Turbo Buick Forums banner
1 - 20 of 25 Posts

·
Registered
Joined
·
103 Posts
Discussion Starter · #1 ·
Well, I thought I had this fixed, but it's back. Hi RPM WOT stutter especially in 2nd gear, even an occasional backfire. New plugs, new MAF (LT1) with Translator, new TPS, new coil pack, module only a year old, tried two different chips (94 octane and 108 octane), hi octane gas, lo octane gas, alcohol injection working fine, no anomalous readings on TLink, boost between 18 and 21 depending on gas (alky injection prevents knock at 18psi with 91 octane).... Nothing makes a difference. I've made no significant changes recently.
149K on the engine, but only about 15K on the valve springs. Leaks some oil but no smoke.
THIS THING IS DRIVING ME CRAZY!!!! Help me before I lose a race! So far everybody has fallen so far back they can't catch me on the stutter... TurboDave, you out there with words of wisdom?
Keith

Mods:
Fuel--Walbro pump, hotwire, Adjustable FPR, 42.5lb 009s
Air: 14" K&N, Ford Powerstroke FM intercooler, Turbonetics Cheetah Stage III turbo, THDP w/test pipe and cutout, single shot exhaust
Stock ECM with 94 octane 009 chip or 108 octane 009 chip.
 

·
Retired
Joined
·
1,966 Posts
Sounds like you may be suffering MAF drop outs. This is going to be a very difficult problem to capture. You'll really need something very high speed to see it (like Direct Scan).

What brand of chips were the 94 and 108? Not an Extender I assume.

An Extender chip would eleviate the dropouts
 

·
Premium Member
Joined
·
43,735 Posts
altho I hear about maf drop out quite often these days...especially since the extender was developed to ignore it, I have to wonder why cars ran okay before the solution was invented for the most part. :)

How old are the injectors?

Have you tried unplugging the cam sensor with the engine running to see if the problem is still there?
 

·
Resident Idiot
Joined
·
11,462 Posts
Changing plug wires is easy and cheap enough. This fixed a high RPM/high load problem for me....

powersix
 

·
Registered
Joined
·
1,838 Posts
My first thought was plug wires...
 

·
Retired
Joined
·
1,966 Posts
Steve Wood:
altho I hear about maf drop out quite often these days...especially since the extender was developed to ignore it, I have to wonder why cars ran okay before the solution was invented for the most part. :)

How old are the injectors?

Have you tried unplugging the cam sensor with the engine running to see if the problem is still there?
IMNO, the problem existed all along. It's just that only in the last few years have we been able to properly diagnose what was happening and it's cause (more ECM experts available).
Really quick/fast cars are becoming more prevalent these last few years, and when conditions are right, it starts happening more often than it used to.
 

·
Premium Member
Joined
·
43,735 Posts
I would suspect that if a bad maf that dropped a number of consecutive frames is present, then it might be noticed. Frames in the DS sense.

On the other hand, Keith is running a modern maf and is running in the low 12s. Unless the translator is prone to amplifying the problem, which I doubt, I would doubt that it is a very high percentage chance of being a problem. :)

All things should be considered, tho. Kinda hard for me to suggest expensive solutions until the easy things have been eliminated.

These cars can be very frustrating on any given day. :)
 

·
Registered
Joined
·
1,247 Posts
Steve Wood:
Unless the translator is prone to amplifying the problem, :)
:)
Steve: You master of innuendo... headbang I've noticed the same problems/comments over the past few years. Maybe that's why the EXTENDER CHIP was necessary...sorry...I mean dunno OPTIONAL... Just an observation. usa

I'd try to borrow an EC to see if that's it. yup
 

·
Fuhgeddaboudit
Joined
·
18,613 Posts
I would swap out known good coil/ module and plug wires first. I also got bad spark plugs new out of the box, 3 sets. Must have been stored underwater in some warehouse. Try a new set gapped at .032 from a new supplier. If that fails. With 150K miles and new valve springs I would do a compression/ leakdown test. You may have a cylinder not up to par. Also the new springs may have started to wipe out the cam lobes. What springs did you put in? If they are too weak it can cause popping. and to stiff you get lobe damage.
 

