Just thought I would stir up the pot a bit. I noticed this when I wrote my BPW article but didn't mention it then in order to avoid that post turning into a MAF/SD debate. However, I thought that I would take the time to post it now ... after returning from the National F-Body Event in Indianapolis and having to read through the code in order to search down another problem.

Here is the original post and I would recommend everyone read throught my first 2 postings in this post...

https://www.thirdgen.org/techbb2/sho...hreadid=214156

What I didn't say in this post is that when calculating the Grams of Air per Second the ECM will limit the GMS/Sec to 255 if the actual calculated value is over 255. This is very important. Basically, you run into the same problem that the MAF cars have. You lose resolution. For those in doubt please review the code starting at LCC6B. This is extremely important in cases where you are running a big cubic inch motor. If during part throttle the ECM calculates that you have more than 255 g/s of airflow then it will limit the airflow to 255 g/s.

This has probably been known by the long standing chip guys for awhile .... and hidden ... but I wanted to bring it out into the open. I especially wanted to bring it out into the open for the MAF guys.

Tim

 

not trying to start anything but I mentioned that years ago. IMO 255 bits is nothing, the maf sensor's restiction is why I switched to map.....

 

I'm not trying to start anything either but if this was mentioned years ago then it would have been used in the MAF vs. SD debate. Guess what - it never was. Furthermore, your experience with reading source code is, well, little to none ... especially years ago (my previous thread on BPW illustrates this). The only way you could have mentioned it was if somebody told you about it - and I'm struggling to figure out who would have told you. In any case, it's important for me to bring it up and make a point out of it since it has been hidden for too long.

Tim

 

is this limit in the $58 code as well?

 

And with all the proding done by some, the MAF guys, still haven't done anything to really investigate how to get around the 255 issue, up to and including now, from what I've seen.

You act like it's some conspiracy or something, with using the word HIDDEN. Nothing's been HIDDEN, it's all in the code. The 255 is a limit?. For anything done with math there is sometimes a limit. There is no such thing as an infinity large number. Sometimes a rounding has to be made, which can be seen as a lose of resolution, but the trick is to do it so there is a min lose of resolution, or done in such a way that it really doesn't even matter.

So in bringing this all out into the open, have you thought of an answer for the MAF guys?, or are you just truely trying to stir things up, without having any definitive answers?.

The answer was mentioned on another list, about the time this list got started. The info., is that old. At least for the fueling part. He failed to follow thur on how to fully impliment it, IMO.

Like I've said for years, the answer is in being able to understand and in some cases modify the source code. Nothing new, and nothing's been hidden. It takes alot of work to comment a hac, and get it to assemble, and I fully understand why the guys that do it, don't want to share the months of work it takes to develope Source code. Especially when so many are willing to take their hard work and make a buck off of it.

The 255 limit, is, only if you let it be.

 

Guess it depends on the ecm used. In my 8063 its used mainly in the PID algorithm for the O2. As a matter of fact, its limited to 64 gms/sec. The fuel/spark and other related lookups are done purely by MAP. I dont think its as much of a problem like it is with the MAF.

 

Interesting. Didnt know that, as if you didnt know I didnt know that.

On a related note, when it does hit this fictitional limit, does it add fuel based on PE as well to compensate for what it doesnt understand?

 

I am trying to stir things up in terms of people's thought processes ... not in terms of getting people to be argumentative. Have I thought of an answer for the MAF guys? Sure - Rewrite the source code and convert over to a 16bit value. Have I thought of an answer for the SD guys? Sure - Rewrite the source code and convert over to a 16bit value. However, as is widely known, this is easily said but not easily done.

By the way, Thanks for turning my informative post into an accusation where you are turning the tables and putting the focus on me with regard to me doing stuff for the MAF guys ... or, for that matter the SD guys since THEY are straddled with the 255 limit also. You could have just said ... 'cool - the general public is finally getting the full picture'. But instead - you speak negatively about me because I am mentioning information without presenting some sort of solution ... as if it is wrong for me to present data without a solution. Stop being negative toward me and maybe try working together. Pessimism and negativity will only get you so far.

Tim

 

I tend to disagree. The BPW is calculated via ...

