Thoughts on this thread???

[SSpdDmon]

http://www.ls1tech.com/forums/showthread.php?t=726607

This isn't the first time I've heard what this guy is saying. However, I keep hearing it with a lack of explanation. I hear the "I've been in the source code of the PCM" and the "that's not how it was designed to work" statements. But, I don't hear anything more than that because explanation would expose 'trade secrets.' Are we full of $h!t with the VE/MAF manipulation process we go through or is he with his manipulation of the commanded fuel in open loop while leaving MAF/VE stock?

[joecar]

Jeff, you may have answered you question in post #19...

I don't know...

I personally like my actual AFR to match my commanded AFR...

Sure, maybe GM didn't intend for the PCM to run MAF-less, but it seems to run ok (there may be a few side effect tho, for example trans torque calc may be off, and other things)...

Source code does not necessarily reveal the original intent...

It's funny to think that the original intent was to run with mismatched AFR's... and consider that GM cams are not big enough to cause problems of false wideband readings.

[joecar]

...Source code does not necessarily reveal the original intent...Source code may even obscure the original intent (even to the original author)...

Question: for what reason would someone want to tune by "indirect" methods...?

[picnic_george]

Very interesting reading. I figure as long as the car starts, idles, cruises and does everything else well. Then it's pretty good, at least for me.

I like the fact that you don't really discount anything he says but you do ask for proof. You are obviously a student who wants to learn more. I think that's better than saying you helped invent the thing and you can't tell anyone how it works. If the information is there then why not share it?

I don't see anyones "secret" formula working noticably better than what the normal person who knows what the heck is doing.

[Maggie]

"that's not how it was designed to work" :exactly:

Now that statement pretty much sums up the fine art of building a hot rod.
:jump:

[5.7ute]

I wonder if he is a member of teamZR1?:)
IMO there is many tables that need to work together to get the fuelling right for all conditions. To fudge one or two & then call it a tune is just ripping off yourself & the customer.

[Maggie]

I wonder if he is a member of teamZR1?:)

I was wondering the same thing, he doesn't sound contemptuous enough for teamZR1 though.


IMO there is many tables that need to work together to get the fuelling right for all conditions. To fudge one or two & then call it a tune is just ripping off yourself & the customer.

agreed.

[SSpdDmon]

I just thought it was 'funny' because I've heard the same arguements from someone else....almost word for word.

[joecar]

Jeff, you posted good responses in a scientific manner... :cheers: ...and he fails to respond to you in a like manner...

lol, it does sound familiar, but without contempt...

Having "privy-ness" to source code... that's like name dropping... :muahaha:

Experiments:
- If you alter the VE (g*K/kPa) table, what changes do you see in actual AFR, and why...?
- How can we "sanity check" our wideband...?

[ViolatorTA]

I'm still awaiting a reply............................................. ................

[Bruce Melton]

If you could run AutoVE "behind" the stock MAF table ie MAFed AutoVE then you could address it all. Everything we are doing to improve performance is aimed at changing VE as I see it.

[Doc]

Notice the new screen name.

Backup VE tables?

Stock Commanded Open Loop Fuel tables.

The 5gas analyzer, that one really was over the top for me.

I think you all are right...I think somebody got lonely and had to come back to Cartoonville.

Amazing steadfast resolve Jeff. The only problem is when you are trying to explain something to a group of people while having a goon drool on you...

This guy sounded like ALgore's Inconvient Truth.


Custom OS #5, LC-1, and the RR are all I need.

[hquick]

I immediately thought JR from TeamZR1 and I haven't even looked at the link yet. Going to have a look at it now.
I have to admit...JR makes some thought provoking statements.
Not long back on Monodax.com he was sprouting whatever but he did say most tune issues come from air leaks and the like. That inspired me to check over my motor. I found bolts were missing in the upper intake manifold. Fixing these made a big difference.....now, to go and read that link!

[dfe1]

I've been doing this stuff since the early 1990s, so I have different perspective on things because the early Mass Air systems had no MAP sensor and the early speed density systems had no MAF sensor. When you tune these vehicles, you have comparatively few "tools" because there aren't many options (relatively few tables). With the MAss Air systems, you have to work primarily with injector flow rate and PE fuel. There is no VE table in these systems, and while you can change the MAF voltage versus grams per second calibration, it's usually a waste of time because the MAF will only register up to 255 grams per second because of software limitations. More than 255 grams/second will flow through the meter, but the system can't indicate more than that. So once you peg the air flow meter, you have to add fuel through the PE versus RPM table-- in essence the system becomes speed density at that point. (There are some other tuning options, but they're usually too time consuming to be practical.)

You also have the issue that if you're running closed loop and the engine has a healthy camshaft, you have to alter the injector flow rate setting to compensate for the fuel that passes through the combustion chamber at low engine speeds (a result of valve overlap) and causes skewed readings from the narrow band O2 sensor.

With these relatively prmative tools and techniques, it's possible to get some very radical engines to start, run and idle properly. I used an early Mass Air system to tune a 411 small block that made 605 on the (engine) dyno-- through a descreened stock diameter Mass Air Sensor.

