[Speargun]

EFI experts, step right up!

tl;dr:
Engine is running pig fat & will not lean out.
None of the usual causes found.
Same problem on 2 different Fitech units & 3 different O2 sensors.


I put a new engine in my car a few weeks back and it has been running rich, although it drove okay. I pulled the plugs yesterday and they are fouled pretty good so I put new plugs in. It's now running so rich that if I rev it at idle, it will smoke out my shop. The new plugs were already fouling and were wet.
I did a complete reset on the 1200pa unit thinking that maybe it was still set for the old 469cid it was on, but no change. I also swapped out the O2 sensor (which was covered in smut) with one I had from another system I had. (It was used so no guarantee it is good.)
The only thing that I can think of is that both O2 sensors are bad. Am I missing something?
Here's a link to my log file not long after the reset.
FITECH LOG FILE

Here is a short video of the car running.
https://youtu.be/R_RFrNt8x0s

Pontiac 400, 9.5:1, Voodoo '702' cam, #16 heads, HSD intake.

More info.......

So I put the new O2 sensor in and still have the same problem.
I can't find any vacuum or exhaust leaks anywhere.
I removed the PCV hose & capped the port.
I have adjusted the IAC using the front & rear screws. (Maybe 1/8 turn each)
The AFR is holding close to the the 14.5 target.
The AFR learn is at -34% and the AFR trim is between -10 to 5. iirc
Fuel learn is also in the negative 20's or 30's. (Maybe more)
The cam is set to 1 & I have 13" - 15" of vacuum at idle.
I reset all of the learn values and it still gives the same numbers.
.
I'm running NGK heat range 5 plugs, which should be fairly hot, gapped at .042".
The ignition is an off-brand billet "Ready to Run" HEI with a MSD Blaster 2 coil.
*
*** If I understand it correctly, the fuel learn is pulling fuel because it thinks it is getting too much; which it is. If there were a vacuum leak, wouldn't it be adding fuel because it thinks it's not getting enough? And wouldn't the AFR learn & trim be on the positive side?

Here are some photos I took this morning of some of the settings with the engine running, but still not warmed all the way up.
Fitech Setting Photos

[Lee]

Tach signal coming from the HEI? Is your indicated RPM bouncing around a lot, possibly a LOT more than what the engine is actually fluctuating? If so, then wire in some sort of CD ignition box - let the HEI trigger the box, and let the box supply the tach signal to the EFI.

[Scott65]

The vacuum leak question depends. How is your fuel system set up? If it's vacuum referenced, vacuum leaks will really mess with you. Vacuum is decreasing fuel pressure when working correctly. When not working correctly, you're running with more fuel pressure. This "could" be what you're dealing with. A blown diaphragm in a vacuum referenced regulator will REALLY mess with you. Been there done that. In that case it's blowing fuel through the vacuum port into the engine. Check the vacuum connections for evidence of wet fuel for this. Of course with the evaporative rate of today's fuels if you haven't run it recently you'll need to run it again to check for this.

[djustice]

Hmm, What does fault code 76 translate to in the Fitech manual?
Z-column is filled with uninterrupted 76 fault.

[Speargun]

Originally Posted by djustice

Hmm, What does fault code 76 translate to in the Fitech manual?
Z-column is filled with uninterrupted 76 fault.

Fault 76 = FAN 2 OPEN
It was trying to turn on a fan that wasn't there. I shut it off in the setting later.

[JLMounce]

Based on the log, your read AFR and your commanded AFR are pretty much inline. However, if you look at your AFR learn, compared to the trim, you see a pretty big discrepency happening. Your Learn is in some spots at it's maximum negative, meaning how the system has learned is that it needs to pull out fuel to meet it's commanded AFR. However your trim's are running positive, sometimes by a decent bit, which means your short term tables are having to add fuel.

There's honestly only really one reason why this would happen and it's an exhaust leak upstream, or very close downstream to your oxygen sensor. The o2 sensor is seeing fresh oxygen and making the system think it's lean, so it's adding fuel to compensate.

Fault 76 pertains to fan2. If you don't run a multi-stage fan, or two fans, you can disable fan 2 in your controller. In this case it's of no bearing on your current situation.

