156 2.0 TS Big Flat Spot

That makes sense as there was a distinct improvement in the reluctant revving above 4k with the new knock sensor fitted.

I am still totally baffled as to why the remaining flat spots and jerkiness occurr at the same revs consistently in every gear. This is not typical of electrical or fuel pressure/flow problems which are more erratic/random or flow/rev related rather than occurring at a discreet point in the revs every time.

Of things that should occur or start to occur at set rev values;

I pegged the inlet manifold actuator fully open and it made no difference (as did disabling the solenoid) - so it's not that.

So that leaves the dreaded variator then.

Only other thought is that the P0655 fault has tripped the ECU into 'limp home' mode - which has a crap fuelling curve as well as reduced power ? I've seen comments that the limp home mode has a power surge at 6.5 k - exactly as I have.

Anybody any experience of what to expect from the 'limp home' mode ?

Just had a horrible thought whilst contemplating how many of you have immediately suggested MAF - but I have had the MAF changed and belled out the wires to the ECU ...

What if the original MAF is the wrong one and the garage put a wrong new one back in beacuse they read the number and ordered a new one from that ???

OK Just checked the one in the car it's a 0 280 218 019 which seems to be consistent with the one for the CF2 engine as discussed on other threads.

False alarm then - it's the right one.

p0655 code is usualy caused by the ecm or wire fault try chcking the connection to the ecm ...... do you know if the engine hot lamp works?

Originally Posted by rt05492

That makes sense as there was a distinct improvement in the reluctant revving above 4k with the new knock sensor fitted.

I am still totally baffled as to why the remaining flat spots and jerkiness occurr at the same revs consistently in every gear. This is not typical of electrical or fuel pressure/flow problems which are more erratic/random or flow/rev related rather than occurring at a discreet point in the revs every time.

Of things that should occur or start to occur at set rev values;

I pegged the inlet manifold actuator fully open and it made no difference (as did disabling the solenoid) - so it's not that.

So that leaves the dreaded variator then.

Only other thought is that the P0655 fault has tripped the ECU into 'limp home' mode - which has a crap fuelling curve as well as reduced power ? I've seen comments that the limp home mode has a power surge at 6.5 k - exactly as I have.

Anybody any experience of what to expect from the 'limp home' mode ?

Limp home mode, as i understand it will not allow you to rev the engine to 6.5k I maybe wrong but is that not the whole idea of it?
all cars i have ever seen in limp mode will not rev past 3.5k

Thanks for the latest advice, especially about limp home mode which I am now confident is not the problem.

Since you replied I had the codes read again.

They are now;
Temp signal - open circuit
Temp lamp - open circuit
Knock sensor - signal unfeasible
Cam sensor - open circuit or short to battery +ve

All these were reported as intermittent but I didn't ask what that really meant - presumably they cancelled but with the engine running they would keep coming back again.

To answer your question, yes the temp lamp comes on with all the idiot lights and then blinks once at start up. The temperature gauge also works although it is always very slow to warm up despite the fan operating regularly just now.

I replaced the knock sensor a week ago, so this code is inexplicable barr a wiring fault.

With this in mind I stripped all the cloth tape of the loom and did some checking for continuity from the relevant plug to the ECU pins;

The 2 wires from the knock sensor go direct to the ECU gave perfect continuity - eh ???

Of the 3 wires from the temperature sensor 2, gave continuity and one 125 ohm on the blue wire which actually goes into the idle actuator plug - not direct to the ECU.

Of the 3 wires from the cam sensor 2 got continuity and one got 125 ohm, but get this - that was measured at the same ECU pin as for the 125 ohm measured for the temp sense blue wire. However this wire is a brown wire which also does not go back to the ECU but goes to the big 8 pin 'connector' that sits on the manifold bracket next to the crank sensor plug.

OK - without a wiring diagram I can believe that the 125 ohm thing is probably OK, but what the hell is going on with the temp gauge, warning lamp and new knock sensor (another £60 apparently wasted) ?

