question on maft settings, and autotap o2, a/f readings

[John_D.]

I've been playing with the autotap and maft today, and the more I look at the numbers, the more confused I get.

I thought the car was running a little rich, based on the reported a/f ratio. Also the ltrims were about -7 (also indicates rich).

So I changed the base setting to 5 lean, and then 10 lean. The ltrims came in to about 0 to +3. I also watched the reported a/f ratio after making the base setting, and then changed the WOT setting, first to 4 lean, then 6 lean.

But, I started looking closer at the o2 readings I logged. They're all over the place. Is this typical, or are my sensors about shot? They will run anywhere from .1x to .8x

I thought I could use the reported air/fuel ratio from the autotap. I've been trying to get 12.3 at WOT instead of 10.x But, the o2 readings do not support the a/f readings, so I'm confused.

When I'm getting 10.1 to 10.4 (pretty rich) the o2's are .83 to .85, which is supposedly leaner than 12.3. So which one is right, the o2 or the a/f?

Where does autotap get the reported a/f ratio? Is this a target number the pcm is shooting for, or is this a measured number?

[ June 02, 2002: Message edited by: John_D. ]</p>

[John_D.]

Ok after researching this some more, I'm not nearly as confused.

I found out that the .89-.90 is not the 14:7 like I originally thought. It's the rich reading that I'm supposed to be shooting for at WOT. So I was pretty close after all. And the all over the place readings are to be expected during normal (not WOT) conditions, because .5 is 14:7 so it will bounce on both sides of .5, not hover around .9 like I was thinking. And under deceleration, it may drop way down, even all the way to zero, and that's ok too.

I'm just going to ignore the reported air/fuel ratio from autotap and go with the o2 sensor readings.

I also figured out what the references to fuel cells 1-19 meant. I thought there would be like 19 columns of 02 readings in the spreadsheet. Apparently there is a "fuel cell" parameter I can pull back from autotap, and I should only include the o2 readings if the fuel cell value on a row is 1-19. With 20 and 21 being decel and idle, and 22 being WOT.

[Magnus]

John, yea, forget that a/f reading..

When I tune for the MAFT, I ONLY look at engine RPM, timing advance, knock retard, and o2's... thats IT.. you get much more data this way.

I STRONGLY encourage you to leave the base settings at zero unless you have severe driveability issues. Changing the base setting mutliplies the error you will see as the #'s increase.

Leave it at zero.

Then just play with the WOT settings to get your desired O2 reading.

Your car will run MUCH more consistant this way.

[John_D.]

Thanks Magnus.

Over the weekend I found an earlier post from you where you said you left the base setting alone. But in some of my other research I found some information saying the ltrims do get factored in to the WOT, as an offset to the table. So I decided to try it...

Makes sense to work on the WOT only though. Since the pcm will manage part-throttle anyway. Then I only have one variable I'm changing instead of two, to hit my WOT target numbers.

I have two more quick questions for you:
1. Do you also get tps position or fuel cell back from autotap, to make sure you analyze WOT rows only? Or do you just log at WOT only?
2. I saw you upgraded to maft+. Did you need to change the timing very much? I see kr coming in at WOT, so I can't see bumping the timing up, unless it is just to get it higher at certain mid-rpm ranges where there is no kr currently, and still let the kr continue to take it back out in the problem ranges.

[Magnus]

1. I know when i'm in WOT from looking at the log.. RPM's will go to 6K and o2's wont cycle. [img]smile.gif[/img]

2. I got a tenth or so by adding timing. Definatley helped.

You really shouldn't be seeing KR on stock style cars... I hope your running 93 octain though.

When do you see KR, what is your o2 readings at that point?.. what is your timing advance also?

[DEE97GTP]

I believe the A/F ratio from autotap is the DESIRED ratio, not what it is measuring.

DEE

[John_D.]

Thanks for the feedback Dee, that's what I'm thinking too. It's definitely not a reliable number for tuning.

Magnus - here are the numbers I have on the kr's, to answer your question:
(I wasn't logging o2's or ltrims before the exhaust change though, so that is missing on the first tuning run I have listed...)

On takeoff, before exhaust, before maft:
2-4.6 kr, 12-17 timing, 3800-5000 rpm, 1st gear WOT
2 kr, 14-16 timing, 3800-4500 rpm, 2nd gear WOT

Interstate cruise, after exhaust, maft base 5% lean, WOT 4% lean (3rd tuning run):
2-3 kr, 14-16 timing, 4000-4700 rpm, o2 .81-.83, WOT
2-4 kr, 14-16 timing, 60% tps, 4000-4700 rpm, 02 .62-.72
(ltrims -6,-9)
Interesting that the kr was the same at 60% tps as at WOT, while the pcm was holding 14:7 vs. the rich WOT target.

