Using tuning to diagnose a problem

SOLVED: visit here for the solution

--------------

Hello tuners,

 

I've been having an elusive problem with my car. The symptoms are as follows:

 

1. Cold start, the car idles like this --------v------v----v--------v--------

with "-" means smooth, and a "v" means a dip

dipping every 5-8 seconds or so until she gets warm.

 

2. Once the car is warm it doesn't display the problem

 

3. IF i disconnect the MAF it works without issue.

 

What Ii've done so far:

 

1. Cleaned maf

2. Try another maf

 

Both giving me same issues

 

Now, I was gonna try a third maf but the guy suggested it could be my Coolant Temperature sensor. Saying when my MAF is disconnected it forces the ecu into a different mode. Sounds like it makes sense.

 

My question tonight is this.

Instead of hunting down a coolant sensor to borrow, could I set values in the CT sensor scaling table that tricks the ECu to believe it's running

at warmed up temp and see if that diagnoses the CT sensor as the problem? I wouldn't drive on this, just a cold start

 

 

http://s2.postimg.org/h8iksehd5/Untitled.png

 

I'm thinking if i start the car fine tomorrow and it runs smooth then that would diagnose it?

 

OF course, I wouldn't be crazy enough to try this without consulting u guys

is this on a 05 Legacy GT? Have you tried testing for leaks?

Vacuum/inlet leaks?

 

Somewhat. What would be the best way to do that though? Boost leak is easy to test but i'm not sure about intake/vac

Radiator leaks..no, that's solid

 

But still, if i had a vacuum leak would it go away after engine warm?

I know with the maf unplugged, it goes into a speed density format for fueling, which cancels out any vacuum/boost leaks. So its possible its a leak causing the rough idle, however it would be strange that it runs normal once its warm. Unless it is something like an intake manifold gasket leaking on cold start, and once it expands, it isn't leaking..

 

Do you have a way of logging?

Yes I can log

Wonder if it's something as simple as a coil pack connector. I've change 3/4 connectors already

*smacks you* why would you mess with your ECT sensor scaling to try to trick your ECU when you can just monitor and log the ECT sensor reading and see if it's messed up?

Did you pull an LV?

 

Here is an easy way to check for vac leaks plus some other good info from my tuner.

*smacks you* why would you mess with your ECT sensor scaling to try to trick your ECU when you can just monitor and log the ECT sensor reading and see if it's messed up?

 

Thing is, everything, all sensor outputs appear fine in the logs except when the revv drops. When it drops the only thing i noticed is that the timing goes up and the left AVCS goes up 1 degree, VACUUM also drops a bit too.

 

Here is a link to a log i did when it is misbehaving

http://www.romraider.com/forum/download/file.php?id=20740

*smacks you* why would you mess with your ECT sensor scaling to try to trick your ECU when you can just monitor and log the ECT sensor reading and see if it's messed up?

 

Thing is, everything, all sensor outputs appear fine in the logs except when the revv drops. When it drops the only thing i noticed is that the timing goes up and the left AVCS goes up 1 degree, VACUUM also drops a bit too.

 

Here is a link to a log i did when it is misbehaving

http://www.romraider.com/forum/downl...e.php?id=20740

Did you pull an LV?

 

Here is an easy way to check for vac leaks plus some other good info from my tuner.

A quick look this morning showed fuel being pulled -1.2% in the idle area, the other areas have some small amount of pulling too, but i don't know or think that is significant. 40+ range is 0.

 

I'll try that vacuum leak test, it seems pretty easy

Thing is, everything, all sensor outputs appear fine in the logs except when the revv drops. When it drops the only thing i noticed is that the timing goes up and the left AVCS goes up 1 degree, VACUUM also drops a bit too.

Then it's not your ECT sensor.

A quick look this morning showed fuel being pulled -1.2% in the idle area, the other areas have some small amount of pulling too, but i don't know or think that is significant. 40+ range is 0.

 

The consensus is that if the fuel trims stay within +/- 5%, you *should* be free from vac leaks. Still does not hurt to do the quick vac test.

 

Have you logged cylinder roughness? I am no tuner expert, but it may not hurt to check them.

one known issue is the original intake manifold to tgv (or risers if ur jdm) o-rings leak air when cold. after warming up a bit they swell and the leak stops... until next "cold" start. new o-rings are much taller and fix the leak.

one known issue is the original intake manifold to tgv (or risers if ur jdm) o-rings leak air when cold. after warming up a bit they swell and the leak stops... until next "cold" start. new o-rings are much taller and fix the leak.

 

Do you think this issue would occur in Jamaica's climate?

 

In any case, here is the thread referring to what heiche described.

thanks for the link.

 

i suppose as the o-rings continue to shrink over the years, the issue would occur in progressively warmer weather "cold starts", maybe even in Jamaica.

Well, UPDATE:

 

Took the car out for some logs and I noticed that now it idles funny even when it's warm!

 

There was an issue with this car smilar to this a few weeks ago where a plug on underneath the turbo water tank (gray connector) wasn't plugged on properly. It would misfire like hell once it was warm.

 

Do you guys know the name of that sensor/plug on? It's gray and it's on both sides of the engine

camshaft position sensor

Ok. That's what I was told, Idk guys, I feel like buying one and changing it to see if that makes a difference.

 

THEN AGAIN WHY IS IT WHEN I DISCONNECT THE MAF IT IDLES FINE (a bit low though)

 

... I just bought a cheap maf off ebay for diagnosis purposes

UPATE:

 

Weird that a friend of mine had a similar problem that was solved today. It was his exhaust cam angle sensors. I believe the intakes are on top and the exhaust are lower near the timing belt?

 

Do you guys think it's a good idea to pull out all 4 and clean them? Worth a try right?

Thing is, everything, all sensor outputs appear fine in the logs except when the revv drops. When it drops the only thing i noticed is that the timing goes up and the left AVCS goes up 1 degree, VACUUM also drops a bit too.

 

Here is a link to a log i did when it is misbehaving

http://www.romraider.com/forum/downl...e.php?id=20740

 

A quick look this morning showed fuel being pulled -1.2% in the idle area, the other areas have some small amount of pulling too, but i don't know or think that is significant. 40+ range is 0.

 

I'll try that vacuum leak test, it seems pretty easy

 

The ECU will vary the timing and throttle plate angle to maintain a constant idle RPM, and I suspect that explains why you see those changing when RPM dips.

 

-1.2% in the AF Learning is negligible.

 

I don't have any theories on what's causing this, but I look forward to finding out...

So we decided it could be weak ignition blowing out the spark and decided to gap the plugs low. Now, upon removing the plugs we found one of them with a big split along the ceramic insulation!

 

Car runs like a beast now, but we are not how much a difference the gap made

 

The VVT and ignition timing changing on the misfire event was just a by-effect and not the cause of the problem.

Have you logged cylinder roughness? I am no tuner expert, but it may not hurt to check them.

 

^Did you ever log the above mentioned parameter? I believe you would have noticed some misfires in the cylinder where the damaged spark plug was.