damn that is nasty

 

You must have a K&N. And have had one that is over-oiled for a long time.

 

MAAAN! THis picture is 1.5MB!!! My laptop has just a dial up modem and it toke me 5 min to load your dirty MAF
I was thinking the same about cleaning it with brake cleaner or similar stuff.
How did you managed to get it so dirt? I have a K&N FIPK and my MAF is still nice and clean. I never oiled the filter... does it happen when you service it and put too much oil?

My girlfriend is also wondering why I always sit in front of the screen looking at strage curves... I always think "ok this is the last thing then I'm done". Yeah!
In spring time I'll install a water/alcohol injection system (non intercooled Vortech). Imagine the scanning & tuning stuff... Scary!

I went to your homepage: nice job! I'd like to do something similar but I don't have no place and no time and no second car. The blower is grate but more cubic inches are better for sure. Is it normal to have so much dirt on top of the pistons?
You write "the engine is shot, and this is the explanation..."
Sorry I don't understand it, could you please explain?

 

the maf will get nasty if you over oil the filter, all it needs it a light coating and the oil will eventually spread evenly over the whole filter. personally i went back to a paper filter after K&N and the holley pieces, then you dont have to worry about over oiling. too much oil will gunk up the maf and flow less air than a paper one. to clean off my maf i got the gunk cleaner from the hardware store by my house, looks brand new now

 

TAQuickness,

I would ditch the ported MAF and replace with a stock Delphi 85mm MAF with a known cal. This is the only way you can reverse engineer a new MAF cal. to suit your ported MAF. You have to start from a known baseline or you are chasing your tail.

I know its not your fault. Anyone who ports a MAF does not know what they are doing. GM designed these MAFs with an aerfoils to laminate airlfow over the sensors. Without it you get fluctuations and eddies around the sensors which will cause erratic g/sec measurements leading to unstable fuel trims. Porting a MAF will MAY seem to generate more power by virtue of the fact it MAY lean out your mixture. But you are only introducing random variables which the O2 sensors wil do their damdest to squash.

You want rock solid trims? Ditch the ported MAF and replace with a stock 85mm MAF with a known cal.

Absolutely. Sit tight for a sec.

These are HPT functions. They must be available in LS1Edit under Closed Loop Fuelling somewhere - perhaps under a different name. I don't own LS1Edit so I am sorry I can't help you more.

I can't post up pics. of my MAPS as I am working through a firewall. However, I will try to explain more clearly.

Closed Loop fuelling is all about O2 sensors reading an AFR around stoich and making small adjustments to the injector base pulsewidth to maintain stoich when either rich or lean conditions are detected. The rich/lean switchpoint is determined in the "O2 Rich/Lean vs Airflow" tables for both injector banks. 451mv is considered stoich and any O2 mv reading less than (lean) or greater than (rich) are countered by either adding or removing fuel depending on which condition is detected.

The size of each adjustment is determined by multiplying the "Closed Loop Proportional O2 Error" by the "Closed Loop Proportional Airflow Mode". To increase the fuel charge to maintain stoich in either lean or rich conditions simply increase the "Closed Loop Proportional Airflow Mode" values accordingly. This is only needed to maintain oscillating O2 sensor readings. You want just enough fuel charge to force oscillation. Too much and you introduce hunting. This is evident when hunting for idle RPM.

There are 9 Airflow Mode values [0-2-4-6-8-10-12-14-16]. Thus the O2 sensors only have 9 intervals over their oscillating range with which to maintain stoich.

Each Airflow Mode [0-16] is scaled directly to a g/sec Mass Airflow reading in the "Airflow Mode-Mode vs Ariflow" table.

You can indirectly correlate the Ariflow Mode to your LTFT MAP boundaries by loggging either your MAF g/sec flowwrate or VE derived Dynamic Air against the Manifold Absolute Pressure or MAP.

You want to determine what g/sec is evident at each MAP LTFT boundary. For example, say your LTFT MAP boundaries are set at 37-57-77 MAP. You need to determine the g/sec MAF readings at those points. in turn this will allow you to resolve the Airflow Mode applied to control closed loop fuelling.

The stock cal. for Airflow mode is a best guess at that. Its all over the place and for every plumbing change you make you need to ensure the Fuel Trim Cells are given max. resolution via the Airflow Mode to control closed loop behaviour.

Now you have g/sec at each LTFT MAP boundary you can resolve your current Airflow Mode via the "Airflow Mode-Mode vs Ariflow" table.