BPW = GMS AIR / (AFR * Injector Flow)

GMS/AIR for MAF is the value calculated via the MAF voltage.
GMS/AIR for SD is the value calculated via the Ideal Gas Law.

No matter how you look at it ... it could be limited to 255 for each system.

Tim

 

I'm not totally following the question. Can you ask in more detail? PE Fueling is only added in PE Mode. That's it. Thus, if you are not in PE then this 255 limit can prove to be a problem ... but only with bigger cubed engines. HOWEVER, I don't think it's problem given what I have seen with the bigger cubed engines so far.

Tim

 

AFAIK, its not used in the fuel calcs in my mask, just in the handling of the O2. At least in my tbi ecm, its sort of a guestimate anyway since its based on pressure readings. Again, different masks use different things. Ill take a look again, though, just to be sure.

 

Not a problem at all. The SD airflow value is NOT used to calculate fuel. It is just a calculated value used for other stuff. This is not a SD limitation.

Scot

 

What ever.
I'm not going to bother to respond to the none prom stuff you've mentioned.

But as far as the 255 problem, no total rewrite is necessary, searching thur the net, will reveal the answer, abiet, thou, IMO, it would need further work.

But, 255 aside, the MAF systems of the 3rd gens still have issues. But,with source code, they can be addressed. With some clever patching, the ultimate system is possible now.

 

Scot,

THANK YOU. Thanks for approaching this in a manner that isn't negative ... but rather - very much to the point. I now clearly see what you mean. L006F vs. L006B. THANKS AGAIN!

Tim

 

So are you saying it is not a limitation? If it is how do you tell if you are hitting this limit, can you tell by just dataloging?

 

For some reason I feel as though this was a test and we failed.

 

You answered my question anyway.

Better than GM making the same 'mistake' twice.

 

Where could I find an overview of this issue?
.
Is 255, (8 bit) the max count of the A/D converter for the MAF?
Is one count, always equal to exactly one gram?.
........................................................
Is this correct?
Maf output 0 to 5 volts analog.
Where is the A/D converter

 

NO. The airflow calc doesn't work at all on $58 if you select 8 cyls, and it has nothing to do with fueling.

In $8D, I think, If I'm reading the hac right, this "255 limit" will effect some egr operations. But, I may be wrong.

-- Joe

 

It's just the way things were designed.
It's like saying a calculator with a 8 digit display is limited. While in fact a calculator with a 16 digit display is also limited. 4 divided by 2 will be displayed/calculated perfectly with either, so it's just the rounding, or overflow areas that have to be dealt with.
Most every system has some limit or shortcoming, trick is to figure out how to min., the shortcomings impact on actual operation.

In steady state testing, the MAF will read a given way. So in that reguard it's constant. Trouble is that once you modify the engine, the actual VE of the engine changes, and reversion can be an issue. So while the meter is the same, the indicated airflow might be in error. And in some cases, like the 89 TA's with the turbo V6, they can peg the metter.

In an application where the engine uses closed loop, there is alot of correction possible, so in the oem application things work out OK. And for some people, OK, is close enough. Now when you run a big motor or something with alot of overlap, and drastically change the intake tract design, or MAF it's self, then you can get to where the system isn't very accurate. Which, IMO, kind of defeats the purpose of EFI. There are work arounds for getting it all correct with MAF systems, but it's nothing easily done. Just to mention one, is that you need to measure air flow that the engine is actually using. So with a large plenum, you have some air flow that is only used for changing the relative vac level in the plenum, and isn't actual air flow that the engine is consuming, so there is a false rich reading as you crack the throttle open. It takes a fair amount of modeling to correct that. So in some applications a false rich off idle kinda covers other issues. But, in some it'll cause problems. From what I've noticed, most everyone that's converted to MAP systems has mentioned better throttle response, and this just might explain part of it.

Then we can also look at the airflow reading errors, and corrections needed for say, the PCV, CCC, and EGR (both forms).

There's been a few threads here about the later MAP/MAF systems that pinpoint what GM's done towards correcting things.

I guess....

 

Wrong!wrong! wrong!!!

I did mention that the map system is limited to 255. In fact it was a debate on maf vs map. Ya know...the one with SDI in the thread. Oh wait, every maf thread with SDI was a debate.. Actually I was looking for the thread but it's a pain to go though pages of arguements.


