Troubleshooting Help Needed

[mrblyoung]

Okay - back for round 2 guys!

Cars been down since the beginning of May and I am still banging my head against a wall over here.

So I was trying to take the car to a show and about half way there it started running a little rough and wouldn't idle. Ran fine at upper RPM's so I turned around and drove it home and just had to feather the gas pedal at stop signs. Had my laptop with me and everything on the EFI seemed to be working fine, all inputs were good but was just running rough down low.

So I get home and realize that my fuel pressure would drop as soon as the pump turned off - looking at the gauge on my regulator. Prime it would shoot up to 43 and as soon as the prime was finished, it would plummet back to zero. So diagnosing if it was the regulator or pump I had a buddy go into the test outputs of Tunerstudio, run the pump and I clamped the return line and told him to shut it off. Still dropped to zero, so bad fuel pump. Changed it and fixed the pressure issue. Back to normal, right? Wrong.

Still wouldn't idle. So the suggestion was that I might have a dirty or clogged injector. Lucky for me I had 4 additional Access 30# injectors (bought a set of 8 when buying them). Swap them in and now I cannot get the sucker to fire at all! WTF?! Swap the old ones back in, still nothing, barely a cough when cranking the motor. So I start looking to see if I am not getting spark. Had 4 additional spark plugs, so I go back into test mode in Tunerstudio and start testing to see if I am getting spark to all cylinders testing in pairs (wasted spark). Find out that both passenger side cylinders aren't sparking. Found that the power ground for my LS coils that grounded to the cylinder head came out of the connecter. Got that fixed and reconnected. Boom! Sparking all plugs now. Put the new injectors back in (I'm good at changing injectors these days) since I figured that might have originally been the problem. New clean injectors, fresh spark plugs gapped correctly and everything bolted back together. That surely would fix my problem, right? Still wrong.

Now the engine fires, but sounds like its only running on a few cylinders. Runs rough, and for some reason, now the wideband is saying its ALWAYS at 22:1. So I am not running it for long. So I figured maybe the wideband (LC2) is out of calibration. Go through the process of recalibrating it and still....22:1. Ugh. So I figure maybe I have a faulty O2 sensor. If you remember from earlier...I bought extra widebands so I first switch out the sensor, recalibrate and still....22:1. Seriously?! Okay, so I take out the LC2 and drop in an MTX-L which is basically a LC2 built into a gauge. That way I can confirm that the gauge on the wideband and the gauge on the Tunerstudio are seeing the same thing. Still 22:1 on both whether the motor is off (after having the key on for a minute letting the sensor heat up) or if the car is running. Sure the needle bounces around a little bit once the motor is started, but it doesn't go under 22:1.

Any ideas guys? I am pulling my hair out over here. What can I possibly be overlooking. The tune hasn't changed, it ran fine before. No blown fuses, no loose wires, everything is connected and registering on the laptop.

[Piledriver]

Your WBO2 is probably telling the truth.
If you have a second serial input you can run the innovate software and compare digital readout to analog.

Misfires read dead lean, as the oxygen doesn't get used if the fuel doesn't burn.

Replace the plugs on the holes that weren't firing at least.

Probably varnished up, may get away with a soak in berrymans b12 and blow dry.
Don't get it on paint or in a cut, contains acetone, isopropyl alcohol and MEK.
Excellent paint stripper.

Your WBO2 cal is a little off, with a few exceptions, WBO2 controllers max out at 20:1. and 10:1 on the low end.
Knowing that you can calibrate the analog input using the custom linear wideband option.
dead rich should read 10:1, fuel cut should read 20:1.

It's way too easy to concentrate on the electronics/sensors as they provide so much information, but I have found its usually a basic mechanical or electrical issue when things go weird.

[mrblyoung]

Replace the plugs on the holes that weren't firing at least.
Probably varnished up, may get away with a soak in berrymans b12 and blow dry.

Did that. 4 fresh spark plugs installed in the motor. NGK DR7EA's.

Misfires read dead lean, as the oxygen doesn't get used if the fuel doesn't burn.
Your WBO2 cal is a little off, most WBO2 controllers rail at 20:1.

That's what I was thinking. If no fuel is going into a cylinder, it's just acting as an air pump pumping air in and out and skewing the readings. Not sure how I should recalibrate it? Both of the LC2's I have tested on the car and this MTX-L, all with different Bosch sensors will go up to 22:1. Thoughts on how to recalibrate it to the 20:1 ceiling?

[mrblyoung]

It's way too easy to concentrate on the electronics/sensors as they provide so much information, but I have found its usually a basic mechanical or electrical issue when things go weird.

Very true. I don't think it's the EFI that's the problem. I already went though and did a valve adjustment, checked spark and injectors. Trying to determine what else I could be overlooking. It's not that complex of a system. Something small is just causing it to not operate correctly.

[Piledriver]

Thoughts on how to recalibrate it to the 20:1 ceiling?

~Always set up an analog WBO2 input using the "custom linear wideband" option, you can tweak the numbers as needed there.
The selectable by model options are convenient, but unlikely to be accurate.
Its a limit of analog inputs, they ~always need calibration to be accurate.

[Piledriver]

Time to use test modes, check fuel pressure, timing if it runs.
Check injector flow but getting at them in a T1 may be a pita.
Any chance you swapped coils/wires or such?
have a consistently dead hole?

It doesn't take long for a plug in a dead hole with fuel to foul.

[mrblyoung]

Time to use test modes, check fuel pressure, timing if it runs.

Used the test modes in Tunerstudio to diagnose the fuel pressure problem. Also used it to test the spark and helped find the issue.

Check injector flow but getting at them in a T1 may be a pita.

Yes. Somewhat pain to get them in and out. How do you suggest I test injector flow? Pull the injectors/rails out of the manifolds and test if they will spray fuel out the back for example? That thought crossed my mind but man that can be a mess. Haha.

Any chance you swapped coils/wires or such?
have a consistently dead hole?

Nope. All coil packs are numbered and triple checked that the wires are going to the right cylinder. Also checked that the connectors for each coil pack were in correct order. Mario's wiring harness makes it pretty easy as pair 1 have green signal wires and pair two have blue signal wires. So pretty easy to not mix those things up.

[Clonebug]

Do a "save as" on your tune and then richen everything up and see if it runs better. It won't take long to find out if the engine responds.

You could also enable "EGO" if you have a target AFR set up and see if it will try to correct.

It might be best to set autotune on easy and set up advanced settings so it can change the VE at will. Set upper and lower rpms from 500 rpm to 5000 rpm...set CLT temp to 85 degrees minimum.....also set the changeable amount to max so it can do it's thing quickly.

One other way is to highlight the whole VE map and increase all bins by a certain percentage to see if it helps.

[Piledriver]

You can pull the injectors and slip some tubing over the nozzles and run it into a calibrated beaker, but the rails obviously have to be secure to the injectors.
CIS made it easy due to the mechanical hose/injector design.
Spare stock manifolds might work depending on your setup.
You are only looking to see if the flow is ~the same...
Folks have used MS ECUS for dedicated injector testers.