I checked the ECU part number on mine and compared it to a '99. The last digit on mine is a zero, the '99 last digit is 1 ?! Does this mean I have the faulty ECU found on early models ?

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

Quote from 'Kangarooing 156 - now desperate' submitted by sparkoniocv

apparently this is a fault that only exsists on the very early 156's, it is an ecu fault. I have been told by an tuning company that my 98 156 has the prototype ecu it cannot be alterd only a new one at the cost of £560 quid will cure your problem.Slackening the throttle cable helped mine,if you do find a cure let me know

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

You can see my dilema;

Spend £££ on a new variator to find it makes no difference because the ECU is bust, or spend £££ on a new ECU to find it makes no difference because the variator is bust.

Any further advice from the panel would be much appreciated

Originally Posted by rt05492

I replaced the knock sensor a week ago, so this code is inexplicable barr a wiring fault.

With this in mind I stripped all the cloth tape of the loom and did some checking for continuity from the relevant plug to the ECU pins;

The 2 wires from the knock sensor go direct to the ECU gave perfect continuity - eh ???

Of the 3 wires from the temperature sensor 2, gave continuity and one 125 ohm on the blue wire which actually goes into the idle actuator plug - not direct to the ECU.

Of the 3 wires from the cam sensor 2 got continuity and one got 125 ohm, but get this - that was measured at the same ECU pin as for the 125 ohm measured for the temp sense blue wire. However this wire is a brown wire which also does not go back to the ECU but goes to the big 8 pin 'connector' that sits on the manifold bracket next to the crank sensor plug.

(

Earth fault as a first guess, at least thats cheap to fix, assuming Im correct. On my CF3 there is an earth on the ECU mounting, earth on the cylinder head close to thermostat, earth next to coil packs, earth on clutch housing, earth under battery to body. Dont know if I've missed any others?

I have re-done all of these earths as part of my fault diagnosis. Check one of the earth cables isnt snapped by belling them out once they are disconnected. You should have about 1ohm resistance maximum from any earth point to battery ground. I always put the probe close to the connection but not on it to check.

My bella is "chez alfa" today..........

Originally Posted by rt05492

Thanks for the latest advice, especially about limp home mode which I am now confident is not the problem.

Since you replied I had the codes read again.

They are now;
Temp signal - open circuit
Temp lamp - open circuit
Knock sensor - signal unfeasible
Cam sensor - open circuit or short to battery +ve

All these were reported as intermittent but I didn't ask what that really meant - presumably they cancelled but with the engine running they would keep coming back again.

To answer your question, yes the temp lamp comes on with all the idiot lights and then blinks once at start up. The temperature gauge also works although it is always very slow to warm up despite the fan operating regularly just now.

I replaced the knock sensor a week ago, so this code is inexplicable barr a wiring fault.

With this in mind I stripped all the cloth tape of the loom and did some checking for continuity from the relevant plug to the ECU pins;

The 2 wires from the knock sensor go direct to the ECU gave perfect continuity - eh ???

Of the 3 wires from the temperature sensor 2, gave continuity and one 125 ohm on the blue wire which actually goes into the idle actuator plug - not direct to the ECU.

Of the 3 wires from the cam sensor 2 got continuity and one got 125 ohm, but get this - that was measured at the same ECU pin as for the 125 ohm measured for the temp sense blue wire. However this wire is a brown wire which also does not go back to the ECU but goes to the big 8 pin 'connector' that sits on the manifold bracket next to the crank sensor plug.

OK - without a wiring diagram I can believe that the 125 ohm thing is probably OK, but what the hell is going on with the temp gauge, warning lamp and new knock sensor (another £60 apparently wasted) ?

I checked the ECU part number on mine and compared it to a '99. The last digit on mine is a zero, the '99 last digit is 1 ?! Does this mean I have the faulty ECU found on early models ?

when you check the wiring and you replace the sensors and the ecu still isnt happy then you do have to start wondering if the ecu is to blame. i doubt it is, but i cant think of what to try next tbh.

it would be good if you could experiment with a donor ecu, but obviously it would need to have the matching codebox and key in order to actually test it.

it might be worth buying the bits from someone breaking a car on ebay. if it turns out you dont need them you can always sell them.