It doesn't just have to do with source code. You attack my code reading ability. BUT I can read it...just not that well. Kind of like learning a new language. I can understand it but not good enough to do much. It did help me decide on making changes to the maf tables. That was years ago...(maybe 2-3yrs).

Nobody told me about the map systems limit....


I could find the old thread if you insist.

 

I think one reason the throttle response is better is because I don't have a 2 5/8" restiction before my TB. What's the point of having a big TB if the intake tract is smaller.....

 

The 255 is NOT a limitation on SD systems. I should delete this thread so that people don't get cornfused. But, I won't since it is important for me to show newbies that I make mistakes too.

<font color="RED"><H2>Speed Density Systems are not limited on airflow based on 255.</H2></font>

My response to Scot holds the answer

 

I went back and read my old thread. I didn't say it SD was limited to 255. Someone else said that. BUT I did say the 730 is limited in the same manner as 165. Because map and maf both are 5v. Both will read 5 volts at wot.(my conclusion was that both sensors had the same resolution because both are 5 volt) Both systems, wot is just a calculation. I think there use to be a myth that (WOT) SD tuning was more complexed than maf. Really it's the same thing. The 730 is just slightly more advanced for wot, if that.

I think the maf wars are long gone. There was only a handfull of diehard guys. I was not one of them. Sure I defended the system but always knew I would change over. No way my engine was going to be stuck breathing through a tiny straw....

Besides, my assembly lang skills are not so well(as you pointed out).



you gave up that right guess you do still have some pull

 

At small throttle openings, ie where you'll really notice the better drivility of a particular set-up, the MAF's resistance is minimal.

If you want a challenge, try optimizing your intake tract. Try, 2.25, 2.5, and 3" tubing, from the air filter to the TB. And then with several different sized TBs. While the 3" tubing, and 3" TB on my car was kind of neat, it really wasn't any faster. With the big piping, the air flow just off idle was great, but after about 1/3 throttle there was poor throttle modulation. The throttle become more of an off-on switch. The 2.25 was nifty but just down of HP at the upper levels. So I've chosen the 2.5 tubing, with a 62mm TB, so as to be able to have the best average.

And if anyone's running a MAF it does matter where it's located within the piping. Especially if your trying to run screenless, or play for getting the most accurate readings. Again, moving it around and a careful eye for what it reads will show when your onto something.

 

And to contiune the "TRAX is wrong" ....
Most if not all of the c3 TBI (speed density of course) don't use the Ideal Gas Law for fuel calculations. Like Dim said, we're also limited to a calculated 64gms/s. Believe me when I say that this number is usually pegged yet the sync, async, and qsync are all still calculated based on the VE, BPW and other values. I just learned this weekend (even after reading the source code) that the BPW in the c3's can limit the final calculated injector pulse width! That's because the injector constant (aka BPW) is multiplied to get a starting pulse width which is then subtracted from through table lookups. Hold on... what about the injector bias. Is that added in after.... no, sorr about this, I need to go look again to get the facts straight. In other words.... this stuff is CONFUSING.
So let me get things straight. Past a MAF's 255 limit it has no other way of calculating fuel delivery but if in PE mode there is the AFR table that adds additional fuel!?!?! Sounds like a major problem for large MAF systems.

 

The newer MAF stuff isn't limited to 255, like the 3rd gens..
It'd be interesting to see if the bad ZO6's eventually peg their MAFs.

FWIW, there's some 89TTA setups in the 10s using the LS1 MAFs, and still not pegging the MAFs.

 

I didn't mean what I said with regard to the C3's. I was talking about the 730 ... but, it doesn't matter anyhow since I was wrong about that one too

Tim

 

I was saying stock MAF but since you said it, how is an 89TTA in the 10's without going over 255gms/sec? That's not a lot of airflow for a 10 second car.

 

*********
FWIW, there's some 89TTA setups in the 10s using the LS1 MAFs, and still not pegging the MAFs.
*********

They're using the LS1 MAFs that go to like 400+, gm/sec.. The TTA use the Bosch Freq MAFs, and then a *Translator* to go from KHz to the Bosch Low Freg..

Looks like yahoo don't like me, try emailing me if you would.

 

I thought that's what you ment but wasn't sure. As for the e-mail, it works, I just replied.