Lets say you logged 48gs/ec at LTFT cell boundayr 77 MAP. Lets also say the Airflow Mode table shows that for more than 48g/sec Ariflow Mode 16 applies. Now go to the "Closed Loop Proportional Airflow Mode" table and determine how many of the 9 cells contain values for 16. Just 1. So you only have 1 cell to control fuelling behaviour under acceleration.

No wonder your trims are whacked for half the histogram. There is not enough granularity for closed loop to apply any other fuel charge than the highest base modifier. Depending on the voltage differential from the O2s the gain error can modify the fuel charge somewhat. But by and large you will get larger fuel swings hence wildly fluctuating STFT corrections to maintain stoich. If the VCM overshoots the fuel swing needed to maintain closed loop it will go hunting, leading to loss of fuel economy and railing O2 sensors. This has a snowball effect for the next few seconds as it tries to stabilise. However by then you have let your foot off the gas changing the load characteristics.

And so on it goes. Your fuel swings carry into other LTFT boundaries resulting in further overshooting rich/lean conditions. The result? STFT and LTFT that are constantly changing.

You want solid fuelling behaviour across all LTFT boundaries.

You want no less than 2 Airflow Mode corrections to be applied for each LTFT MAP boundary. So, you need to readjust your "Airflow Mode-Mode vs Ariflow" to achieve this. This is what I have done:


Cell count__Airflow_Mode__g/sec__LTFT_boundary__driving style
3___________0-4___________7______37_____________idle & cruising
3___________6-10__________28_____57_____________off idle, acceleration & cruising
3___________12-16_________84_____77_____________accleration

Using these desired boundaries, alter your "Airflow Mode-Mode vs Ariflow" to suit. Use an evenly rising linear scale for each boundary. Insetad of a stepped graph you should see a smoother curve rounding off as it nears Airflow Mode 16.

Now you have reliable and stable trimming you can rescale your VE or MAF tables with confidence. All TRIMS should now rescale by whatever % you apply to the VE or MAF. They will be rock solid and you may find your STFT are hardly fluctuating from 0 - perfect. Your PE AFR will be stable and you can now move onto the fun part. WOT tuning.

 

Dial up...primitives... You know your computer would run faster if you blew the vacuum tubes off with air every once in awhile.

Just kidding...that picture was a PITA to download.

 

so getting an 85 mm MAF would be better than the stock 75? i have yet to read your whole post but for adjusting that would you just use scaling? or would you not have to touch the maf tables at all since the maf is unaltered?

if thats the case where is the best place to get an 85 mm from, might take out a lot of the issues that come up

 

That I have done, please see my next post

Amen

 

My recommendation is that you use the stock 85mm MAF and cal. to stabilise your fuel trims then right at the end rescale the whole MAF cal. by whatever average % your LTFTs are out. This I determined by looking at the LFTF histogram and deciding "yeah, it looks like 2% might just do it". 2% is not enough to drive your previous efforts at LTFT stabilising off.

Now if its off by 8% or more you may have other issues, like leaks, bad/non-stock IFR's or a bad MAF cal. to start with. You could try it and see - same principle applies - but you dont want to rescale to much or you will negate the efforts you have gone to to stabilise your closed loop fuelling. In that case I guess you could just revisit your Airflow Modes to suit and check again. Either way, you want rock steady fuelling in closed loop before you rescale.

Chasing variable fuel trims is like chasing the wind. Control their behaviour and you have target to shoot at.

 

ok so that sounds a lot simpler, just tune it for everything else, im leaving the stock maf on for tht, then swap out maf sensors and then scale accordingly in the maf tables? as long as its not off by a lot?

im trying to find where ic an get one at a good deal, any idea?

ls1speed has one for 149.99 that says its calibrated for 98-02 ls1 cars. is that what i would need?

 

Yes. As long as the MAF cal is right. I prefer to stick to GM MAFS and cals. as I believe they are more accurate due to their big development and production costs.

You don't need the stock MAF if you have an aftermarket one with a good known cal.. You can apply the same principles with any MAF. Just avoid MAFs that have been tampered with and violated - your car could get crabs.

 

MNR-O, no way to do this with edit
I see what you mean: slow down the STFT reaction in order to avoid continues changes. When everything is fine I have +/- 3 STFT, also at constant load.
They just fluctuate. No matter if I have a MAF or not. Only @ idle and WOT they stay constant.
BUT... should I first go in open loop? Maybe this is the difference!! during my tuning I stay closed loop and see what the LTFT are doing during at least 30 min logging...