Obviously, the engine wouldn't have produced that kind of power if the air/fuel ratio was very far off. And it that was possible to accomplish that with no VE table and no MAF sensor recalibration.

That's a rather lengthy, somewhat off-topic explanation, but I think it adds some perspective to tuning. In this case, you had to "lie" to the ECM because you had no options. With the LSx systems, that's not the case because you have so many more options. But I think a lot of people miss the point that these aren't pure Mass Air systems-- they're hybrids. The system may run off air meter readings, it may use MAP readings to reference the VE table (speed density mode) or it may use a combination thereof. Anyone who thinks the VE table is of no consequence is using non-prescription pharmaceuticals. Try cutting the values in the VE table in half and see if you can get the engine to start and run.

I think the thread on LS1tech is incorrect and misdirected. If I'm interpreting it correctly, wide band readings are useless, and you have to use a 5-gas analyzer to tune properly. I think real world results tell a different story. If the wide band is reading properly, what you see is what comes out of the combustion chambers. I don't think it's relevant whether the air/fuel mixture entering the chamber is 10:1 or 20:1. If the wide band indicates 14.7:1, that's what the engine has processed, and that is what's relevant.

I think the experiments that SSpdDmon did pretty well document reality. As for "the best" way to tune, that's a matter of personal preference, but I've always found that fudging values is ill-advised unless there are no other options. GM spends a lot of effort calibrating air flow meters, and I from what I've seen, the calibrations are accurate and consistent. Why would you want to alter them? If the meter says air is flowing through it at 100 grams per second, and it is, in fact flowing at that rate, why change it? If you leave it unmolested, you have an accurate indication of the amount of air the engine is using, and that's worthwhile information. On the other hand, if the meter has been modified and is no longer accurate, it should be recalibrated. However, I doubt that most of these "experts" have the equipment or expertise to recalibrate properly.

Bottom line-- use whatever method you're comfortable with and provides the desired results. And take most of what you read with at least one grain of salt. I've found that a lot of "tuners" are very knowledgeable about software and computer systems, but don't really have a clue about engine operation. A trained monkey could put a car on a dyno and tune WOT. But he'd have a hell of a time getting idle and drivability right.

Now that I've rambled on for far too long, I'm done.

[dfe1]

One more thing-- Doc, you're right on target, I like your approach, and you are officially awarded a "smart-ass" attaboy. If you'd like your very own smartass smiley, PM me and I'll send one to you.

[SSpdDmon]

dfe1,

It's funny you say that about the MAF. I sometimes wonder what my SLP unit really flows....almost to the point where I want to disassemble from the air lid base back to the bellow that hooks to the throttle body and take that assembly to a guy here in town who has a flow bench for MAFs. You know...just to see where the differences lie.

The problem I have with leaving the MAF table stock (as I mentioned in the LS1Tech thread) is it seems the grams/cylinder calculation doesn't change. I show a max of .88 grams/cyl. with my SD tune, but only ~.73grams/cyl. with his stock VE/MAF, tune the other tables method. On a car with dramatically altered airflow characteristics making 25% more power, does it make sense that the same amount of air is getting in the cylinders as compared to a stock vehicle? I don't doubt GM does a lot of research on the MAF units. But they're observing flow on a stock lid, a stock MAF, and a stock head/cam engine. If you can eliminate the VE with B0120, then (assuming you can keep IAT's constant) you should be able to recal. the MAF to accurate numbers based on the AFR. If there are more influential factors within the source code of the PCM that I don't know about, well...that's my ignorance for you. But, I think the guys who designed our software did a great job by including all the descriptions for each table/cell that we can manipulate. If it weren't for those, I'd feel rather lost and unable to keep up with people like LS1curious who say, "I have the source code, I know what it's supposed to do, but I can't tell you any more than that."

[dfe1]

My point with the MAF calibration is that it measures air flow in grams per second. From that standpoint, it doesn't matter whether you're talking about a bone stock engine or one that's highly modified-- air flow is air flow. The grams/cylinder reading is a measure of load, not overall air flow, and it changes with rpm-- that is, the same amount of air flow (in grams per second) at different rpms will result in vastly different grams/cylinder readings. I built a conversion table way back when we could only monitor grams per second. If you like, I'll dig it up and send it to you.

As for the difference in grams/cylinder readings between the mass air and speed density calibrations, I'm pretty sure that's a result of the data processing that goes on in the PCM. If you take a car that's running normal mass air and convert it to speed density and make no other changes, the VE table will be way off. That's obviously because the system is interpreting data differently depending on whether the Mass Air sensor is in the system or not. Don't forget, in these systems, speed density is a back-up mode intended to be used in the event of a MAF failure. I'm sure some data handling is involved, but alas, I am not privileged enough to have access to the source code.

[Doc]

Aye captain, we be sailing the seas of cheese.

[joecar]

I'm still awaiting a reply............................................. ................He still hasn't replied to you...

[ViolatorTA]

He still hasn't replied to you...


I'm patient ;)