Alfa Romeo Forum banner

21 - 30 of 30 Posts

·
Registered
Joined
·
132 Posts
Discussion Starter #21 (Edited)
the overheating happened like a year ago and a lot faults happened since then, so I couldnt connect these 2 things. Also there was no white smoke, but black smoke from overfueling and the oil cap looks normal, no mayo on it.

Time to go to the mechanic and open the engine :(
 
G

·
Guest
Joined
·
0 Posts
Get rid... It's screwed and you haven't exactly helped matters by driving it for the past 4 month. It is literally going to cost three times the car is worth to sort, you can buy a good example for the cost of repairing it.

If I had to guess, you have multiple issues that have been caused by an initial compression failure in one or two cylinders. As a result you now have possible valve issues and o2 and cat issues due to unburned fuel detonating further down the manifold and exhaust system.

Sometimes, learning by your mistakes and finding a better example of your current car is the only course of action....
If I had to figure out the cost of the advice that is already been given in this thread + labour, you could afford a Gta and run it by now:rolleyes:


Sorry but just being honest.
 

·
Registered
Joined
·
132 Posts
Discussion Starter #23
you are right. luck is i can get anY Part second Hand very cheap. just like i found all the other Part (coils , lambda, ecu and Body for under 100 euros). will See if i can switch the whole head, if the car works ok i will keep the head.
i will sell the car as soon as i fix it.
 

·
Registered
Joined
·
6 Posts
Hi, I have exactly same problem on my Alfa 147, same engine. I tried replacing everything electricbased but in vain. When car is conected to diagnostic and you delete all the errors (lambda1 and cat1 error, multiplication), car immediately starts to work fine, no discomposedly work, fuel consumption is normal but after a driving some distance (10 - 50 km's) everything is back to old. My car wasn't overheated and consuption of water and oil is minimal. Is it possible that is head gasket? Smoke from my exhaust is black and back door have black dots from exhaust...
 

·
Registered
Joined
·
132 Posts
Discussion Starter #25
So I changed the lambda sensor again few days ago, and it worked ok for an hour and started shacking again (not firing fuel). When I reset the ECU the car works ok for a few minutes, when the engine is cold the car is ok, when the lambda starts working it starts to shake. Something is killing my lambdas really fast. Anyone?
I even changed all the wires on the engine.
 

·
Registered
Joined
·
1 Posts
I have the same problem! My engine not working on #1 and #4 cylinder. I have checked coils (primary coil approximately 0.9 ohm and secondary #1: 14.98 kohm, #2:14.03 kohm, #3:14.75 kohm, #4:14.70 kohm). I put new main and small spark plugs and checked all cables.

Engine goes to missfire usually on low rpm and whan engine is hot. Injectors cleaned and checked.
I didn't check compression, but I doubt it because it sometime works fine, sometimes works very bad. Probably it is something electrical, bit I can not figure out what...

Any idea?
Thanks!
933127
 

·
Registered
Joined
·
2,556 Posts
Hmm, overfuelling on cylinders 1&4.

Firstly I'd disconnect the fuel pump, ideally at the tank multiplug under the seat. That would be better than disconnecting the fuel pump relay as that would enter a fault code. That way you could remove the spark plugs and test them if they are fitted back into the coils and the ground electrode is earthed.

I don't think that is the fault though. I think it points more to a partially collapsed catalytic converter (1&4 cylinders). If you remove the heat shields, you could tap the 2 pre-cats. I'd expect one of them to rattle (I think the one nearest the gearbox).

If it is the maniverter and you need a new one, make sure the one you get has full size ports at the cylinder head flange as some new ones have overly small, restrictive ports.

Alternatively, a garage airline may be able to do a rather poor flow test if the front pipe is disconnected from the maniverter.

Final thought; the spark plugs are cheap alternatives so won't last like OE NGK ones.
 

·
Registered
Joined
·
788 Posts
I’ve just seen something very similar on a friend’s Peugeot 308. Plugs/Coils/Injectors/Compression all swapped/changed/tested with no result, but I caught a whiff of exhaust fumes in the cab when accelerating. Using a torch in the dark showed a leak around the manifold when the engine was revved so current theory, yet to be tested is that air is getting in (the pressure goes negative during part of the cycle). If air gets in the lambda probe tells the ecu that the engine is lean on the affected cylinder(s) so it adds more fuel and those cylinders run rich. What happens then depends on how the ecu handles it. The Pug shut down cylinder 4 injector and signaled a misfire. This always happened when the engine was hot and usually when idling.
 

·
Registered
Joined
·
466 Posts
"I don't think that is the fault though. I think it points more to a partially collapsed catalytic converter (1&4 cylinders). If you remove the heat shields, you could tap the 2 pre-cats. I'd expect one of them to rattle (I think the one nearest the gearbox)."

If the cats (either precats or the primary cat) have a loose core, then they might or might not rattle when tapped. I don't think that failure to rattle is reliably diagnostic of the core matrix being tight, or indicative that it is loose (though a cat that does rattle when tapped would be highly suspect...).

Looking for loose cat cores;
For the precats take out the O2 sensors and look inside. Shine a light through each sensor orifice, can you see the core matrixes in correct position? If you carefully probe each matrix with a pointy implement (through the sensor orifice), does either one move at all?

For the primary cat, you might as well just take it out and have a good look inside, shake it and carefully prod the matrix. To take the cat out, it's just two nuts at one end and two bolts at the other end of the cat. So long as the nuts and bolts aren't seized it's an easy job. You don't even need to disconnect the rubber exhaust hangers (which can be a bit of a pain), the metal chassis hooks just unbolt from the chassis pretty easily.

