thanks to behlinla for some of this help and info!!

First, watch trims at idle - you're looking for them to be between -5% to +5% just like Jeff said above. You are looking that they don't go outside of that much during any driving.

If you see the LTFT much higher than that, then you have a probable vacuum leak. If it's super high (+10 to +25%), especially on both banks, then it's most likely a vacuum leak!

Of course record all that and both banks....

Okay, to test the O2 sensors functioning correctly do this test (thanks behlinla):
Find an empty stretch of road, or somewhere you can do this.... idle for 60 seconds and drive normally around 40mph. We're watching that the O2 sensors are switching correctly. They should be moving from around .1v to around .8v at least every 2 seconds, ideally at least ever second.

Next stop the car, then go ahead and go WOT all the way through first gear. The trims may drop out to very low numbers for a second, but all 4 O2 sensors should jump up to .9v or so the entire time through WOT. This means that the correct amount of fuel is being dumped into the motor at WOT. The duratec really likes to dump a lot of fuel at WOT. This also shows the O2 sensors are working correctly, the switching is really what you need, but this shows how much fuel you're burning by doing this

 

I will do this tomorrow morning, thanks to Sam, behlinla, and Jeff! Good info here

 

What are fuel trims all about?

generally, if there below double digits there in the ball park.
When you see over 10% trims, or one bank radically different from the
other, its time to investigate further.

 

Got the spreadsheet. Original was messy, but here's one that I cleaned up, and its much easier to read. Has all relevant categories.

https://skydrive.live.com/redir?resid=35734B8B157017A5!112&authkey=!AAbhVP9pLNytu70

 

Okay, I took some time to look at it.

My first question is how old are the O2 sensors? Original? My second question is what is your fuel economy? What percent city/highway?

Your idle trims look perfect. The trims are all within ±4%. No vacuum leaks! It looks like this wasn't a cold start (168° ECT), so I can't tell you how well the O2 heaters are working. The only thing I didn't like is the Bank 2 Sensor 2 pegging at zero until it warmed up.

The cruse speed data looks okay except for the accelerations. It looks like you were accelerating moderately hard during driving. I don't like how that fuel trim is spiking up under acceleration and staying there for some time. It looks like the engine is really dumping fuel. This is even more apparent at WOT. The mixture goes rich as it should, but the fuel trim is going up to +9. When you took your foot off the gas afterwards, the O2 sensor took almost 13 seconds to switch once, then it took another 14 seconds or so to get back to normal switching operation. This tells me the enrichment is excessive and it seems like you would really be wasting fuel.

I'm not sure what to make of the enrichment yet. It could just be PCM programming, but you might have some kind of leak upstream of the TB like a loose clamp or cracked rubber intake tube. Or it could be something with the MAF.

So if you have time, can you do a few more tests for me?

I would like a WOT run with MAF data. Only select RPM, MAF, MPH, and B1S1. Start at warm idle, do a WOT 0-60+, then take your foot off the gas and just coast back down to about 40 MPH before stopping. In park, let it idle a minute then do some throttle snaps. It's important to go from foot off the gas to pedal to the floor as fast as you can. You can let off when the engine revs up to 3-4k. Do this about 5 times in faster succession. That should be all for that trace.

I would also like a single O2 sensor voltage trace to see if they're switching properly. Deselect everything except the B1S1 voltage PID. In the corner you should see an update rate of about 50-60 PIDs/sec. If it's slower than that, you need to disable the fuel economy logging (which selects PIDs in the background and slows it down). Once you get a good refresh rate, then start the collection. Let it just idle for 2 minutes. Then do a few throttle snaps as above. Finally moderately accelerate (less than 1/2 throttle) to a normal cruise speed (1500-2000 RPM) and hold it steady for an additional two minutes.

 

In my observation I've noticed that these cars really like to dump way more fuel than they should at heavy throttle, especially at WOT.

It shouldn't take nearly that long to recover. My Tec' recovers back to switching O2 sensors after a second or two when letting off from WOT.

I do agree that the trims at idle look spot on!

The B1S2 sensor never starts switching again after WOT. It stays at +.8 or so until the recording stops.

 

It shouldn't take nearly that long to recover. My Tec' recovers back to switching O2 sensors after a second or two when letting off from WOT.