General rule for the oxygen sensor placement is as close to the main collector as possible to see the maximum number of cylinders, in as straight of a pipe as possible with a minimum of 18" of pipe behind it before exhaust exit to atmosphere.

Look for leaks in the header flange, and tube connections, at the oxygen sensor bung, at the gasket between the header collector and intermediate pipes etc. If you are still using the supplied sensor plate with clamps, you need to go get that welded. The plate provided is designed to work with a 2.5" OD pipe. If you're clamping it to a 3 or 3.5" header collector, it's not ever going to seal because of the radius difference.

Additionally, considering your learn values are near or at their max, towards lean, I would consider changing your cam selection up by 1 number. 3 if your on cam file 2 for example.

[Speargun]

TACH signal is solid & the tach wire is shielded.

Fuel system is an internal Walbro 255 lpr with a -6 feed & return.
Fuel pressure is rock solid & is not vacuum referenced.

Everything says vacuum leak, but I can't find it.
Is it possible that it's in the ignition system? The distributer is a chicom knock off ready to run type with a MSD Blaster 2 coil.
Everything APPEARS okay with the ignition system, but maybe it's worth looking at....

[Formulajones]

That sure seems like it's 02 related.

Your learn and fuel trims are out of whack. Like the 02 is seeing an air leak.


Crap, Jason beat me to it.

[JLMounce]

Vacuum leak after the unit will not cause a fouled plug, there's unmetered air entering in that case and you'll more than likely read a lean plug as a result.

Ignition related issues failing to light the mixture will cause a rich read at the oxygen sensor.

My course of treatment at current would be the following, in this order.
1. Verify and correct any leaks upstream or near downstream of the oxygen sensor
2. Replace plugs
3. Clear all learn values
4. Move to Cam profile 2
5. start the car and let it idle for 10 minutes or so to get a decent idle learn baseline.
6. Check the plugs and pull a log

[Speargun]

Originally Posted by JLMounce

.....Look for leaks in the header flange, ...

Vacuum or exhaust is really the only thing that makes sense, but I just couldn't find a leak. I guess I need to do a more detailed inspection......
It's always something simple, but maybe "I can't see the forest for the trees."

I have Remflex flange & collector gaskets. The O2 bung is welded, located behind the collector +/-18" & clocked at the 10/2 position with a full exhaust behind it.

[Tom Vaught]

I have seen Steve Morris and others clean O2 sensors that were acting erratic and yet looked fairly clean externally.

Before I did more I would learn how to clean the O2 sensor properly vs assume the
"NEW" sensor (actually a used different sensor vs a real box stock NEW sensor),
is the issue.

Tom V.

[lfdsteve]

Fwiw I have the Powejection 3 unit from Retrotek and despite all attempts to convince me otherwise, the near identical look along with the same people with the same names in the Sema booth tells me it’s a next gen unit with new branding to get away from old deficiencies and complaints (my opinion anyway)
However, I digress. I had a similar situation with constant fowling of plugs and terrible fumes of rich running, especially at idle that I just couldn’t “tune out”
Last month I finally found a welder that could weld solid my clamp together 3 inch pypes system and after several weeks, seems to be cleaning up. Fwiw

[JLMounce]

Originally Posted by Speargun

Vacuum or exhaust is really the only thing that makes sense, but I just couldn't find a leak. I guess I need to do a more detailed inspection......
It's always something simple, but maybe "I can't see the forest for the trees."

I have Remflex flange & collector gaskets. The O2 bung is welded, located behind the collector +/-18" & clocked at the 10/2 position with a full exhaust behind it.

Just to clarify here, your oxygen sensor is located about 18" downstream of the collector?

What kind of exhaust system is on the car? Is the sensor close to any clamped together sections that aren't welded? Does it have a cross-over? How close is the oxygen sensor to a possible cross-over?

The further away from combustion chamber that the oxygen sensor gets, the slower it's response will be to changes and accuracy can diminish as the spent gasses cool. If the sensor is sufficiently aft in the exhaust system, there may not be enough heat to keep the sensor working properly and accurately.

