tryme96Z

I am encountering high LTFT on bank 2. LTFT at idle are 25%. At 3K rpms the LTFT on bank 2 drops to acceptable trims. See attached screenshots.

From my research this points to a vacuum leak or an exhaust leak before the o2. I tried the propane method of finding a vacuum leak around the top side of the engine and was not able to impact the LTFT. I did check the vacuum lines to the Opti also.

How can I diagnose a leaking header gasket on bank 2? I read that people typically listen for a pinging sound, but I cannot distinguish the exhaust "ping" from the noisy 1.6 roller rockers.


1996 Z. Manual.

Engine
Arizona Speed & Marine 58mm Throttle Body
Port & Polished the MAF sensor
throttle body by-pass
!CAGS
SLP Cold Air Induction
Comp Cams 306 cam (230 / 244, .510 / .549 112 LSA)
LT4 Timing Chain
Port and Polished Heads
Mr. Gasket Head Gaskets (.026 thick)
Port and Polished Intake (and 58mm Throttle Body ready)
Comp Cams 1.6 Roller Rockers
Comp Cams Hardened Push Rods
Comp Cams "R" Series Hydraulic Roller Lifters
Crane Dual Valve Springs
Comp Cams Guide Plates
Taylor Pro Spiral 8mm wires
1LE Elbow
LT4 Knock Module
Fuel Pressure Regulator (T.P.I.S. Top Hat Kit)
CSI Electric Water Pump
MSD 6AL ignition box

Exhaust
Hooker Long Tubes with JetHot 2000 coating
Pacesetter Y-Pipe
MagnaFlow 59959 catalytic converters
Hooker Aero Chamber Cat-back


Suspension
Eibach Proline Springs
T.P.I.S. Strut Tower Brace
Boxed PanHard Rod
Spohn Adjustable Torque Arm

Injuneer

Appears that the 3,037 RPM data is with the engine revved, but not under any load at all (vehicle is not moving, VSS=0). That could put the PCM in a cell that is not fully developed, because you don't necessarily drive under those conditions, at least not very often.

The correction at 1,031 RPM may be even worse than it looks. The LTFT and the STFT are both multipliers. In effect, the LTFT is at 160 (160/128=1.25), and we don't know the STFT. But once the LTFT hits the upper limit (+25%), the STFT will need to increase if it's still running lean (true lean, or false lean). You need to look at the STFT at the same time you look at the LTFT.

There would have to be a huge vacuum or exhaust leak specific to Bank 2 to require 25% extra fuel.

Still thinking you have a bad injector, or a lot of misfires.

Just for reference, for anyone looking at the background on this, and discussions to date:

https://www.camaroz28.com/forums/exh...0/#post6994060

tryme96Z

Engine is not under load. Vehicle is still elevated from the exhaust work. Keeping it high off the ground has been useful to keep an eye on the cats to ensure they do not get to hot. Could take it down if I need under load data.

More screenshots with STFT. CSV of core data points attached.

I will test the injectors next. Based on other posts (example: https://www.camaroz28.com/forums/lt1...95/#post779043 ) it sounds like unplugging the suspect injectors one-by-one and watching for a change is the best way to do this.

Injuneer

There are 18 LTFT cells. You are probably only seeing two of them (16 = idle; 07 = 30-50 kPa MAP, above 2000 RPM, unless tuner changed cell boundaries). It's beneficial to see as many cells as possible, including the high load cells - 12 thru 15.

MAF reading at idle looks high. 1.7 #/MIN converts to almost 13 grams/second. That's about 2X the normal reading for a stock engine at idle. Evidence of the impact of high cam overlap, as is the 16 to 18"Hg MAP readings with 0% throttle. As you increase throttle/RPM under no load, the effects of the overlap disappear, dropping MAP to 9.8 - 10.4 "Hg.

I need more time to look at the log.

PS: what data logging software are you using?

tryme96Z

http://www.obd-2.com is the scan tool. Back in 2002 is was a solid tool for the price. These days there are far more options for far less. I have the Android Torque app too, but find the desktop tool more powerful.

I can collect many GM specific data points not in the log posted. This is due to those additional data making the log very hard to read (creates many many duplicate rows for a single reading). So if there is a pid needed, let me know and I can collect it.

GaryDoug

Fred, I have never seen any reference to fuel cell numbers in OBD2 documents. As far as I know, there is no data (SAE or GM specific) to be extracted to give such a number. Are you sure OBD2 uses fuel cells?

Injuneer

To my knowledge, the OBD-2 fuel control strategy for the (96-97) F-Body is identical to the OBD-1 strategy, with respect to the use of fuel cells, the number of cells, application of cells, and the boundaries. Back in the days of AutoTap, Auterra, and obd2.com, I never gave it a thought, because the fuel trim data looked the same as OBD-1. The only change that was made was to change the nomenclature from "Block Learn Multiplier" to "LTFT" and "Integrator" to "STFT". And, rather than use absolute numbers (108-160) as the data output, they converted to +/- %.

