What causes Freq. Spikes in my MAF?
04-10-2007, 05:12 AM
#1
What causes Freq. Spikes in my MAF? PLS look at this graph:
RED = MAF Frequency at idle after cruising or after a warm up (engine warm).
ORANGE = LTFT’s + STFT’s resulting from the RED MAF signals.
BLUE = MAF Frequancy at idle after a spirited driving or in traffic (engine pretty hot).
LIGHT BLUE = LTFT’s + STFT’s resulting from the BLUE MAF signals.
As you can see the hot MAF (or hot wires, or hot PCM…) produces frequencies spikes that influence the fuel trims. Pretty badly.
After a while the spikes disappear and the Trims line becomes flat again.
Is that my MAF?
A connenction?
The PCM?
The MAF has an insulation and doesn’t get hot. I have an extension between the MAF and the original connector.
The MAF is stock and it’s the original piece (1998, 75'000 miles).
Any idea? I see the spikes not only at idle and this messes with my tuning
Thanks
Stefano
RED = MAF Frequency at idle after cruising or after a warm up (engine warm).
ORANGE = LTFT’s + STFT’s resulting from the RED MAF signals.
BLUE = MAF Frequancy at idle after a spirited driving or in traffic (engine pretty hot).
LIGHT BLUE = LTFT’s + STFT’s resulting from the BLUE MAF signals.
As you can see the hot MAF (or hot wires, or hot PCM…) produces frequencies spikes that influence the fuel trims. Pretty badly.
After a while the spikes disappear and the Trims line becomes flat again.
Is that my MAF?
A connenction?
The PCM?
The MAF has an insulation and doesn’t get hot. I have an extension between the MAF and the original connector.
The MAF is stock and it’s the original piece (1998, 75'000 miles).
Any idea? I see the spikes not only at idle and this messes with my tuning
Thanks
Stefano
04-10-2007, 10:31 AM
#2
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
I think the spikes may be symptomatic, but not
responsible fot the LTFT drift; they are short-time
and the trims drift long after, and sometimes
without any preceding MAF-blip.
I believe that Dynamic Airflow and Dynamic Cylinder
Air are what are truly used for fueling, and these
should be smoothed of any such spikes. How do you
see these looking, compared to the raw value?
responsible fot the LTFT drift; they are short-time
and the trims drift long after, and sometimes
without any preceding MAF-blip.
I believe that Dynamic Airflow and Dynamic Cylinder
Air are what are truly used for fueling, and these
should be smoothed of any such spikes. How do you
see these looking, compared to the raw value?
04-10-2007, 10:45 AM
#3
ah... I use Autotap and see only MAF raw frequency and MAF G/sec.
G/Sec follows the frequency pretty well (same trend, no delay).
OK, in my graph you don't see a direct relationship between frequency and trims, but in general there is one.
OK, let me disconnect the MAF and see what happens. The SD tune shows almost the same trims than using the MAF. Let's see if then the trims are more constant.
What I don't understand (and this happens since years) is why as soon the engine gets hot the trims start to make weird things (mostly go negative by -5 / -7%). As said: hot means stuck in traffic or after driving with fun
If it was a bad MAF cable I would see no signal, or a lower signal... but there are spikes!
A real weird idea: with the hot engine there is a certain amount of backfire that tricks the MAF and the trims
The MAF isn't directional, so a reverse flow is seen as "correct flow".
I hope this remains a weird idea... but it's the only thing I can imagine in case it's a mechanical issue (and not electronic).
G/Sec follows the frequency pretty well (same trend, no delay).
OK, in my graph you don't see a direct relationship between frequency and trims, but in general there is one.
OK, let me disconnect the MAF and see what happens. The SD tune shows almost the same trims than using the MAF. Let's see if then the trims are more constant.
What I don't understand (and this happens since years) is why as soon the engine gets hot the trims start to make weird things (mostly go negative by -5 / -7%). As said: hot means stuck in traffic or after driving with fun
If it was a bad MAF cable I would see no signal, or a lower signal... but there are spikes!
