Fun with MAF's :-(

budwich

Well-Known Member
My 2008 4.2 has been good but fuel mileage has been reducing. I have change out the front o2 sensor (would have done the back but its tough to get at the harness... maybe next time... probably doesn't matter in terms of fueling but it does throw a "weird code" often). I decided to give a new "aftermarket" MAF (ebray... he ha) after cleaning of the existing one didn't do much. It looked ok anyways.
Plug in the replacement, fired up torque, started the vehicle and things (trims, air volume, etc). So OK. But the next drive out, was monitoring things and see the LTFT start to "climb" negative instead of sitting around 0 (old unit and initial run with new). The air intake "volume" is running similar to the old unit at around 6gm/s (I think that the units). Of course, eventually the LTFT for a code of running to rich. I measured the IAT side of the sensor to compare with the old sensor... they both ran about 6M ohm. I then ran a measurement on the ground the signal pin pairs. The old unit is around 2kohms while the new one is around 1kohms. Maybe a design.... potentially doesn't matter since the signal is "pulse based". Further, a diode test of the same pin set give .9 v for new versus 1.9 for the old.

The thing that is unusual is that torque isn't registering air intake flows all over the place like a bad sensor issue..... so then why is the LTFT running away negatively? Any wisdom with a MAF?
 

MRRSM

Lifetime VIP Supporter
Even though you don't mention getting a P0171 Code... there is some additional information on the O2 Sensors and Scanner Diagnostics that could prove useful here:


 
OP
OP
B

budwich

Well-Known Member
there is no codes normally. the one code that I get on occasion is the code for the second o2 sensor (post cat) which give 013b (strange but true). When the new MAF is in, eventually the rich code is set as the LTFT goes to -25 (and over) for a period and thus the system can no longer compensate, throwing the code.

I guess my question is if the o2 were so out of whack, why with the old MAF in does the system run "normally" (ie. trims are in normal ranges, along with air intake)?

still further, if the intake air "volume" is the basically the same with either sensor, where is the air going that engine isn't getting it?
 

Mooseman

Moderator
Was the replacement MAF an actual branded one or just some sort of CCC? What about the O2 sensor? This engine is sensitive to the parts used on it and ACDelco or Delphi may be required.

And the post cat sensor just reports if the cat is working correctly so no point in replacing it for engine running issues, unless you just want to get rid of that weird code.
 
OP
OP
B

budwich

Well-Known Member
I cleaned the throttle body... it didn't make much / any difference. Basically, at idle, I am seeing -15-20% lean on the LTFT which if left idling eventually moves to -25 which will code with a rich running code. If I push on the throttle, the LTFT goes "instantly" down (up) to around -2-3... it seems cracking the throttle body causes a significant change of 10-15% which is strange for "long term"... STFT are oscillating as expected. Idle is good at 600 (610-600). Overall running seems ok. O2 front sensor appears to oscillate fine as expected... with slow oscillation at idle and "choppy" (ie. higher frequency at speed). Post cat is relatively flat but does move / drift "plateaus" depending on throttle / speed. The front sensor was a acdelco (at least that was the box). It was replaced last spring.

The replacement is supposed to a "new" but was unmarked so it may be the issue. I am still not quite understand how it functions though since the intake air volume seems the same / similar to the old. So how can it be that the thing is giving the pcm the incoming air volume so relatively stable yet the "mixture" have such drastic changes. To me this has got to be some sort of "table fubar" whereby the "running table" has significant "differences" in some areas that it is indexing (by what) yet the "compute table" for display is saying something different.

I have tried the 3 minutes of idle / 1 minute off type cycling that the GM SM says can be used instead of the scantool reset. The idle does eventually adjust to "normal" (600ish).

Tomorrow, I will probably drop back the old unit and see what happens.

One last question, the "cells" being used by the pcm for various settings are adjusted at idle to get an appropriate idle speed. How are the "running" cells updated or for that matter, erased to start over from "factory" or whatever. I see that the LTFT value is a "average" of the cells in the "range" that is being used.... so to me that "range" has some pretty high values... but how did they get there and where is "there"?
 
