Help with Rough Idle when Engine is Hot

One of the easiest things to start with is to check for a vacuum leak.

 

Might be the O2 sensor. When the car starts cold, the O2 sensor signals are ignored by the cars fuel computer (ECU). Once warmed up, the O2 sensors feed their reading to the ECU and it takes control of the mixture using that feeback from the exhaust. If the O2 sensor is dirty or worn out it will send the wrong signals and the computer will not give the correct mixture. If it has not been replaced, it is a good time to do so, unlike most parts for Ferrari they are not very expensive. That's my best guess.

Most of the cars fueling complexity is to allow the car to have low emisions at idle. So when there are issues, it usually presents at idle - off idle the car can run fine as there is higher tolernance for fueling issues at high levels of fueling.

I would take out a spark plug and see if there is signs of fouling, ie. black oily soot being a rich mixture. Might give some further clues. Mostly when things go wrong the fuel mixture is too rich.

Except for the vaccum leak idea, which will cause things to be too lean. There are various hoses that go the intake plennum, and are part of the auxiliary air valve system, make sure there are no cracks in hoses and snug up any clamps. Too rich or too lean can both cause a rough idle.

As noted there is an AAV, auxiliary air valve, the opens and closes based on the car warming up and is what creates the cold idle being higher to start before settling down to about 1000 rpm? Is your cold start idle higher?

 

Ah yes, my mistake for some reason I was thinking about the 328, not the 308.

 

Failing ignition coils could be the problem, once hot, shut off and check resistance. Compare to cold values.

 

If the ignition system can be ruled out as a problem, another issue for warm idle on the K-Jet without Lambda models can be a change in the WUR's behavior. When cold, the WUR control pressure is quite low = adds richness which makes ICEs run well, and any small change in this added richness has little noticeable affect. However, when warm, the WUR control pressure is much higher = leans things out, and the Mechanic sets the mixture screw in this condition to be just slightly richer than stoichiometric. Consequently, if the warm control pressure from the WUR changes a little to be slightly higher than it was when the Mechanic tweaked it up (and a WUR is an analog, open-loop thermo-mechanical gizmo so can age/change) = can go from being just slightly richer than stoichiometric to slightly lean below stoichiometric = too lean to warm idle well. Anyway, just wanted to add that once the ignition system is ruled out, the next usual diagnostic steps for the symptoms on your model (which are all done with the same test gauge set-up) would be:

1. Confirm the system regulated supply fuel pressure is OK.
2. Confirm the WUR cold control pressure is OK.
3. Confirm the WUR warm control pressure is OK.

If all in-spec:

4. Confirm/Tweak the mixture screw setting when at warm idle.

Good Hunting!

 

Try an internet search using "spark plug appearance chart" and have a look at some of those images -- looks most like "ash fouling" to me.

 

That plug does not look like all is well in the combustion. It is not terrible, but I suspect enough deposits and gap issues to cause a miss. I would change the plugs to regular NGK's, the iridiums (I think that is what those are) do not necessarily work best in these older cars.

The cars can idle quite well with one entire bank not firing, so it does not surprise that one cylinder disconnected does not cause an obvious impact. It would be felt under load for sure though.

It certainly does not look like you are running rich, which is probably a good sign. Perhaps your mixture was not set correctly in the first instance, it is a fussy itterative process to juggle the very small movements of the mixture screw setting and the air bypass screw.

Many do not advise playing with the mixture screw, but I think it can be used to help diagnose things. eg. richen the mixture by turning the screw clockwise 1/16 of a turn. Just keep track so you can move it back the same amount. It might solve for things, if indeed you are running lean.

The main clue is the rough running only when hot. I wonder if the thermo switch sensor is failed or corroded/covered in coolant deposits? I have the later 3.2 so I am guessing on your car, but I think the fuel injection system relies on a temperature sensor to tell the computer what do do when the car has warmed up.

 

Re possible vacuum leak - keep in mind that when the engine is cold the mixture is richened automatically because cold engines need rich mixtures to start/run. A slight vac leak can be unnoticed due to the richer mixture supplying sufficient fuel to disguise it. Once the engine begins warming, the extra richening is no longer added and a vacuum leak has a greater affect.

If the engine runs ok under load/higher RPM and not at idle, that's another symptom of a possible vac leak. Ignition misfire problems generally do the opposite - OK at idle/low load, worsening as load increases.

 

Bad Gas Probably?

 

I vote for a vacuum leak. the hoses can get brittle, and if / when things get colder during fall/winter - change or Spring / Summer change the hoses contract or expand. ... creating the leaks. It's a pita to chase the leaks, but the hoses to the AAV are cheap and easy to change, also the crazy elbow hose from the cold start injector is under INTENSE pressure vacuum so it may be collapsing when warm...

Its not bad gas... never really is.
its usually not the CIS - KJet is pretty simple and reliable. I would not touch the CO mix until you have gone through all the electrics, and vacuum hoses first. ALOT of folks go right to the CO mix or CIS system while the real issue is electrical.

if you have a CAT and O2 sensor - I just disconnected mine and let it run in closed loop... it usually idled smoother, and did not do that surge that they all do with the O2 sensor plugged in.

 

Looks like spark plug gaps had worn too wide.

Might tighten up the hose clamps on the various hoses that are part of the air injection, and any others that get plennum vacuum. The clamps tend to loosen over time as the hose clamp area compress and take a set over time and heat cycling.

 

In the marine world - with systems using LOTS of hose clamps - I was always amazed at how they would loosen. At the spring "re commissioning" it was a standard thing to check/tighten the hose clamps. Some were quite loose.

Related to that...I was once standing in one of tthe engine rooms of a 53ft Hatteras yacht when I heard a "SPROING" sound. I looked in the sound's direction and saw the end of a cooling system hose clamp slowly waving back and forth. The engine was not running and had not been running since the previous summer - the clamp just snapped with absolutely no additional strain or "input" of any kind. Ya just never know...

 

What's the exhaust look/smell like when hot Brian? Any chance you have somewhere that can test the exhaust when hot? Ignition or rich/lean might show up in the exhaust. How's the coil 12+ to coil mount voltage when hot?

 

I wouldn't go that far with it (though I suppose too late, at this point.) A basic propane torch from the hardware store often works well enough.

 

I think the gap for the iridiums is set by NGK at .032. So you might have gapped them too close.

Why not try changing out to the non iridium NGK's gapped per the tighter factory spec.

I would want to be definitive to rule out the plugs before doing more things.

I would also take of the easiest access dist cap and take a look, see if the rotor is ok, and no cracks in the cap, they can crack and short.