A real weird idea: with the hot engine there is a certain amount of backfire that tricks the MAF and the trims
The MAF isn't directional, so a reverse flow is seen as "correct flow".
I hope this remains a weird idea... but it's the only thing I can imagine in case it's a mechanical issue (and not electronic).
04-10-2007, 10:48 AM
#4
hot air is not as dense, thus you'll get less airmass. if you're commanding the same amount of fuel as before though, you will get an effective rich condition.
airdensity=MAP/(287.05*TEMPk)
where TEMP is a blend between ECT and IAT, in proportions depending on airflow (the faster you go, the more IAT-ish it's going to be)
airdensity=MAP/(287.05*TEMPk)
where TEMP is a blend between ECT and IAT, in proportions depending on airflow (the faster you go, the more IAT-ish it's going to be)
04-10-2007, 12:12 PM
#5
Originally Posted by RedHardSupra
hot air is not as dense, thus you'll get less airmass. if you're commanding the same amount of fuel as before though, you will get an effective rich condition.
airdensity=MAP/(287.05*TEMPk)
where TEMP is a blend between ECT and IAT, in proportions depending on airflow (the faster you go, the more IAT-ish it's going to be)
airdensity=MAP/(287.05*TEMPk)
where TEMP is a blend between ECT and IAT, in proportions depending on airflow (the faster you go, the more IAT-ish it's going to be)
The commanded fuel quantity is for Lambda = 1.
Actually if it was 100% MAF and the MAF would bring me the correct quantity of air the idle trims would be the same for a warm or a super hot engine.
But you have reason: in SD I should see a difference. With the MAF I don't think so.
At idle it may be 50% MAF and 50% SD (?). maybe it switches back and forth from MAF and SD or stay somewhere in between. but not producs spikes (!!)
However: it's about the MAF signals. Why the heck do I see spikes?
2400 Hz is about 6.1 G/sec, the spikes are around 3000 Hz = 10.5 G/sec (17% higher!). The spikes last 1.5 - 2.5 seconds, so it's not only an electrical issue... I don't think so
Assuming the MAF and the wiring are OK ,what kind of mechanical issue can cause the spikes?
Assuming the engine is mechanically OK, whak kind of sensoric / electronic issue can cause them?
04-10-2007, 01:25 PM
#6
the fact that it commands lambda 1, doesn't mean it knows what airflow it gets. if maf isn't perfectly calibrated, it's going to throw off fueling. maf is calibrated for a particular intake tract. if i remember correctly, you have a supercharger, which definitely counts as a modification of the intake tract 
thus, your maf IS off, you just need to figure out how much and which way.
because of the dual nature of the stock operation (MAF/SD hybrid), you cannot change one and ignore the other. what if it will use the calculation/lookup you haven't tuned? what if it uses the one you want but it still does a 'sanity check' with the other, and if they're not really close, it's going to freak out again.
on top of that, there's always noise, especially at lower maffreq. it's extra bad if your maf was descreened. that's why we dont run purely off maf, it's just not precise enough for small amounts of airflow (ie. idle)
another thing is i remember you're one of the guys that refuses to go away from arbitrary IFR tweaking as main means of tuning. this way you will never get your airflow straight, thus your MAF/VE will never agree, thus you will always have some strange situations where the PCM freaks out and goes into panic mode. time to get away from the autotap/ls1edit. autotap itself is spectacularly slow as far as scanning is concerned, so the spikes might be a result of the crappy scanner, not a car problem.
thus, your maf IS off, you just need to figure out how much and which way.because of the dual nature of the stock operation (MAF/SD hybrid), you cannot change one and ignore the other. what if it will use the calculation/lookup you haven't tuned? what if it uses the one you want but it still does a 'sanity check' with the other, and if they're not really close, it's going to freak out again.
on top of that, there's always noise, especially at lower maffreq. it's extra bad if your maf was descreened. that's why we dont run purely off maf, it's just not precise enough for small amounts of airflow (ie. idle)
another thing is i remember you're one of the guys that refuses to go away from arbitrary IFR tweaking as main means of tuning. this way you will never get your airflow straight, thus your MAF/VE will never agree, thus you will always have some strange situations where the PCM freaks out and goes into panic mode. time to get away from the autotap/ls1edit. autotap itself is spectacularly slow as far as scanning is concerned, so the spikes might be a result of the crappy scanner, not a car problem.