Last edited:

MRRSM

Lifetime VIP Supporter
Please do not take offense to this clarifying explanation...

The MAF does not actually alert the PCM as to what the "Volume" of the Air moving through is at any given moment. Rather... it WEIGHS the passing Air using Fractions of Grams to advise the PCM as to How Much Fuel to Deliver based upon its MASS (Hence the MAF Acronym... versus using the Acronym of VAF...for VOLUME AIR FLOW if that were the case for using Air Volume in its F/A Calculus.

The reason for this distinction is that you can have the same VOLUME of any given kind of matter ...and both will be capable of having a greater or lesser amount of MASS. For example... One Cubic Foot of Marsh-Mellows weighs substantially less than One Cubic Foot of Gold. However...their VOLUME within the 12"X12"X12" Cubicle will naturally be... Identical. Thus the MAF's ability to make these precise "scale weight" type of assessments of the Air Mass versus the Engine RPM and other parameters must be quite precise for the PCM to determine whether or not it is achieving a stable, Stoichiometric balance of 14.7 to 1 Air to Fuel Combustion ratios.

The other reason for using the MAF has to do with the fact that not all vehicles have NA engines that breath in available Air...at Sea Level. Consequently, it is the MAF that helps the PCM make variable adjustments of Air Weight for the normal running of the Motor if say... that vehicle traveled from the Pacific Shore all the way up to Pike's Peak where the MASS of the Air would be much "thinner" and so Weigh less. Volumetric Measurements would only work in either one place at one Altitude or another, but not be capable of making these distinct adjustment in the absence of making constant reads of ambient Temperature and Air Pressure for its F/A Calculations.

And so the Quality and Precision capabilities for NON-OEM Delphi or NON-OEM ACDelco Mass Air Flow Sensor will come into question right along with the Quality of the Lead, Upstream O2 Sensor that may not be Kosher. One thing is certain... The only way to conduct a valid scientific solution in any such investigation... is to Only Change One Variable (or Sensor) at a Time. Working with more than One Variable in this case can turn out to be One Too Many.

Take a very close look at the Electrical Harness of the MAF and the IAT Combo for any loose Connector Insert Wires that may be separating when the engine reaches a certain operating temperature or sustained engine vibrations that might be making that MAF act in sketchy ways that cannot be revealed by static probing and testing with the engine OFF.


 
OP
OP
B

budwich

Well-Known Member
No problem. I agree that the quality of precise of the unit is a question. I guess that I am wrestling with the "precision" at a given "instance". It appears that its "precision" is very bad such that the "value" it sends is 15-20% off... and then slightly cracking the throttle body appears to get it back to perhaps a "somewhat normal" response... maybe.

Of course, as you indicate, it is truly hard to compare things in time as changing / switching out the MAF takes time along with potential that monitor process (torque) has its own "degree of variance" with what can be monitored / displayed at any given instance.

As you indicate / suggest, wiring could be a culprit... I have tried the "wiggle / pull" test at the connector and wire route along the way while the engine is running and monitoring torque.... this is relatively easy to do since the issue (high negative values of LTFT) occur at idle ... thus I don't have climb on the engine while it goes down the road... :smile:

I haven't tried a "frequency check" of the output yet to see if there is an indication of poor correlation (I don't actually know what I would be looking for) as I don't like "pin pricking" leads on signal wires especially thin ones as this can lead to further problems later in "life". I am concerned that the meter checks of the MAF side of things gives electrical parameters that are different when compared to the old device.... but again, not sure if they matter at the operational level if the value of interest to the pcm is something like frequency "density" (ie pwm ish) as opposed to voltage level / resistance variance.