That's what I would think is normal. During the time those upstream O2 sensors are stuck rich, the PCM is essentially "blind" to the actual mixture (it can't tell the difference between a little bit rich and a lot rich). When that happens the fuel economy usually suffers until the O2 sensors start switching again. So the question is why is it going that rich? It could be either from a fuel/air metering problem, or due to a slow response from the O2 sensors. If the O2 sensors don't respond fast enough, the mixture will keep going rich and when the sensor finally starts responding the fuel trim will have overshot.

 

Old sensors will take more time to react, because their membrane is almost completely clogged. They are still "good", but they waste gas at accelerations.

 

First off, thank you guys for taking the time to review. I can do those extra tests tomorrow, behlinla. And I usually get between 18-21 mpg mostly town/city driving, maybe 15% highway. I guess I may have a heavy foot but that was pretty much normal accelerating for me. All four O2 sensors are OE to my knowledge. So that would make them 8-9 years old now.

I can also do a cold start tomorrow, what data would you like on that?

Sent from my iPhone using AutoGuide.com Free App

 

For cold start, do RPM, all four O2 sensors, ECT, and one bank of fuel trim data so I can see when it goes into closed loop. You can do two fuel trim banks if you want. If you can, start it and let it idle for 4-5 minutes until it comes all the way down to warm idle speed.

 

^^^^ Yes, Sam, the PCM will set a slow switch code if switch rate gets too low, but the PCM programming looks at this from an emissions standpoint, not a MPG standpoint. If you sensors are original at over 100K miles, I would replace both upstream sensors. Dont mess with the downstream unless they give you trouble. Downstreams last much longer than the upstreams and have zero effect on fuel economy.

 

Yes, you will have MPG effects long before the PCM will set a code for a slow O2 sensor. To really evaluate the O2 sensors you need high res data (fast scan tool or oscilloscope) to look at how fast the O2 sensors cross threshold. But it's a good idea to change the sensors as PM at about 100k.

 

Then I'm sure I need to do them. I see nothing about chaining O2 sensors in my maintenance records... How hard is it to get to them and get them out? Just curious. I know they run about $50 a piece. Specific brands that are good?

 

Not sure how difficult they are on the Duratec. The front one is usually pretty easy to get to, but you might have to take the cowl off to get at the rear one. The hard part is breaking them loose from the manifold. You will need a special O2 sensor socket or use a 7/8" flare nut wrench or crowfoot on them. Spray them with penetrant several times in the days before the replacement. As for brands, stick with Motorcraft, Bosch, or NTK. On the replacement sensors use a very small amount (rice grain) of anti-sieze and do not get any on the sensor casing (also avoid touching the end of the sensor if possible).

 

If they have never been out in 10 years Plus+, be prepared for
a cussing session even with the right tools. The sensor on top of
the exhaust pipe below the trannie hump is especially a challenge.
I have broke off more than one on the various bulls i have had.

 

Oh great.... Then I'm thinking about having my mechanic do it. He gives reallllllllyyyy cheap labor if I give him the parts, like $50/hr. Not bad at all. This will be a project in January. Maybe my gas milage will go up haha!

Sorry about hijacking the thread Ben

Also, Ben's car only has 44k miles IIRC. Do you think his O2 sensors have an issue with the slow switching?

 

It's about to hit 57k. They're also going on 9 years old too..

 

Here you go Lee, cold start

https://skydrive.live.com/redir?resid=35734B8B157017A5!117&authkey=!AIxDbet6ku8nLho

 

Took 40 seconds to go into closed loop and then 4.5minutes to settle down. A tad long for closed loop but if it was super cold I can see that.

 

It was like 40 ish degrees out


Sent from my iPhone using AutoGuide.com Free App

 

Data analysis and graphs

Okay Ben, I got them all graphed for you. The warmup time for the front sensors was 36 seconds, which is a little bit longer than I would like but coolant temp did start out at 30°F. Downstream sensors also warmed up just in time, so I think the O2 heaters are not 100% but they are still pretty good. Lag time still seems like it might be a little long, so I'll have to look at that more carefully in the O2 sensor-only trace if you do it.

Idle was nice and stable so I think your IAC is working very well. There was a nice drop in idle to warm idle speed when the ECT got to 51°C. Fuel trims at warm idle also tightened up nicely.

 

So it all looks pretty good then? Still want me to do that WOT 0-60 one?


Sent from my iPhone using AutoGuide.com Free App