Sanity check, please...

[joecar]

True (you left out some of the Holden/HSV products)...

but when you also include the "plain" engined vehicles models that GM has made, quite a few of those came without a MAF.

But, as you said, by using a MAF, GM must be seeing advantages.

That's a good pic... that has integrated MAF and IAT and what else...?

Only performance V8 vehicles that GM used Speed Density for was the 1990-95 LT5 ZR-1 Corvette. 1990-91 L98 350 and 1992-93 LT1 Corvette, 1991-92 TPI 305/350, 1993 LT1 Camaro/Firebird. All the turbo V6's TTA,GN, GNX were all MAF, as were all 1985-1989 TPI 305/350 Camaro/Firebird/Vette and 1994-2002 and 2010+ Camaro, and 1994-2013 Vettes all use MAF's. I'm not counting TBI which is indeed Speed Density, but not a performance system IMO. Why is GM holding fast with the MAF system? Here is a pic of the GEN 5 SBC MAF for the 5.3 L83 and 6.2 L86 MAF. They have eight (8) wires. peace Hog

Bret has a possible advantage for GM using MAF's, to save time by avoiding fine tuning:

...

And as far as why GM uses the MAF - I personally believe it is because they don't want to take the time to map out the VE table the way it would need to be in order to get the same driveability and throttle response. It is much easier to use a blend of MAF and VE like they do.
...

[joecar]

I'm sure there have been zillions of discussions about the pros and cons of MAF vs speed density tuning. I've been reading what I can, and each camp appears to be as firmly entrenched in their beliefs as the other.

I guess what it all boils down to is if there is any difference in accuracy of the data the PCM needs to correctly determine how much fuel is needed for optimum combustion for each instantaneous NEED of the engine. Has anyone created a map to log MAF frequency data in a RPM x kPa matrix? I'd be curious to see if every cell has a completely unique value in it. If so, then they both may just be different ways of solving the same problem of how much air is available for combustion when the fuel injector needs to be fired. But then again, perhaps all of the tables are just too granular for any realistic analysis in that manner.

Hi Rich,

yes, there have been many many debates over MAF-vs-SD.


More correctly, since we're talking MAF andor VE, that is the data the PCM uses to correctly calculate how much air filled the cylinder... and then fueling is determined from that based on the fueling tables (i.e. indirectly);

( I know you already know that ... to keep learning about the PCM we have to think/speak strictly in terms of what the PCM does )

airmass is tuned, fueling is stated (i.e. fueling follows airmass)

( of course you have to know what to state, but the tuning of this is different, requiring dyno and dragstrip)


I'll dig up some logs containing MAF and I post some pics of MAF x RPM x MAP (noting that MAF is uniquely determined by MAFFREQ).

[BLK02WS6]

Here is your posted tune with only the commanded fueling changes (B3613, B3618, B3647). When I say commanded fueling - I think of it this way - commanded is the target you are trying to get to... the way you get there is by the VE and, if used, MAF tables. So, you set your targets at some sane reasonable value for your setup. Then, you correct the VE table based on your BEN readings until actual fueling (from your wideband readings) equals commanded readings. Then, after you have actual equal to commanded, you put it on a dyno and tweak commanded fueling (actual will follow at this point because you have corrected it already) to obtain optimum torque and horsepower throughout the RPM range.

With timing, start off conservative and raise it on the dyno to find optimum timing throughout the RPM range based on torque and horsepower readings...

[BLK02WS6]

I think Joe cleared up most of your other questions already (or maybe confused you more )... Joes speaks PCM better than I do... I try to put things in the simplest terms possible cause that is the way I learned. Whatever you do, don't get discouraged! You probably know more than you think... If you don't get something, just ask - Joe is one of the most patient people on the face of the earth and will continue to try and help you pretty much forever... and I try - between the day job, kids, and trying to tune on the side, I don't have a ton of free time...

[joecar]

lol...

Rich, the method is to adjust the VE table until the wideband shows the same fueling as stated/commanded...

[ this requires all other source of airmass to be disabled (i.e. MAF), and it requires all trimming to be disabled (i.e. CL/LTFT/STFT/SOL) ]

[Rich Z]

Here is your posted tune with only the commanded fueling changes (B3613, B3618, B3647). When I say commanded fueling - I think of it this way - commanded is the target you are trying to get to... the way you get there is by the VE and, if used, MAF tables. So, you set your targets at some sane reasonable value for your setup. Then, you correct the VE table based on your BEN readings until actual fueling (from your wideband readings) equals commanded readings. Then, after you have actual equal to commanded, you put it on a dyno and tweak commanded fueling (actual will follow at this point because you have corrected it already) to obtain optimum torque and horsepower throughout the RPM range.

With timing, start off conservative and raise it on the dyno to find optimum timing throughout the RPM range based on torque and horsepower readings...

OK, thanks for helping me with this. I'll take a look at the tune you gave me and see if the pieces fit together in my head.

