Possible MAF issue w/IATS Reading High

[wvicary]

2005 V6 Cayenne. Outside temp -10C.

Key on,engine off- IAT +36 C

KOEO- IAT +28 C

Engine RPM@3000- IAT +18 C

MAF sensor is new and engine was at coolant temp of +85 C. Sorry I don't know what the conversion to F is.

[wvicary]

How many miles/drive cycles do you have on the vehicle since all of this swapping of parts and such? Maybe it just needs a chance to relearn. Put 20 start/stop drive cycles on it without clearing any code or changing anything and monitor how it operates.

[logray]

Excellent, thank you again. About 50 miles on the rebuilt engine. About 20 miles since the last battery reset and parts swap.

I plan to drive it heavily over the next several days, and continue monitoring. Probably at least 200 miles, but I have the first oil change coming up in about 150 miles.

One interesting thing I found is the idle fuel trims keep pulling back, now at -0.07 IIRC. And the long term trims are going up a little. Over 1 now by maybe a few hundredths.

Edited February 10, 2012 by logray

[Ahsai]

So the natural question is do those COLD air intakes help? Would be nice if someone who has one could provide some log data.

[JFP in PA]

So the natural question is do those COLD air intakes help? Would be nice if someone who has one could provide some log data.

From a stoichiometry as well as a volumetric efficiency perspective, the cooler and more consistent the incoming air charge, the better off the power output should be. But this runs into several large caveats along the way, such as where is the intake charge temp measured, how much of the actual engine intake runner system is allowed to heat soak, and how fast the intake charge is moving. As there is a huge mass of heat soaked intake system, as well at the throttle body itself due to its location, quite often these ultra slick looking cold air systems add absolutely nothing to the car other than more noise.

[logray]

Having owned an EVOMS WAI, I can certainly say I miss the noise.

[logray]

Back to MAF troubleshooting again. Still original MAF (78k miles). I already returned the new MAF since it sounds like the high IAT is perfectly normal for other cars (even though whne you read the tech manuals they say a coolant temp over 90C and IAT over 30C will cause a lean condition).

After the DME reset 200 miles ago things are slowly going downhill. At cold startup and idling the engine runs very smooth, since it's getting what seems to be the right amount of fuel. At first after the reset the engine was running great, warm idle was very smooth and power was great. While the car still pulls harder than I'm used to with the increased displacement, it seems to be slighly down on power in the mid-high RPM range, and at warm/hot idle it's back to rumbly and stumbly rough running again, almost like fuel delivery is wrong.

I bet if I reset the DME again it would run fine for a while.

I pulled out the trusty fuel pressure tester, hooked it up to the fuel rail and got a solid 3.3 bars at idle and 3.8 bars engine off/fuel pump on.

Did some more MAF testing. Since I had already completely tested the wiring I did some more data logging.

The more I drive it, the more the TRA (short term idle fuel trims) seem seem to keep pulling back to negative/rich, about -0.1 beyond spec at idle, currently reading -0.19 on BOTH banks. The set point at idle is 0.00 +/- 0.1 for TRA. But it's not throwing a code, not sure with the upper end is to throw a code.

The FRA (long term mid/high range fuel trims) are leaning towards lean, FRA of 1.05/1.06 for bank 1 and 1.06/1.07 for bank 2. The set points for bank 1 is 1.02 +/- 0.04, and 1.03 +/- 0.05 for bank 2. So still within spec, but just barely.

I've been doing some reading about MAF failure affecting fuel trims. In my reading, when a MAF over estimates air flow, it will cause short term trims to go negative and long term to positive.

"When a MAF sensor gets contaminated, it skews the transfer function such that the sensor over-estimates air flow at idle (causes the fuel system to go rich) and under-estimates air flow at high air flows (causes fuel system to go lean). This means Long Term Fuel Trims will learn lean (negative) corrections at idle and learn rich (positive) corrections at higher air flows. Short term fuel trims will run even more lean (negative) at idle as well as they are the quick-acting fuel trims so to speak."

Here's where it gets strange...

The MAF, at no load idle is pretty much steady at 17-19 kg/hr, 1.35-1.45 volts.

At no load 3000 rpm it's pretty stead between 65-75kg/hr.

Both of which are pretty much on spec.