·
Premium Member
Joined
·
43,735 Posts
CaPPY:
[Steve You master of innuendo... headbang I've noticed the same problems/comments over the past few years. Maybe that's why the EXTENDER CHIP was necessary...sorry...I mean dunno OPTIONAL... Just an observation. usa

I'd try to borrow an EC to see if that's it. yup [/QB]
Innuendo? Huh? what the heck are you talking about?

Maf drop out is real. And it certainly can be a problem if incurred in large doses. On the other hand, people ran quickly for years without serious problems from the occasional blip. Therefore it is hard for me to tell someone they need a different chip until the basics are all eliminated.

The extender is the second chip that eliminates the problem, anyway.
 

·
Registered
Joined
·
1,247 Posts
<strong>
Steve Wood:
[qb] Unless the translator is prone to amplifying the problem, )
yup
If it did amplify the problem and the extender chip eliminates the problem (as it was designed to ignore the dropout)...should it have been NOT an OPTION! Ca-ching! dunno I read something that only my mind silly being so highly developed could conjure. eek! I "innuendoed" when I shouldn't have even "innued." nono

Tis folly to be sure! rant :D
 

·
Premium Member
Joined
·
43,735 Posts
I don't think the translator amplifies the problem....I think it just translates. I am surprised that the modern mafs would have the problem...and not knowing anything about ecms, I wonder why it was not a problem in LT1s/LS1s? guess, if it was a problem, they (GM) fixed it in the ecm?

At any rate, if he had DS and could look at the injector pulse widths, he would know if that could be a problem. And if he had DS, maybe he would come to my house and make mine work consistently, eh? :D
 

·
Registered
Joined
·
1,247 Posts
I know as soon as I say this rant I'm going to have a problem with it.

But, I went to an electronics store and bought this stuff that they use on electronics that are exposed to the elements. It seemed like a type of wax. I sprayed my ECU contacts and DS connector innards and the entire guts (what I could get to) and I haven't had any problems with my DS yet. It's about 3 years old.

Now I have to throw some salt or whatever over my shoulder, so I didn't hex myself. silly
 

·
Retired
Joined
·
1,966 Posts
Steve Wood:
I don't think the translator amplifies the problem....I think it just translates. I am surprised that the modern mafs would have the problem...and not knowing anything about ecms, I wonder why it was not a problem in LT1s/LS1s? guess, if it was a problem, they (GM) fixed it in the ecm?

At any rate, if he had DS and could look at the injector pulse widths, he would know if that could be a problem. And if he had DS, maybe he would come to my house and make mine work consistently, eh? :D
For your first paragraph, a very easy explanation is in order. Yes the Translator only translates. It doesn't much matter what MAF it's running, the limitations that cause MAF dropout are all in the ECM (which is why the LT1's and LS1's don't suffer the same problems, much newer and more sophisticated ECM's).
It can only recognize/count to a maximum of 255 (decimal) FF (hex). Therefore the MAF limit of 255. And when driving substantially MORE than 255gps (A lot more) through the electronics, the buffer in the ECM goes into "overload" (more correctly called overflow) and starts resetting its self, hence the dropouts.

This manifests its self as a slight bucking, and possibly a backfire or two in high boost, high rpm situations. The more air the engine is injesting, the more pronounced the problem!!!
Assuming everything else is in tune that is wink

It takes DS to see these occurances because they happen randomly and only last a few mileseconds, but when they happen you'll see several things occur all at once. MAF resets (very low flow), and of course this takes the LV8 down with it, and a few milliseconds later, you'll see the injectors close down for a few milliseconds. It's all very nasty, and happens too fast for an ALDL scan tool to catch (unless you're incredibly lucky)!!!

<small>[ April 07, 2003, 07:55 PM: Message edited by: TurboDave ]</small>
 

·
Premium Member
Joined
·
43,735 Posts
Okay, :D I understood until I started to think, then I had more questions.