04-10-2007, 03:17 PM
#7
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
The MAF frequency signal can be corrupted by any
EMI (electrical noise) from nearby cabling, injectors,
etc. With the MAF running at the low end of frequency
there is more of a "window" between the real pulses
for impinging noise to be picked up and register as
an extra pulse; one fake in between two real ones
would make effectively a doubling of frequency, for
that sample.
When hot you may just have the added activity of
the fans, adding load and load-dump spikes to the
wiring harness and putting static on the MAF output.
Separating the MAF wires from the rest of the
harness is something the Helm manuals sugest in
troubleshooting kooky MAF readings.
EMI (electrical noise) from nearby cabling, injectors,
etc. With the MAF running at the low end of frequency
there is more of a "window" between the real pulses
for impinging noise to be picked up and register as
an extra pulse; one fake in between two real ones
would make effectively a doubling of frequency, for
that sample.
When hot you may just have the added activity of
the fans, adding load and load-dump spikes to the
wiring harness and putting static on the MAF output.
Separating the MAF wires from the rest of the
harness is something the Helm manuals sugest in
troubleshooting kooky MAF readings.
Trending Topics
04-10-2007, 03:40 PM
#8
But I LOVE Autotap 
I tried: log trims at idle with and without MAF.
With MAF it's +/- 1%, without MAF it's 0 LTFT and 0 STFT.
Obviously (because I was watching it) the MAF frequency was spikes free (during 5 min... however).
In SD (MAF disconnected) the closed loop range of the VE table gives +/- 2% fuel trims with the actual IFR. Same results when the MAF is connected.
This should avoid the PCM to freak out
The MAF is stock: not ported, not polished, not descreened.
OK: Autotap may be the cause of the spikes but why always when the engine is hot?
Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so.
Again: hot causes spikes and fuel trims to go negative by 5/7 %
What can be the cause?
I tried: log trims at idle with and without MAF.
With MAF it's +/- 1%, without MAF it's 0 LTFT and 0 STFT.
Obviously (because I was watching it) the MAF frequency was spikes free (during 5 min... however).
In SD (MAF disconnected) the closed loop range of the VE table gives +/- 2% fuel trims with the actual IFR. Same results when the MAF is connected.
This should avoid the PCM to freak out
The MAF is stock: not ported, not polished, not descreened.
OK: Autotap may be the cause of the spikes
but why always when the engine is hot? Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so.
Again: hot causes spikes and fuel trims to go negative by 5/7 %
What can be the cause?
04-10-2007, 04:25 PM
#9
Originally Posted by jimmyblue
The MAF frequency signal can be corrupted by any
EMI (electrical noise) from nearby cabling, injectors,
etc. With the MAF running at the low end of frequency
there is more of a "window" between the real pulses
for impinging noise to be picked up and register as
an extra pulse; one fake in between two real ones
would make effectively a doubling of frequency, for
that sample.
When hot you may just have the added activity of
the fans, adding load and load-dump spikes to the
wiring harness and putting static on the MAF output.
Separating the MAF wires from the rest of the
harness is something the Helm manuals sugest in
troubleshooting kooky MAF readings.
EMI (electrical noise) from nearby cabling, injectors,
etc. With the MAF running at the low end of frequency
there is more of a "window" between the real pulses
for impinging noise to be picked up and register as
an extra pulse; one fake in between two real ones
would make effectively a doubling of frequency, for
that sample.
When hot you may just have the added activity of
the fans, adding load and load-dump spikes to the
wiring harness and putting static on the MAF output.
Separating the MAF wires from the rest of the
harness is something the Helm manuals sugest in
troubleshooting kooky MAF readings.
Something is still strange: I can let the engine warm up at idle until the fans start to cycle, watch the signals during minutes and everything remains quite and constant.