So, could it be that the MAF is bad after all, despite these numbers, because of what's happening to the STFT and LTFT? And it is misreporting the air flow, even though it thinks it's seeing the right amount of air, perhaps it is actually mis reporting it, which explains the poor running?

edited for accuracy

Edited February 14, 2012 by logray

[logray]

I reset the fuel trims back to default 0.98 LTFT and 0.00 STFT. In other words, no longer lean long term (requiring more fuel/positive) or rich short term (negative short term), but instead using the default fuel maps.

I unplugged the MAF before even starting it.

Then drove the car around for a while, and it seemed to run smoother and have similar power to when I reset the computer earlier, but maybe not at it's best when it had time to adjust the map somewhat.

Warm/hot idle seems to have dramatically improved as well, rough running values went back to normal and it's not stumbling at all and having trouble holding onto 680 rpm idle with high rough running values constantly spiking. Now it sounds like it's getting the fuel it needs. I hope all this rich or lean driving I've been doing during break in hasn't hampered the ring seating process.

Bad news is, I need to buy another MAF now. :( Good news is, I think it might be a bad MAF. :)

edited for accuracy

Edited February 14, 2012 by logray

[Ahsai]

...it seems to be slighly down on power in the mid-high RPM range, and at warm/hot idle it's back to rumbly and stumbly rough running again, like it's running lean.

Um...from the -0.19 TRA, your engine is actually running RICH at idle (DME is removing fuel, more than spec, trying to correct tthe RICH situation). That is also consistent with the MAF failure description you quoted.

[logray]

Yes, thanks I had it backwards in the posts above and edited for accuracy.

Edited February 14, 2012 by logray

[logray]

New MAF arrived and installed.

About 40 miles on the car with a full DME reset.

First started out and LTFT were are 0.98 and STFT at 0.00

After some miles, the STFT went to 0.02 and LTFT to 1.02.

After more miles, the STFT flipped to negative and the LTFT positive.

By the end of the 40 miles, the STFT went to -0.08 and the LTFT to 1.06/1.07.

The car still seems to be performing well, although possibly not as good as it did when I reset the DME again and it was using the default fuel maps for a while, I need to drive it more to find out for certain.

Could this just be due to the fact that the stock map thinks this should be a 3.4L and because it is now 3.6L, requiring more fuel, that it's sensing it's running leaner than the stock fuel map in the mid/long term range, and therefore dumping more fuel (positive LTFT) and then at idle/short term pulling it back to compensate?

It's too soon to tell though because running without the MAF and after a DME reset the car ran great too.

I'm wondering if there is something else going on like too much fuel vapor being sucked into the intake skewing the fuel trims or perhaps a back pressure issue?

The engine is basically stock at this point, aside from upgraded 3.6L cylinders and pistons, fabspeed headers and stock 3.6L mufflers. Stock intake, air filter, etc. etc. etc.

All new Bosch 02 sensors were installed a few thousand miles ago, and the 02 graphs look spot on. I'll post some more data later on.

Fuel injectors were cleaned and tested, and all performing well by the tester and report.

[logray]

After letting it sit for a while, then starting it again and driving it around, the car is once again back to stumbling on idle, nearly to the point of stalling and performance is down.

Rough running is all over the place.

FRA is 1.07 and TRA is -0.08 for both banks.

I'm 100% positive now that if I reset the DME and fuel trims back to default it will once again run like a pro and not stumble at idle. But that doesn't seem to be a reasonable way to operate the car every 30 miles, or to drive around without a MAF.

So, since I replaced the MAF, tested the MAF wiring, replaced the O2 sensors, what's next?

I'm going to try to read about the EVAP system and see if perhaps the intake is somehow sucking excess vapor in, causing the fuel trims to skew out of place so rapidly. Does anyone have any good links on this one? I'll search around some. I'm wondering if I can just temporarily clamp off the evap purge valve at the intake.

Here's some O2 data, looks normal to me.

(click to enlarge)

Edited February 15, 2012 by logray

[dcdrechsel]

I am sure you have thought of this but................Jake Raby probably knows the answer or has seen this before .Perhaps you could contract for an hour of his time ?

[wvicary]

You can test the EVAP flow control solenoid, it should be closed under internal spring pressure and opened via DME. I have had them leak under vacuum on VW.

[logray]

I clamped the evap vacuum line going into the throttle body and unplugged the control solneoid. DTC for unplugging the EVAP solenoid, but don't care.

Reset the fuel trims to default again, 30 miles of driving.

FRA/FRA2 now at 1.06/1.04

TRA/TRA2 now at -0.05

I'll have to drive it more tomorrow to see if it keeps going downhill again. That's about all I can do today.