Beginning with a stock ecm and conventional chip/oem maf, if 255 is the largest number that can be written, are we saying that the stock maf can actually send a number larger than 255 which would then fold over and the lower order bits would then cause the ecm to write a very small number? If this is possible, wouldn't it be reasonable to simply take any number larger than 255 that comes in and just write FF rather than 100 or whatever the larger number? Just a simple maf lock.....?

Then, secondly, if one was using a conventional chip/translator/late model maf, the translator has a limit position in the third dip switch which, if I understand correctly, prevents the ecm from seeing a number larger than 255. In this case, how does one incur drop out if the problem resides in the 8 bit A/D of the ecm?

I appreciate you patience! :)
 

·
Registered
Joined
·
1,838 Posts
I appreciate you as well Steve! I didn't know that patience was term of endearment however.

Ok then, good-buy for now sweety! :D

<small>[ April 08, 2003, 07:43 AM: Message edited by: Turbo_Tim ]</small>
 

·
Premium Member
Joined
·
43,735 Posts
headbang :D
 

·
Premium Member
Joined
·
43,735 Posts
Anybody? :)

Okay, I understood until I started to think, then I had more questions.

Beginning with a stock ecm and conventional chip/oem maf, if 255 is the largest number that can be written, are we saying that the stock maf can actually send a number larger than 255 which would then fold over and the lower order bits would then cause the ecm to write a very small number? If this is possible, wouldn't it be reasonable to simply take any number larger than 255 that comes in and just write FF rather than 100 or whatever the larger number? Just a simple maf lock.....?

Then, secondly, if one was using a conventional chip/translator/late model maf, the translator has a limit position in the third dip switch which, if I understand correctly, prevents the ecm from seeing a number larger than 255. In this case, how does one incur drop out if the problem resides in the 8 bit A/D of the ecm?
 

·
Registered
Joined
·
350 Posts
The code that interprets the pulses (frequency) coming from the MAF is in the embedded rom on the ecm board (not the removable prom), so it cannot be changed. However, airflow is limited in the hard code to 255 already and cannot be exceeded, so that is not the problem. I cannot tell you exactly what the problem is, but I believe it must be either a) the maf sensor itself flakes out. b) the pulse accumulator flakes out. c) directscan is causing the glitches.

Here's how it works:
There is a "pulse accumulator" that increases in number for every pulse that comes from the MAF. I don't know for sure how high the accumulator can count, but since it seems to be a 16-bit number, I would assume 65535. It is always increasing (or counting) since there is always pulses coming from the MAF (normally).

When the program gets to the "MAF processing" code, here's what happens:
1. The "old" MAF count (frequency) is loaded into memory. This is the count from the last cycle through the code.
2. The "new" MAF count is loaded into memory from the pulse accumulator.
3. The old is subtracted from the new and you get the DeltaMAF counts. This is the number used in the MAF tables that correlate to the grams/sec.
4. The DeltaMAF counts are checked to make sure they do not exceed 1791. If it does exceed 1791, then it is forced to be 1791.
5. The value is used to look into the MAF tables and get a "grams/sec" and also a table scalar.
6. The tables value and scalar are applied and we get XX grams/sec.
7. An MAT(airtemp) correction is applied.
8. Some delta airflow code is run to alter how fast the airflow can rise or fall (which doesn't do alot in the factory implementation).
9. The airflow is checked against the "Max Airflow vs. RPM" table on the prom. (It cannot exceed 255).
10 We're done.

When "MAF dropout" occurs, I see that the airflow usually drops to around 30-40gr/sec. This must mean that the pulse accumulator provided the wrong count. Why did it provide the wrong count? Thats the question. Maybe its somehow resetting like Dave said. But why only at high airflows? The accumulator is always counting, surely it must reach its limit many times under normal driving, yet we see no dropout when just cruising around. It would make more sense if the maf sensor itself was overloaded at high airflow and goes funky. I dunno. :)

Fortunately for us, the injector pulsewidth is calculated on the removable prom, so we can write patches that tell it to "ignore" the maf in certain cases.
 

·
Premium Member
Joined
·
43,735 Posts
Thanks, Eric, I appreciate the time and effort. :)
 
1 - 20 of 25 Posts
Top