 

Sorry to hear you can't do this with LS1Edit because it really does work. Perhaps you can loan someones HPT to assist you there.

I have now logged 4 runs into the City and back, cruising, WOT and everything in between from ambient temps. 15*C to 30*C. I have observed no change or fluctuation in either my LTFT or STFT. They constantly sit at +/-1 all the time. The killer is my STFT. I could never get them stable - always reading anywhere from -12 to +8. Now - all 0! I just can't destabilise them!

I will post some proof and screenshot my LTFT and STFTs. I'll even make available my 4 logs for download.

As for open loop, even if I nail you nail it I cant shift them.

It doesnt matter whether you are open loop or not. All you need to do is determine what g/sec corresponds to your LTFT MAP boundaries and rehash your Airflow Mode table to deliver 3 Airflow Modes per Fuel Cell of resolution. After that, drive around in closed loop in 20 mins to allow time for your logged STFTs to settle. Then rescale your MAF by the entire average LTFT % observed during closed loop.

Even after you provide 3 cells resolution per Fuel Cell, your TRIMS should stabilise. This is, of course, assuming you have a correctly calibrated stock MAF and MAF cal. If you still get fluctuations I would revisit your O2 sensor oscillations to ensure they are not flatlinging or railing.

 

Here are my 4 logs so far. For those with HPT right click and select "Save As". Apologies to those who don't have HPT.

http://users.bigpond.net.au/thompson...mnr0_34o_1.zip
http://users.bigpond.net.au/thompson...mnr0_34o_2.zip
http://users.bigpond.net.au/thompson...mnr0_34o_3.zip
http://users.bigpond.net.au/thompson...mnr0_34o_4.zip

Here is a snapshot of my O2 oscillations at idle and light throttle take off. You need happy O2 oscillation in order to get correct fuelling at idle and in closed loop.



Here is a screen shot of the first logs Avg. LTFT.



And the same logs Avg. STFT.



Now for the second log. As you can see, its a mirror copy of the first log. Note I have DFCO and Lean Cruise enabled so you will expect to see transient rich spots coming off lean cruise and going into DFCO and idle conditons as closed loop hunts for stoich.



And the same logs Avg. STFTs.

 

Here are snapshots of my Airflow Mode settings. My O2 R/L settings are the same for each Bank. My O2 Error tables are stock, as are my Closed Loop Integrator Delay Airflow Mode and O2 Error tables.





I had to set my Idle Airflow Modes higher from 128 because they weren't oscillating after fitting 2.5" shorty headers and opening up the CATS to 2.5". The extra airflow cooled down the O2 sensors.



And finally the Base Airflow Mode which executes closed loop trimming. Note that I provide linearly increasing values corresponding to each Fuel Cell in g/sec. My MAP Fuel Cell boundaries cross over at 7/20/38g/sec for 37/57/77 MAP. This yields 3 Airflow Mode cells per Fuel Cell - the max. even resolution possible.

 

I see, I have the wrong software... hehe! But my goal isn't so high so I'll stick with what I have.

I can get a nice LTFT table like yours, but STFT not. But wait a minute: are you talking about averaged STFT during a long period? My average is also about 0 but if I log them their curve follow somhow the O2 sensors. Amplitude maybe -3 / +3. A little larger band if I drive no MAF.
To correct the MAF curve I log for LTFT, cells, G/min and frequency. Put everything in a pivot table (Excel) and aim at -1 / 0 LTFT.
This is the only thing I can make using Autotap and Edit.
All in all I'm happy with the result, the only thing I hate is to have a perfect VE table today and see how it's out next day. Not so much, maybe 1% more or less on the whole RPM vs. MAP table.

 

ok so im still confused about the 85 mm, is one from a spnosor ok or do i want one from like sdpc from a z06 that hasnt been fooled with?

 

i might have you get me through this step by step once i get hp tuners, should be here today i would hope. being the first time i have heard of this i dont fully understand the process, so if you dont mind maybe giving me a step by step for dummies lol thanks for all the great info though!

 

Be wary of workshops saying the MAF is calibrated for your car. Unless you are supplied the MAF cal. with the MAF I wouldn't trust it.

 

do they tell me the calibration or is it already built in? like do i have to change something in the tables or no? and i just ask them about it?

MNR- do you think when you have time you could either post or pm me your whole process of tuning? and it is done before the VE tuning?