Only after a REAL long time or after I
around for a while the idle trims (and other low trims too!) become negative and the MAF "makes spikes".Maybe a connector that doesn't like high temperature... (I have an extention between the original harness and the MAF).
Is it a good idea to splice the original cable? Should I solder it?
04-10-2007, 04:46 PM
#10
" OK: Autotap may be the cause of the spikes but why always when the engine is hot?
Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so."
So try HPTuners and see if you get the same results. Can you borrow from a buddy or something, just to eliminate autotap?
but why always when the engine is hot? Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so."
So try HPTuners and see if you get the same results. Can you borrow from a buddy or something, just to eliminate autotap?
04-10-2007, 05:02 PM
#11
Originally Posted by Viper
" OK: Autotap may be the cause of the spikes but why always when the engine is hot?
Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so."
So try HPTuners and see if you get the same results. Can you borrow from a buddy or something, just to eliminate autotap?
but why always when the engine is hot? Hot doesn't mean overheated, just well warmed up as wen you are stuck in traffic during 1 hour or so."
So try HPTuners and see if you get the same results. Can you borrow from a buddy or something, just to eliminate autotap?
Maybe some so called tuner (expensive).
I have a fixed installed WB, but I think it doesn't help... unless I force the PCM to run open loop and see what happens with the AFR (?)
I'll check the MAF extension... a bad connection or the cable too close to other electrical devices can explain the spikes.
04-10-2007, 05:57 PM
#12
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
Can you log system voltage with ATAP? I know it's
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
04-11-2007, 01:00 AM
#13
FormerVendor
iTrader: (9)
Join Date: May 2003
Location: Jacksonville, Fl
Posts: 1,573
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by RedHardSupra
the fact that it commands lambda 1, doesn't mean it knows what airflow it gets. if maf isn't perfectly calibrated, it's going to throw off fueling. maf is calibrated for a particular intake tract. if i remember correctly, you have a supercharger, which definitely counts as a modification of the intake tract thus, your maf IS off, you just need to figure out how much and which way.
because of the dual nature of the stock operation (MAF/SD hybrid), you cannot change one and ignore the other. what if it will use the calculation/lookup you haven't tuned? what if it uses the one you want but it still does a 'sanity check' with the other, and if they're not really close, it's going to freak out again.
on top of that, there's always noise, especially at lower maffreq. it's extra bad if your maf was descreened. that's why we dont run purely off maf, it's just not precise enough for small amounts of airflow (ie. idle)
another thing is i remember you're one of the guys that refuses to go away from arbitrary IFR tweaking as main means of tuning. this way you will never get your airflow straight, thus your MAF/VE will never agree, thus you will always have some strange situations where the PCM freaks out and goes into panic mode. time to get away from the autotap/ls1edit. autotap itself is spectacularly slow as far as scanning is concerned, so the spikes might be a result of the crappy scanner, not a car problem.
thus, your maf IS off, you just need to figure out how much and which way.because of the dual nature of the stock operation (MAF/SD hybrid), you cannot change one and ignore the other. what if it will use the calculation/lookup you haven't tuned? what if it uses the one you want but it still does a 'sanity check' with the other, and if they're not really close, it's going to freak out again.
on top of that, there's always noise, especially at lower maffreq. it's extra bad if your maf was descreened. that's why we dont run purely off maf, it's just not precise enough for small amounts of airflow (ie. idle)
another thing is i remember you're one of the guys that refuses to go away from arbitrary IFR tweaking as main means of tuning. this way you will never get your airflow straight, thus your MAF/VE will never agree, thus you will always have some strange situations where the PCM freaks out and goes into panic mode. time to get away from the autotap/ls1edit. autotap itself is spectacularly slow as far as scanning is concerned, so the spikes might be a result of the crappy scanner, not a car problem.
And further the 98 Backup VE tables reduced resolution is not helping in SD, nor in these "transitions" you are going thru.
04-11-2007, 05:57 AM
#14
Originally Posted by jimmyblue
Can you log system voltage with ATAP? I know it's
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
You have reason, this is a possible problem: sometimes I see a short "low" in certain lights than can be related to a short "low" in the system voltage (headlights, radio...).