Interstate cruise, after exhaust, maft base 10% lean, WOT 6% lean (4th tuning run):
2-4 kr, 15-16.5 timing, 96% tps (??), 3800-4300 rpm, o2 .62-.75
2-4 kr, 15-17 timing, 4100-4900 rpm, o2 .68-.80, WOT
5 kr (max kr logged ever), 13 timing, 4400 rpm, o2 .71, WOT
(ltrims 0 to +3, both banks)
I don't know about the 96%, I must have been playing the car a little at that point instead of running wide open.

I've since pulled the WOT back to 2% lean, base back to 5%, after figuring out that the a/f ratio is not what I should have been looking at. I'm probably lucky I didn't hurt anything on the last tuning run. I know the power was down from normal, it wouldn't hold speed uphill, usually it will still gain speed on that hill. Haven't logged any since then. When I pull the base back in, I'll probably have to tweak the WOT again.

So, does the timing number reflect total actual timing with kr factored in, or is that the number the pcm would be shooting for if there was no kr in the mix?

Oh, and all this is on regular 87... I figured I didn't need anything more unless I specifically tuned for it (more advance, etc.).

[Magnus]

Start using 93... especially if you like to drive your car at WOT.. [img]smile.gif[/img]

If you put your base setting at 5% lean, that is factored in at WOT too.. and then your changing your WOT to the lean position as well which just leans you out even more.. if your base setting is lean, most likely your WOT setting has to be rich.

Just keep your base at zero for now.. and your WOT at zero to... drive 15 miles so the car can learn, vary your throttle.. then log some WOT.. and see what your o2's are then.

I'd also run the 87 out and put in some 93 and then monitor KR again.

[John_D.]

<blockquote>quote:</font><hr>Originally posted by Magnus:
If you put your base setting at 5% lean, that is factored in at WOT too.. and then your changing your WOT to the lean position as well which just leans you out even more.. if your base setting is lean, most likely your WOT setting has to be rich.<hr></blockquote>


Magnus - I have to run this by you, because it sounds backwards to me, and I have to understand it instead of just take it on faith... This example is based on a starting point of negative ltrims (like I had).

1. Strims and ltrims are a form of maf calibration.
2. If the maf reads more air than actual, the pcm will add too much fuel.
3. The pcm will detect a rich condition via the o2 sensors, and generate negative ltrims to correct for the maf deviation.
4. If the pcm thinks the maf is reading too much air, then it will assume at WOT there will be too much fuel as well, if not corrected by an amount similar to the ltrims.

5. If the uncorrected ltrims are allowed to be factored into the WOT mode, then the pcm is going to lean out the WOT mixture. Because the pcm thinks the maf is off, and has negative ltrims (less fuel) as an input parameter.

6. If the maft is used to correct the maf reading in the first place, then the ltrims won't be negative, and the pcm will not be trying to also lean out the WOT mixture by a similar amount. If anything, with the maft set to lean the car out, it should cause ltrims to go slightly positive, adding fuel.

7. If the ltrims go positive, to add fuel, then the same thing should happen at WOT as well. So the pcm would actually make WOT richer as a result of leaning the maf reading?!

This is all based on the pcm not having any way of being aware that I have leaned the mixture. It only knows that it has to add fuel for some reason... (as evidenced by its own slightly positive ltrims) Since the pcm only knows ltrims, at WOT it actually will do the opposite of what I have told it to do via the maft, because it is compensating for the maf readings.

So do you see any holes in this, or does it sound right (although I have to admit, it is very counter-intuitive).

[ June 03, 2002: Message edited by: John_D. ]</p>

[Magnus]

<blockquote>quote:</font><hr>Originally posted by John_D.:
1. Strims and ltrims are a form of maf calibration.
2. If the maf reads more air than actual, the pcm will add too much fuel.
3. The pcm will detect a rich condition via the o2 sensors, and generate negative ltrims to correct for the maf deviation.
4. If the pcm thinks the maf is reading too much air, then it will assume at WOT there will be too much fuel as well, if not corrected by an amount similar to the ltrims.

Very nice. LTrims are form of MAF calibration or air/fuel ratio calibration. It is a factor that is used with the MAF frequency to help determine the pulse width.