It's very annoying because it seems to really do much better idling at about 0.00 or 0.02 TRA when it first starts out.

Power delivery seems better when it's closer to 1.02.

IAT stayed more around 30-35C, but it's a cool day only about 48F while driving around.

[logray]

I am sure you have thought of this but................Jake Raby probably knows the answer or has seen this before. Perhaps you could contract for an hour of his time?

Thanks, I posted over on his forum and offered to pay if he's willing to help and knows the answer.

[dcdrechsel]

This is going to sound stupid but.............the MAF is unidirectional could it be backwards ?

[logray]

It can only go in one way, the screw holes don't line up otherwise.

[logray]

I'm going to do some more testing today like eliminating (plugging) more vacuum consumers such as brake booster and SAI.

I'm wondering if intake vacuum is insufficient causing a fuel pressure regulator problem. I will test intake vacuum as well.

I also forgot to mention this issue was present before the new cylinders and pistons.

Before the rebuild, FRA was 1.04 (normal) and TRA -0.16 (abnormal)

(injectors had been professionally cleaned and still produced these numbers)

Perhaps I should buy the newer white injectors which are two generations newer than what I am running which are the original blue colored injectors off my 1999 M96-01. I had the injectors professionally cleaned and the results after cleaning were 61 ml / 36 ms and 38 ml / 3 ms.

Of course, even after the cleaning 61 ml/36 ms works out to 131.76 cc/min. Maybe that is part of my problem, since it's still way below even original bosch flow specs? Something doesn't line up there though with the injector cleaning report, if they were only providing nearly half the fuel they are supposed to something would be seriously wrong. I doubt the TRA or FRA would be close to spec. Here are the injector specs:

0280150455 (orig bosch blue p/n)

21 lbs/hr

220.7 cc/min

0280156053 (latest bosch white p/n)

22.75 lbs/hr

239.1 cc/min

A scan of the injector cleaning report here:

http://img851.imageshack.us/img851/3015/drinjector.jpg

[wvicary]

When it's hot and running poorly can you capture primary and secondary O2 sensor data. Not sure if this is the case with your DME but sometimes the secondary O2 data is used to aid in long term fuel trim calculations with some manufacturers.

[logray]

Unfortunately Durametric cannot monitor post cat sensors on 5.2.2 DMEs. I think ME 5.2.2 only uses them to monitor cat efficiency, not affect fuel trims, but I could be mistaken.

I might be onto something with the fuel injectors being inadequate however.

Would it make sense that the long term trims would go more positive (add more fuel) because the stock 3.4L injectors cannot dump enough fuel for 3.6L displacement. If the O2 sensors are reading long term is lean, and more fuel is required this might support the theory. Then at idle or short term, it keeps pulling fuel back into negative territory thinking it's running rich (which it probably is since it has to open the injectors longer for the larger displacement).

[dcdrechsel]

Another idea........Prodriver built a motor very similiar to yours .In fact I think he started with a Boxster S .Probably wouldn't hurt to compare notes on what injectors he used and if he has seen any related issues with his build -specifically did the DME compensate automatically .

What is curious is that the default maps seem to work fine and it's only in "real time" that the issues show up .This would suggest that the injectors might be ok .

I'll try not to ask any more stupid questions

[logray]

Good idea.

I think bypro built the longblock, I'm not sure it has been fired up yet (still in longblock form). I think I remember asking or he stated that he generally cleans the injectors during rebuilds.

Yes this has me somewhat concerned too, and am still questioning the injectors. That why does it seem to run fine at default and then go downhill from there. I'm wondering if even though it has good power and idles very well perhaps it's not actually running fine with the default map (O2 sensors sense it is running lean and start to dump fuel). So to start out it runs lean, and seemingly well... then as the O2 sensors see the stock injectors aren't injecting enough fuel for the larger dispclament it starts enrichening the mixture to achieve the proper stoichiometric mixture. And maybe it's just borderline not keeping up, or maybe if I drove it long enough it would throw a code.

Also, if it was doing this before the rebuild (to a lesser degree, not as high long term but just as low short term) then perhaps it is something else.

I'm going to monitor the injection time and see if it is drastically higher than the 3ms it should be at hot idle.

Edited February 16, 2012 by logray

[dcdrechsel]

One more......since this appears to be a preexisting condition -could the DME code be a suspect ?If the default maps work and it shows up in processing real data that might be a clue .

[dcdrechsel]

I was typing and didn't see your last response .ProDriver has A Boxster S that he built an engine for .The motor for sale is a second one .