As for timing, I don't have a dyno available. So I have to do what I can on the street. Luckily I'm in an area where long stretches of pretty deserted roads are easy to come by. Of course, these kinds of roads also have the hazard of a deer deciding to jump in front of you. So from what you can see, does the timing being logged look OK, regardless of how it got there?

Hopefully the rain will take a break so I can get the car out and play with this some more. I did spend some time with the VE table in the lower rev range, so I've got to take a stab at some boost readings to get some data logged there.

[Rich Z]

lol...

Rich, the method is to adjust the VE table until the wideband shows the same fueling as stated/commanded...

[ this requires all other source of airmass to be disabled (i.e. MAF), and it requires all trimming to be disabled (i.e. CL/LTFT/STFT/SOL) ]

OK, I think that makes sense to me. Kind of. The VE table is telling the PCM what the air charge is expected to be. The fuel tables tell the PCM what we want the AFR to be based on that air charge. The differences between the commanded AFR and the actual AFR determined by the wideband sensor are the differences between what the VE table IS and what it SHOULD be. Does that sound right?

So in my case, I actually have TWO VE tables to calibrate. One at naturally aspirated MAP values (B0101), and the other at boosted MAP levels (A0009).

Oh, not to change the subject, as this is jumping ahead a bit, but I'm presuming that I will be enabling the STFTs after the VE tables are kicked into line. But what about LTFTs? I was told that I shouldn't use them because they will interfere with boost fueling. Yes or no?

Thanks for the help.... I'm getting there, I think.

[joecar]

OK, I think that makes sense to me. Kind of. The VE table is telling the PCM what the air charge is expected to be. The fuel tables tell the PCM what we want the AFR to be based on that air charge. The differences between the commanded AFR and the actual AFR determined by the wideband sensor are the differences between what the VE table IS and what it SHOULD be. Does that sound right?

Yes, you got it.

So in my case, I actually have TWO VE tables to calibrate. One at naturally aspirated MAP values (B0101), and the other at boosted MAP levels (A0009).

You can think of the two VE tables being joined at the 105 kPa column... going above 105 kPa is boost.

Oh, not to change the subject, as this is jumping ahead a bit, but I'm presuming that I will be enabling the STFTs after the VE tables are kicked into line. But what about LTFTs? I was told that I shouldn't use them because they will interfere with boost fueling. Yes or no?

STFT/SOL can be enabled... (SOL is simply STFT on any stoich cell in B3647); STFT/SOL are instant (they are applied immediately);

LTFT's can also be enabled if they are going to be zero (if VE is accuratly modelling cylinder airmass, then LTFT will be very close to zero)...

the problem with LTFT's is that positive LTFT's are added to fueling in PE mode...

LTFT behavior in CL mode::
- PE not enabled: CL trimming adds LTFT to fueling;
- PE enabled: the last LTFT is rounded up to zero and then added to fueling (on top of PE fueling), see note in next paragraph;
- LTFT's are remembered (learnt) in a 4x4 + 8 array (there are 4x4 MAPxRPM cells plus 8 additional cells for things like idle and WOT);
- a new LTFT is stored (learnt) in a particular cell when the STFT takes an excursion past some threshold (positive or negative)
(and the STFT is then set back to zero);
- the LTFT from the last operating point in the 4x4 array is stored in the WOT cell and added to fueling;

note:
the second point above is saying this:
negative LTFT is rounded up to zero, positive LTFT is retained, the resulting LTFT is added to fueling;

note also:
in CL mode, going to PE retains CL mode (in a manner) and does not go to OL mode
(i.e. B3647 does not become active; the only active fueling table is B3618).

see post #4 here: Summary-Notes

[ ok, so what was GM thinking...? The consensus is that they were thinking of protecting the engine as best they can (implying that they were expecting people to hack into their PCM's) ]

Thanks for the help.... I'm getting there, I think.

No worries... I hope I haven't confused everything with the description of LTFT's.

[joecar]

Actually, the LTFT array is 4x4 (MAPxRPM) and there are an additional 8 cells (i.e. for WOT, for idle with/without AC, ...).

[Rich Z]

No worries... I hope I haven't confused everything with the description of LTFT's.

Well, honestly a lot of that didn't sink right in. But I think the gist of it is, I think, that if I want to use LTFTs, then to keep them from affecting the AFR at WOT I would have to fool them into being irrelevant by making them show lean. That way they won't affect the WOT fueling. Or perhaps if the tuning is already pretty close to spot on, then the LTFTs won't be a significant influence anyway.

BTW, a question that I asked has been overlooked I think. Based on my log posted, how does the timing look? Honestly, when I looked at my very first log, I was shocked at how high the timing values were. But as best I can tell there is no *real* KR being logged, so I guess it's OK. But I would be quite a bit relieved just to hear some folks TELL me that it is actually OK.

Thanks.