The "fun" will continue for a bit before I decide on what to do with the ebray purchase in terms of return. I am concern that the unit has no product stamps codes which leaves one open to the seller's thoroughness in item quality and "cross-reference"... of course, you get what you pay for... :smile:
 
OP
OP
B

budwich

Well-Known Member
fun continues.... So not much has changed overnight in terms of the operation / response of the LTFT.... I was lightly hoping that the "good mechanic witch" would happen by and magically adjust / fix things. Guess I didn't leave enough money under the vehicle. :smile:

Anyway, go out my "pin probe test lead" as recommended by here a while back.... and tapped into the signal lead. Not sure how responsive my meter is to frequencies especially one that is likely a always positive waveform (no zero crossing) along with not "phase locked" (ie. changing). So the reading at idle was bouncing around 55-65/70 ... could be the meter trying to figure out the waveform. Running the rpms up into the low 1K range or there abouts (helper's foot might be tough to control).... the frequency went into the 110-115/120 range. Not sure what the actual range is for this device. Further, light reading seem to indicate the actual values don't matter just whether is moves up "in step" with the throttle..... BUT comparing with the old, which when put back in, the engine returns to "normal reactions" on the LTFT. The equivalent frequencies are 45-50 at idle and at mid 1k 80-90. It would appear perhaps that frequencies do matter... although my results may be taken with a "grain of salt / accuracy" as the meter isn't very good in this mode of signal.

The other fun now is I lost one of the screws holding the maf in the engine area somewhere... :-( no big deal... except for "its sunday".

Guess it is time to go back to the seller and do that hastle.

Comments?
 

MRRSM

Lifetime VIP Supporter
As for the Lost Fastener over The Engine Bay... Fortunately... unless you imparted any push one way or another when you dropped it... Gravity is your friend... and a Slender, Telescoping Pocket Magnet will soon reveal that it is probably closer to you than you think. Poking around directly under the MAF/IAT and around on the frame areas and wheel well locations will usually reveal where it fell... and you will very likely find OTHER Lost Fasteners as well.

Avoid using the Telescoping Magnets having the Neodymium inserts the size of a Buffalo Nickel... They will stick to the closest thing made of any Ferro-Metal and are more of a nuisance than an assist.

If you look at the images of the MAP Sensor P0102 and P0102 Codes... even though these have not been thrown yet... their diagnosis will give clues about the expected, nominal performance of the MAF and some possible Causes and Origins of problems similar to what you are dealing with.

That second link to the scannerdanner.com Thread I posted above has what is probably a GOOD WAVE FORM for your comparison... but that fellow is using a much more sophisticated scanner.

You might drop in there and request a Good Wave Form for your analysis, as he keeps a whole library of these available. Read his opening Web Page admonition on "How to Post" first ...as the Members there can be reluctant to jump in and help anyone in the absence of following their "format" for inquiry and problem solving.

I have attached the screen prints of the relevant MAF Code diagnostic break downs that may prove helpful:

MAFP0101.jpgMAFP0101A.jpgMAFP0102.jpgMAFP0102A.jpg
 
Last edited:
OP
OP
B

budwich

Well-Known Member
continue to look around in this area... density versus volume. It would appear that the relationship is somewhat "linear" but in the "inverse" for a given altitude. Basically, as density goes down, volume goes up at the same rate... for an ideal gas. Probably more to it than that, but it would seem to make since that incoming air volume on a vehicle would be hard to ascertain but the density maybe less so and hence why they monitor and use that value. Of course, the "final say" is the O2 sensors that feedback so that adjustment can be made. A 10% error in density will likely cause a 10% air in the volume determination which then would throw the "gross running mixture" out by at least that much. Doesn't mean much other than if the "non-oem" unit has little or poor "certification" to density calibration, the engine / system will be running with one hand tied behind its back. Having said that, is making a MAF sensor really rocket science! :smile: I guess so.
:-( Exchanging emails with seller, they indicate their "remanufactured units" are tested but to what, not sure. Further, a visual inspection of the visible components in the air stream does show quite a different in size and placement which may or may not impact accuracy. In addition, there are some small holes in the chrome plate of the main body of the unit... on the replacement unit, they aren't there... not sure if they are important or are just "inspection / test point access"... but a noticeable difference.... which was in the listing picture but on the unit sent.
Still working with seller on next steps.
 
Last edited:

Online statistics

Members online
13
Guests online
187
Total visitors
200

Staff online

Forum statistics

Threads
19,075
Messages
574,461
Answered questions
1
Members
11,780
Latest member
MILES-2005-4.2
Top Bottom