Maybe there is another component that suffers because of high temperature and causes "spikes" or "holes" in the whole electric circuit.
This would affect the O2's too, right?
If I disconnect the MAF during a scan
when I reconnect it the first signal is a positive spike up to 4000 - 5000 Hz... similar to what I see in my "hot" logs.let me see...
04-11-2007, 06:00 AM
#15
Originally Posted by Doc
+1
And further the 98 Backup VE tables reduced resolution is not helping in SD, nor in these "transitions" you are going thru.
And further the 98 Backup VE tables reduced resolution is not helping in SD, nor in these "transitions" you are going thru.
A 98 PCM in SD is always using the secondary VE?
Is there a way to force it to use the main VE table?
04-11-2007, 11:30 AM
#16
FormerVendor
iTrader: (9)
Join Date: May 2003
Location: Jacksonville, Fl
Posts: 1,573
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by tici
I think you are talking about the secondary VE table...
Originally Posted by tici
A 98 PCM in SD is always using the secondary VE?
Is there a way to force it to use the main VE table?
Is there a way to force it to use the main VE table?
I am currently in Madrid on business until Saturday. Want a once in a lifetime deal? Madrid is not that far from Switzerland-compared to Florida.
Check out this...
http://forum.efilive.com/showthread.php?t=4964
04-11-2007, 04:35 PM
#17
Originally Posted by jimmyblue
Can you log system voltage with ATAP? I know it's
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
an available PID but not whether it's "standard" or
"enhanced". Might be interesting to see if there is
any coincident bounce on the +2V (which MAF is
using for heater power, and any jump or sag may
shoot through the electronics and depend on the
internal control loop to null back out, taking some
brief interval to accomplish).
Of curse everything is perfect whar I watch it...
No MAF spikes, no weird fuel trims...
Assuming the electronic is OK, what mechanical situation would cause fuel trims to go negative when the engine (+ oil + other parts) are hot?
04-11-2007, 04:47 PM
#18
Originally Posted by Doc
Yepper, Backup, Secondary it's all the same.
No and this is the single largest reason why your next mod should be the pinswap. I launched the both barrels and changed the cluster and the tank/sending unit. This (the cluster and tank part of the swap) is no longer needed with EFI Live's capability to swap segments. All you will need now is a 99+up pcm, and the pcm harness pin swap, and EFI Live. Study the new stock GM Operating Systems and you will find many other nice features that will really help your blower setup. The clincher is EFI Live's free Custom Operating Systems that will give you the increased VE table range so you can safely run the higher boost.
I am currently in Madrid on business until Saturday. Want a once in a lifetime deal? Madrid is not that far from Switzerland-compared to Florida...
No and this is the single largest reason why your next mod should be the pinswap. I launched the both barrels and changed the cluster and the tank/sending unit. This (the cluster and tank part of the swap) is no longer needed with EFI Live's capability to swap segments. All you will need now is a 99+up pcm, and the pcm harness pin swap, and EFI Live. Study the new stock GM Operating Systems and you will find many other nice features that will really help your blower setup. The clincher is EFI Live's free Custom Operating Systems that will give you the increased VE table range so you can safely run the higher boost.
I am currently in Madrid on business until Saturday. Want a once in a lifetime deal? Madrid is not that far from Switzerland-compared to Florida...
It's just that I can't explain the fluctuating fuel trims at hot engine and the MAF "spikes"
What about the "once in a lifetime deal"?
Madrid - Zurich is still 1000 miles...
04-11-2007, 06:07 PM
#20
FormerVendor
iTrader: (9)
Join Date: May 2003
Location: Jacksonville, Fl
Posts: 1,573
Likes: 0
Received 0 Likes
on
0 Posts
Last time I checked my math 1k was alot less than 3k miles. 
I just happen to have tools and knowledge relatively close to your time zone for a limited time and thought I would offer some help. If you are happy being frustrated, no problemo. I just thought you were looking for absolution.
I just happen to have tools and knowledge relatively close to your time zone for a limited time and thought I would offer some help. If you are happy being frustrated, no problemo. I just thought you were looking for absolution.