If you're 18" back from the collector, I would consider patching the hole you have and moving the sensor into the collector itself, about an inch in front of the collector to intermediate pipe flange. This is the optimal location for a car with headers so that it sees 4 cylinders but is also as close to the combustion chambers as possible.

[HWYSTR455]

A dirty RPM signal can cause this also, since it keeps trying to sync. It doesn't matter if the wire is shielded if they are right on top of each other. Coil wire is a usual cause.

You may have the throttle plates open too far, or the IAC is boogered (look at steps).

Pretty sure if you turn the screws on the throttle blades, you have to cycle the power in order for it to do a TPS reset.

I have seen where condensation/moisture in the exhaust as it warms up blows right on the O2 and fouls it. You can try a Innovative thermal isolator if that is suspected. Some insist on installing those right off the bat. It will also prevent thermal shock. I just went thru this exact issue on a friend's car.

https://www.innovatemotorsports.com/...-p-n-3729.html

If the tailpipes are dripping, and the O2 is in the collector or right behind the collector, it could be getting a direct blast and fouling the O2.

We tried 3 new O2s, and properly cleaned them, none worked right after they were fouled. He pulled one from another car and installed an isolator and boom, solved.

I sanded the included copper washer to get the port aligned properly.

Oddly, when an O2 fouls, it reads lean most of the time. If you look further down in the log you can see it adding fuel. Since it's not reading right, it's going to keep throwing fuel at it.

.

[ponjohn]

Is it possible there is bad supply voltage and / or bad grounds?

[Scott65]

An observation... Does anyone find it odd that between ~1700 and 2200 the learn and trim values level off close to normal? Also a question, why is the target rpm always 831? Keep in mind I know nothing about Fitech.
ETA Guess I should specify, row 69 through 130. Then it goes back to lots of correction.

[70gtojosh]

What is your fuel pressure and what should it be? Is it possible it is too high? Which would cause a rich condition.

[Speargun]

Thanks for all of the input so far guys!

Things I have done since I last posted:
Moved O2 sensor to the passenger side. (Welded bung about 3"-5" after collector. It was about the same distance on the drivers side.) Triple checked manifold & collector for leaks, but none found and still was running pig rich.

Since that didn't work, the only place that I could think of that there could possibly be air getting into the exhaust is the crossover on the intake. This evening I pulled the intake and found that the block off plates that a buddy made out of a monster can had completely burned through.
Found some smut on the EGR block of plate.
Swapped out the HSD for a T2 intake and installed the block offs that came with the Edelbrock gaskets.

Initial impression is that there is no smoke at idle. I didn't reset any learn values yet but after a few minutes & a few tweaks, it started to idle better with a solid 14"-15" of vacuum.
I tried to record a few logs, but nothing recorded so I'll have to try again tomorrow, but what I can remember is:
AFR trims mostly between -10 to +2
AFR learn -26 ???
Fuel learn Idle 60%-64%
The other fuel learns were around the same, or a little higher, once I ran up & held the rpm at a few spots.
Cam selection 2
It smoked like a freight train once I pushed the rpms up.

I "THINK" I may have fixed the exhaust leak and am now just dealing with dirty plugs and a screwed up fuel table. If I understand it correctly, the fuel learn below 100% is pulling fuel from the table and values over 100% are adding fuel. What is a good range for this?
AFR learn should be close to 0, right. Is +/-10 a good range?

I'll probably take it for a drive tomorrow then see if it will record some logs....

[Scott65]

I wouldn't want it to be correcting any more than 10%. I would think 5 or less should be possible. Do you have a base tune you can start over from? Are there any diagnostic measures to go through to verify you don't have a hardware problem? I swear it sounds like you're pushing un metered fuel into the engine.

[Formulajones]

Like Scott said, and from what I've been told, if you have fuel trims + or - 5% or less that's good enough to lock it down and run with it.

I will say it took me a while to get my trims that good, and most of mine are now + or - 2%, some are zero, but that was several hundred miles of driving and lots of transferring learn to base and then smoothing the tables by hand. It took some time.

I'm not familiar with Fi-Tech but maybe a copy of the tune file might help shed some light?? Maybe someone will spot something unusual.