I have not looked at an LS1 data log, but I suspected that they increased the number of cells. That is confirmed by looking at the "Scan Tool Data List" in the Engine Controls section of the factory service manual. I looked at the 2000 manual, page 6-1384, and it shows:

Parameter: Fuel Trim Cell

Data List: Eng 1, FT

Parameter Range/Units: 0 - 23

Typical Data Values (Engine Idling....) : 16, 17, 20.

I also checked the 1996 manual, page 6E-447 and the cell data is similar to the 1994 manual.

Final check was "How To Understand and Modify Corvette Fuel Injection & Electronic Engine Management" by the late Charles O. Probst, GM Part # 12497977, and his discussion of model years 1986 - 2001, and he shows "simplified examples" of the "memory cells". In his example, there are 16 cells defined by increments of "RPM" on the abcissa and "Load" on the ordinate. Unfortunately, this book has so many errors that I have found - e.g. switching the definitions of open and closed loop, alternatively switching the definitions of LTFT's and STFT's, misquoting the ranges of the OBD-2 % LTFT's, etc.) that it is not the best reference for much other than the actual GM graphics that he includes.

GaryDoug

Yes, I see it now. In case the OP might want to extract that datum, it is PID 1190. While we are at it, here are some more for a 96.

tryme96Z

Was not able to find PID 1190.

Fuel pressure at the fuel rail was a solid 45 psi at idle. Dipped to about 42 psi when revved.

Unplugging injectors while running to observe a difference seemed to yield results all over the place in regards to STFT. Sometimes STFT jumped to 30, then another time it jumped and went back to 0. On the last try of the test STFT jumped to 50 and would not come down, even once the selected injector was plugged back in and no codes were present. A restart of the engine started a downward climb to the "typical" (wrong) STFT. My assumption is that my test method was wrong.

Learnings:
1) I am starting to believe when the heads, cam, exhaust were installed many years ago...it was likely running poorly then. Exhaust smell being the primary evidence.

2) I do have some pinhole leaks in the passenger side header prior to the o2 where a patch was welded into one of the primary's due to a dent.

3) There is a noticeable ticking noise on the passenger side that does not exist on the driver's side. My attempt to get an audio recording that captures the difference between the two sides failed.

Injuneer

Fuel pressure arcidle should be 43.5 PSI with the vacuum compensation line off the fuel pressure regulator. GM accepts readings between 41-47 PSI as within spec. When you reattach the vacuum line, the rail pressure should drop proportional to intake manifold vacuum. In your case, pressure with vacuum line attached should drop about 6 or 7 PSI below the "no vacuum" pressure.

When the throttle is opening, pressure should rise. When the throttle snaps closed at higher RPM, pressure should drop. It would even drop a few PSI when you are holding it steady at 3,000 RPM, due to your 10 "Hg MAP.

I did notice in your short data log, that as it returned to 1,000 RPM idle, even the left LTFT started to increase, reaching ~ +8.x %. Reversing the negative LTFT at steady 3,000 RPM. Other than that I'm not seeing any clues to why you are maxed out at +25% on the right side,

When you were checking for a vacuum leak, did you look around the bottom of the injectors where the o-ring sits in the manifold?

The pinhole leaks before the O2 sensor could be contributing to a "false lean" raising the LTFT. Try wrapping it with some muffler repair tape.

tryme96Z

Problem still not identified. However an interesting change in the LTFT. After removing the throttle body for a thorough cleaning and a new gasket, the high LTFT has moved from the passenger side to the driver's of the engine.

Since this happened I have taken the TB off again and dissembled/reassembled to ensure all gaskets are seated and the IAC solenoid was in properly. I read some threads on here about a larger IAC passages possibly being necessary for aftermarket throttle bodies and how intake idle air goes through a separate passage in the manifold from the IAC. So I also opened trimmed the TB gasket around the IAC intake. I did not drill any holes in the TB to enlarge any passages.

I have also noticed in warmer weather the high idle comes down faster on its own.

My current action item is monitoring the MAP.

Note: Throttle body is 58mm from Arizona Speed & Marine. Maybe these are known for idle issues? https://www.camaroz28.com/forums/lt1...2/#post2545361

Injuneer

Not a lot of feedback on the AS&M TB's.... not many people use them because they were fairly expensive. But where there was feedback, the AS&M was judged to be one of the better aftermarket throttle bodies. Not sure if it had the "split LTFT" problems that most of the other aftermarkets seemed to have suffered from.

What are the new values for Cell 16 LTFT?