If the PCM thinks the MAF is reading too much air, negative LTrims will be generated. LTrims are used at WOT to calculate the pluse width and NOT the o2's... O2's are used during cruse to recalibrate the LT's... its the difference between closed loop (o2's) and open loop (cold start, wot).


5. If the uncorrected ltrims are allowed to be factored into the WOT mode, then the pcm is going to lean out the WOT mixture. Because the pcm thinks the maf is off, and has negative ltrims (less fuel) as an input parameter.
The PCM doesn't really know the MAF is off.. it just knows the calibrations differ from stock. Fuel pressure is another thing that can affect LT's. I don't really think there are ever uncorrected LT's except when you reset the PCM or make a base change with the MAFT. The PCM learns LT's very quickly and will use them at WOT.


6. If the maft is used to correct the maf reading in the first place, then the ltrims won't be negative, and the pcm will not be trying to also lean out the WOT mixture by a similar amount. If anything, with the maft set to lean the car out, it should cause ltrims to go slightly positive, adding fuel.
Yes, this is true. But there is no need for this. I drive with -14% LT's and I think my car does run very well. ;) Don't think of the PCM leaning out the WOT mixture..it actually just uses the LT calibrations to achieve the correct air/fuel mixture.


7. If the ltrims go positive, to add fuel, then the same thing should happen at WOT as well. So the pcm would actually make WOT richer as a result of leaning the maf reading?!
No, just like above.. both cases would achieve similar WOT o2 readings as thats what the LT's are for.. they are calculated during cruise and accel as the factor by which the air/fuel ratio is off.. that factor still aplies at WOT. [img]smile.gif[/img]


This is all based on the pcm not having any way of being aware that I have leaned the mixture. It only knows that it has to add fuel for some reason... (as evidenced by its own slightly positive ltrims) Since the pcm only knows ltrims, at WOT it actually will do the opposite of what I have told it to do via the maft, because it is compensating for the maf readings.
Eh, again not really.. at WOT it will use the calculated LT's to determine pulse widths.

June 03, 2002: Message edited by: John_D. ]<hr></blockquote>

The problem with changing the base setting is it throws off the WHOLE curve by a factor.. so factor for error is multiplied. You will have more inconsistant WOT readigns this way.

I strongly suggest you leave the base setting at ZERO unless you have sever driveability issues. It will be MUCH easier to fine tune WOT this way.

My car dies a lot on cold starts. It idles very rich. I have -14% lt's... yet I leave my base at zero. My WOT readings are VERY precise and its much easier for me to fine tune my WOT O2's.

Even with a 5% base change your WOT o2's will vary thus making it harder to shoot for a target O2.. you basically have to shoot for a range then.

[John_D.]

Ok, I got really wordy, and still must not have gotten the main point across. I think we are saying the same thing about ltrims being a factor in WOT, and the o2 readings are not. (except for tuners like us who use them outside the system)

There is one thing we are looking at differently. And because I need to make sure I understand the concept, whether I end up using the base setting or not, here is the real question/issue:

If I set the maft base to a leaner setting, the o2's will temporarily go lean, until the pcm detects this and bumps the ltrims richer. Am I right so far?

If an ltrim says go richer at part throttle, the pcm will also run a richer mix at wot, because that's what the ltrim says to do.

So, if I say "lean" on the maft, the net result is a richer ltrim than what I started with? Right? If not, where am I taking the wrong turn in the logic above?

(and yes, I would definitely agree that your car runs great!!) I just have that awful need to understand how things work... ;)

[Magnus]

<blockquote>quote:</font><hr>Originally posted by John_D.:
Ok, I got really wordy, and still must not have gotten the main point across. I think we are saying the same thing about ltrims being a factor in WOT, and the o2 readings are not. (except for tuners like us who use them outside the system)
Correct

If I set the maft base to a leaner setting, the o2's will temporarily go lean, until the pcm detects this and bumps the ltrims richer. Am I right so far?
Correct.

If an ltrim says go richer at part throttle, the pcm will also run a richer mix at wot, because that's what the ltrim says to do.
The LT's arent really saying go richer. They are more of a calibration factor.. for example...

Cruising at 40mph...
Stock TB: 5lb/min.. 0 LT, normal Air/fuel ratio
Ported TB: 10lb/min.. 0 LT, VERY rich air/fuel ratio
Ported TB: 10lb/min -5% LT, normal Air/Fuel ratio

WOT at 5500 RPM...
Stock TB: 20lb/min.. 0 LT, .900 O2
Ported TB: 28lb/min.. 0 LT, .950 O2
Ported TB: 28lb/min.. -5% LT, .900 O2

So, if I say "lean" on the maft, the net result is a richer ltrim than what I started with? Right? If not, where am I taking the wrong turn in the logic above?
The net result is a positive LTrim over what you started with.