Putting the cat back in;
One end of the primary cat doesn't use a gasket (just a 'cone' and 'socket' kind of joint). The other end does use a gasket, which is likely to be reusable (if cleaned up, along with the faces of the flanges, and a hi temperature gasket sealant used). I wouldn't use a 'hard' exhaust sealant upstream of catalytic convertors (i.e. a sealant which hardens into a ceramic like material), because as the joint is assembled any excess still soft sealant gets squeezed out and ends up being caught in the cat matrix, where it hardens and partially blocks the matrix.

-------

I don't know what problem this engine has, but sometimes bad head gaskets can seal well enough while the engine is still relatively cool, only leaking when the engine gets up to temperature. Since the engine is known to have seriously overheated at some point in the past, I'd be paying careful attention to coolant level. If coolant is mysteriously disappearing with no signs of external leakage, then it's a good bet it is leaking into one or more cylinders...

--------

As suggested above (by R.T.), a leak in the exhaust upstream of one of the O2 sensors could well be affecting the O2 sensor reading, 'adding' rogue oxygen to the exhaust gas which one of the O2 sensors is measuring. The sensor would then send a signal to the ECU that is indicative of a lean AFR, and the ECU would respond by enriching the AFR, possibly quite a lot (inappropriately since the AFR is probably not really lean, but the ECU can't 'know' this...). An exhaust leak can sometimes be zero or minimal when the engine is cold, but worsen as the exhaust heats up.

If the problem is that cylinders one and four are running very rich, as seems to be the case;
The exhaust gas from cylinders one and four is monitored by the 02 sensor on the right hand side of the 'manivertor' (on the right hand side of the car), so if the problem is restricted to cylinders one and four, then there could be a problem with the right hand side O2 sensor, or, the right hand side of the manivertor could have a leak somewhere (flange gasket is a good bet), or maybe there could be a leak at the head flange of the manifold where the primary pipes one and / or four are attached to the head ports. Any leak could suck O2 into the exhaust and corrupt the data being sent to the ECU by the affected O2 sensor.

And maybe, the ECU will run in open loop mode until the O2 sensors start to work (i.e. when they get hot), and so when the sensors are cold the ECU will run from stored maps rather than from what the sensors are telling it (or are not telling it, since they don't work immediately). This could explain why the engine briefly runs OK when it is cold, then runs badly as temperature rises...??

Regards,
John.

PS The flanges between the CF3 manifold and the manivertor seem to be prone to warp fairly badly. If a straight edge is placed across the width of the flange faces don't be surprised to see that the flanges are a long way from being properly flat. This makes it problematic for the gasket to properly seal the flange joint. These flanges on my car were significantly warped, and I could see sooty evidence of at least slight leakage near where the flanges were most warped. The flanges on my spares car were about the same, with similar evidence of probable slight leakage.

Ideally the warped flanges should be machined flat, but good luck with that. The manifold flange has two downward protruding studs, and the manivertor flange has three upward protruding studs. None of the studs will want to come out, no matter how much penetrating oil or blow torch heat you apply (two are actually welded in place). When you break them (as you may well do), they are made from a very hard steel which is quite difficult to drill out (even after they have been annealed...).

So I had flanges none of which were flat enough, all of which are quite rigid (due to being quite thick), and all of which would have been quite difficult to machine flat. With the warped manivertor and manifold flanges assembled together (with no gasket and not tightened), there was a gap between the faces into which I could fit a 0.9mm feeler gauge, which is a lot of gap. A lot of force would be needed to flex these stiff warped flanges into 'flatness' against each other, and what will actually happen is that the gasket will be unevenly compressed, so the seal will be uneven around the circumference of each of the two flange holes (i.e. two holes ingoring stud holes). This is a recipe for upcoming leaks...

To address this problem what I have done is to cut the manifold flange completely in half with an angle grinder (thin cutting disc). The cut runs through the narrowest flange section, passing from the front of the flange through the middle of a stud hole to the other side where the flange is narrowest (where the flange has a 'V' shaped indentation). The flange is now bifurcated and is effectively two flanges, held together only by the manifold tubes. Unfortunately neither of these flanges is perfectly aligned with the other (i.e. each flange not being exactly in the same plane), but there is flex in the manifold tubes which allows each of the two flanges to move more independently to the other (than is the case with the flange uncut, i.e. there is now a lot more flexibility in the manifold).

When assembled onto the still somewhat warped manivertor flange, this extra freedom of movement makes it easier for the two manifold flanges to accomodate any difference between each other, and also to accomodate the warp in the manivertor flange. This mod just takes some stiffness out of the manifold flange and makes it easier to assemble with less risk of the gasket leaking. The flanges will clamp the gasket more evenly around the circumference of each hole, the seal will be better and probably last longer.

Note that there are three studs protruding upward from the manivertor flange. The cut in the manifold flange means that the centre one of these studs passes through the manifold flange hole that has the cut through it. Probably not essential, but I added a large washer under the nut which threads onto this stud, to make sure the tensile pressure is placed onto a large enough area of the manifold flange (considering the cut over which the nut is placed, it just seemed the right thing to do).

I also drilled a hole through the manivertor flange and a corresponding hole in and between the two manifold flanges (these holes drilled near the point of the 'V' indent, through the manivertor flange, and 'into' the cut now between the two manifold flanges). I then put a bolt / nut through these holes with large washers each side of the flanges. This is probably not necessary, but places more clamping load around each of the two exhaust holes (also just seemed the right thing to do...).
 
21 - 30 of 30 Posts
Top