From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Ping Neon John re MAF sensor.
Date: Tue, 07 Jun 2005 23:01:46 -0400
Message-ID: <ucnca1d6dqcbr1luabgqhafj5va2k8n0qn@4ax.com>

On Tue, 07 Jun 2005 19:58:56 -0400, bill horne <redydog@rye.net>
wrote:

>How does a person such as myself test a MAF sensor? Mine was recently
>replaced, and I seriously doubt there was anything wrong with the old
>one. I have my old one - at least, I think I do. It could be someone
>else's that was actually bad.
>
>Vehicle is a 2003 GMC Safari.

There actually isn't an easy way to test these off the vehicle.  One
mostly has to rely on the diagnostics.

A MAF usually doesn't fail completely.  The little heated thermister
that detects the air flow gets coated with crud and becomes
insensitive.  The output signal at low air flow gets noisy and causes
the short term trim to vary all over the place.  The OBD-2 system has
several MAF-specific diagnostics.  That is, codes that point
specifically to the MAF.

The usual problem is that the engine becomes a bit worn so that there
is some back flow in the crankcase vent.  This carries oil mist to the
MAF.  It coats the thermister.  Upon power up the MAF does a burn-off
cycle in which the sensor heats to red heat to burn off deposits.
This works well over time but in the long term, additives and ash from
the oil build up on the thermister.

To test it for this kind of problem, one would have to power up the
MAF, mount it in a system that could pass a known steady flow of air
through it and then monitor the output for stability.

This is one of the few cases where the best diagnostic procedure is to
replace the unit with a known good one after the PCM points to the
sensor.

Did your mechanic tell you what code(s) the PCM stored?

John

From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Ping Neon John re MAF sensor.
Date: Wed, 08 Jun 2005 12:04:57 -0400
Message-ID: <ga4ea1dtlh33gggas9opgcjv6cv923bjbl@4ax.com>

On Wed, 08 Jun 2005 04:18:17 -0400, bill horne <redydog@rye.net>
wrote:

>Neon John wrote:
>
>> On Tue, 07 Jun 2005 19:58:56 -0400, bill horne <redydog@rye.net>
>> wrote:
>>
>>
>>>How does a person such as myself test a MAF sensor? Mine was recently
>>>replaced, and I seriously doubt there was anything wrong with the old
>>>one. I have my old one - at least, I think I do. It could be someone
>>>else's that was actually bad.
>>>
>>>Vehicle is a 2003 GMC Safari.
>>
>>
>> There actually isn't an easy way to test these off the vehicle.  One
>> mostly has to rely on the diagnostics.
>>
>> A MAF usually doesn't fail completely.  The little heated thermister
>> that detects the air flow gets coated with crud and becomes
>> insensitive.  The output signal at low air flow gets noisy and causes
>> the short term trim to vary all over the place.  The OBD-2 system has
>> several MAF-specific diagnostics.  That is, codes that point
>> specifically to the MAF.
>>
>> The usual problem is that the engine becomes a bit worn so that there
>> is some back flow in the crankcase vent.  This carries oil mist to the
>> MAF.  It coats the thermister.  Upon power up the MAF does a burn-off
>> cycle in which the sensor heats to red heat to burn off deposits.
>> This works well over time but in the long term, additives and ash from
>> the oil build up on the thermister.
>
>Well, this engine had only 37K on it at the time, was running fine,
>the MIL has never come on, mpg hadn't changed, and 5000 miles earlier
>contained no codes. However, the honeycomb is slightly sooty (dry
>soot), but the 3 little things that look like silver mini-diodes are
>shiny.

That's normal.

>
>> To test it for this kind of problem, one would have to power up the
>> MAF, mount it in a system that could pass a known steady flow of air
>> through it and then monitor the output for stability.
>>
>> This is one of the few cases where the best diagnostic procedure is to
>> replace the unit with a known good one after the PCM points to the
>> sensor.
>>
>> Did your mechanic tell you what code(s) the PCM stored?
>
>No. I think I'll put the old one back in and see what happens. How
>long should I wait for a code if the thing runs fine?

It could take several days to run all the non-continuous tests but
more likely it would code in a few hours.

Did I see you mention in another post that your engine just quit?  A
bad MAF will NOT cause that.  If the MAF fails in a GM product the PCM
falls back to speed-density (metering based on the manifold pressure,
engine speed and sometimes throttle position.)  The engine won't idle
very well but it will run.

I'm betting on a loose connection that the mechanic "fixed" in the
process of replacing the MAF.  That's the usual cause of sudden engine
stops.  Reinstalling the old part is a good way of testing.

John

From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Ping Neon John re MAF sensor.
Date: Wed, 08 Jun 2005 16:01:23 -0400
Message-ID: <92jea1dqbv42bv1q06moq968v4ueonqkhi@4ax.com>

On Wed, 08 Jun 2005 12:49:18 -0400, bill horne <redydog@rye.net>
wrote:

>Neon John wrote:

>
>And idling fine. I had just pulled away from a stoplight, cruised at
>about 30 until I let off the gas when traffic in front of me slowed.
>When I again stepped on the gas, nothing happened.
>
>> I'm betting on a loose connection that the mechanic "fixed" in the
>> process of replacing the MAF.  That's the usual cause of sudden engine
>> stops.  Reinstalling the old part is a good way of testing.
>>
>> John
>
>The mech's first thought was "fuel pump", but he discovered a little
>later that there was also no ignition. This eventually led to his
>finding the broken wire (inside the insulation). I don't know what
>led him to replace the MAF sensor - unless it's just "policy" to
>replace one of the most expensive and most easily replaceable sensors
>on the vehicle.
>
>One other thing - two days prior, the fuel gauge got stuck on full.
>It was this, I think, that led them to first suspect the fuel pump or
>its circuit. But the fix also fixed the gauge.

All of this points to a bad ground.  Among under-the-hood problems on
GM vehicles (which are few in numbers), bad grounds seem to be
over-represented.  Too many wire loops under the single point ground
bolt, IMHO.

The first prime clue is that the engine stopped when you touched the
accelerator and the engine torqued up against the engine mounts.  I've
seen more than one instance where the ground was made up through the
exhaust system or something touching the engine that broke when the
engine torqued.  Touch the gas and the engine dies.

The second prime clue is the gas gauge behavior.

The third clue is that there was both no spark and no ignition - both
are controlled by the PCM.  A bad ground that powers down the PCM
would stop both.

Do you recall what color wire it was that he repaired?  GM is pretty
good about wire color uniformity in later model vehicles.  Black is
ground, along with black with a yellow tracer.  Red is primary
ignition power while pink is switched 12 volts for accessories.  The
PCM generally has its own separate feed from the fuse/relay block and
that is almost always red.

I have little idea why he changed the MAF other than the obvious -
ripping off an out-of-towner.  I suppose that it is possible that as
the power fluctuated to the PCM, that it might have stored some
spurious codes.  OBD-2 isn't supposed to do that but....  He'd have
had to have been a parts changer who didn't realize that the PCM needs
resetting after such troubles.

I bet that old MAF is as good as new.

John