(and yes, I would definitely agree that your car runs great!!) I just have that awful need to understand how things work... ;) [/QB]<hr></blockquote>

If you have a -5 base setting.. your LT's will factor the new #'s to achieve the correct air/fuel ratio.... so when you go WOT, your air/fuel ratio will be normal.. .900 or so.. It uses the pre calculated LT's to do this. HOWEVER usually a lean base results in a lean WOT...

Then your leaning out the car EVEN MORE with your wot settings.

Another example...

You say your LT's are -7. At any given MAF value, it has to reduce it by a factor using the -7 value to achieve the desired air/fuel ratio. Even at WOT...

If your LT's are 0, the only thing different is there is no factoring done.

My view is this.. Do not modify the MAF singal unless you absolutely HAVE to... When you modify the maf signal you're mutliplying the error! only fine tune with the 2%,4%,6% etc. settings at WOT.. your car will move much faster. [img]smile.gif[/img]

Also, for NA, try to shoot for a .850 O2 at the track with 93 octain... Any improvements over .900 o2's???? ;)

- Keith

[John_D.]

<blockquote>quote:</font><hr>Originally posted by Magnus:

WOT at 5500 RPM...
Stock TB: 20lb/min.. 0 LT, .900 O2
Ported TB: 28lb/min.. 0 LT, .950 O2
Ported TB: 28lb/min.. -5% LT, .900 O2
<hr></blockquote>

Ok, I'm trying to follow this.

Let's say I had a -10 ltrim to start with. For this example we'll say it pulled the o2 to .850 (following your example of 0=.95, -5=.90, then a rough extrapolation would give -10=.85) So far, so good?

So if start with the -10 with no maft, then I lean the maft and get a -5 ltrim as a result, now the car will run .90 o2. And if I lean it even more, to get a 0 ltrim, the car will run a .950 o2. So the more I leaned the maft base, the richer the car ran at WOT, using your numbers?

Do you see why I'm still stuck here...?


<blockquote>quote:</font><hr>Originally posted by Magnus:

Also, for NA, try to shoot for a .850 O2 at the track with 93 octane... Any improvements over .900 o2's????<hr></blockquote>
Are you asking me to try both these at the track and get times? Just curious what you were saying here. I would like to get it pinned down, but with the ati in the works, I probably won't get too carried away with it, since all that will change anyway when the ati is on it...

I am thinking about trying to hit a .87 or so, which should be closer to 12:5-12:7, instead of 12:3. It's hard to find a good cross reference of o2 voltage to a/f on the far reaches of the range. Even if I did, who knows how accurate a narrow-band with 75,000 miles on it will be anyway...


p.s. the kr's got better with the richer mix. Not gone yet. May try some 89 octane and see what that does first. The problem rpm range seems to be 4000-5000. It would be great to be able to tweak that table just a bit...

p.p.s. never mind about that cell 1-22 stuff I posted earlier. That must be for the v8's only. Autotap reports cells 0-4 only, on mine.

[ June 04, 2002: Message edited by: John_D. ]</p>

[Magnus]

Your confused because your not thinking about the MAF readings....

Lets say the TB doesn't change, and the ACTUAL air flow is the same for EACH of our following cases.... meaning the actual MAF readings are inaccurate...

MAF - LT - O2
15lb/min - 0 - .850
15lb/min - +5 - .900
20lb/min - -5 - .850
20lb/min - 0 - .900
20lb/min - +5 - .950
25lb/min - -5 - .900
25lb/min - 0 - .950
25lb/min - +5 - .990

The ACTUAL air flow remains constant through our tests, yet the MAF READINGS vary... THUS the LT calibration factor is used to achieve the correct O2.

Forget about your base setting.. just leave that at zero and work with the -7% LT... then look at your O2's at WOT and fine tune them with the MAFT WOT setting. [img]smile.gif[/img]

The changes you make to the base setting on the MAFT will be learned by the PCM adjusting the LT's.. So you should run the same air/fuel ratio regardless, HOWEVER this is not the case because as you modify the MAF signal with a MAFT type device, you are just multiplying the end error.

Also, if your car is going to be boosted, don't shoot for .850 O2's. lol.. ONLY NA.

[John_D.]

Ok, uncle...!

Still not convinced that leaning the base = leaner wot, but gotta draw the line somewhere. [img]smile.gif[/img]