Thanks Doc - I checked the earths out as well but will check them again to be sure. Mine was "chez alfa Cumbria" last week - absoultely worthless, don't expect anything other than a big bill for nothing and you might not be disappointed.

Thanks Jug - I could happily live with the ECU faults still there if the car ran properly but until I clear the 'knock sensor unfeasible' I can't be sure it isn't the ECU going onto the knock safe settings you described. It only does 110mph flat out by the way - as well as the flat spots and lurching - so it's running at about 115 bhp at the most I would say ?

Is this level of power loss be consistent with the power loss from the ECU going onto knock safe settings or does this sound more consistent with a goosed variator ? The engine is silky smooth below 3000 rpm with no smoke at all ever. It does rattle like a bar steward for 5-10 seconds after startup though ?

PS Does new ECU and key mean changing the ignition barrel (and the doors and boot if you want to make it permanent) ?

Just remembered something ...

When I first got ther car (when the lurch from the flat spot was really bad), the fuel injection warning light would blink with the lurch. On mine the fuel injection light is above the battery light on the top RH corner of the rev counter.

After I changed the plugs and air filter and the severeness of the lurch reduced it stopped blinking.

Does this suggest anything to anybody ?

Tried changing the following items with no change ;

ECU, injectors, old knock sensor put back in (info on ECU changing available on request).

I'm coming to the conclusion that my fault is mechanical (or 3000 rpm related) and that having the fault codes read has just led me down a blind alley. I went over the engine loom again last night and all the earths are perfect. I don't think these fault codes are even an issue with mine as they are intermittent and don't put the car into limp home mode and don't seem to affect the performance.

My flat spots and hesitation (at constant small throttle openings only)occur at exactly 3000 rpm onwards, never below 3000 rpm - so it has to be something to do with the variator, the variable inlet manifold or the pipe that comes out of the manifold just behind the throttle body and has a 2 pin connector/solenoid (?) attached to it halfway along.

The reason I am convinced that this solenoid is a 3rd device that activates at 3000 rpm is as follows. All 3 devices have at least one white wire at their connectors. If you trace the wires back they all get spliced into one single white wire next to the big flat 8 pin connector (?) that sits on the bracket that is bolted to the manifold just next to the brown crank sensor connector.

Looks like all 3 devices are being triggered by the same signal ? Of these 3 devices, I have ruled out the manifold. The variator is probably the original (130k). If it's now leaking oil, even a small amount, this may be enough for it to lag behind the 3000rpm signal or flutter (?) and give the symptoms I have.

The pipe with the 3rd device is being mole gripped shut today to determine it's effect. I suspect it will be like the manifold - no noticeable difference.

Any thoughts on my variator comments would be appreciated.

It's possible that my throttle body/butterfly is worn and has free play that is causing flutter at small throttle openings but this wouldn't cause a flat spot at 3000 rpm at full throttle and therefore this seems unlikely to me - unless the variator and the throttle body are both gubbed.

No sure way but to change the throttle body first (cheapest) and then the variator.

Have You Had The Compresion Checked? Dose Your Engine Start From Cold First Turn Or Dose It Turn Over A Few Times Before Starting? And You Say That Your Temp Is Slow To Register But The Fan Cuts In,how Long Dose The Engine Run For Before The Fan Cuts In.

Compression not checked but it's on the list. Engine always takes several spins to start and time to fan on not measured yet - what are these parameters supposed to be an indication of ?

Update :

I am coming to the conclusion that the 2 problems I have (reluctance to rev freely above 4k and hesitation and flat spots) may not have the same root cause.

The reasoning is that the reluctance to rev freely above 4k was largely cured by fitting a 'new' fuel pump unit from a breakers. Further enquiries with Alfa about ECU numbers revealed that my ECU isn't even listed for a 2.0 TS ! According to a German website it's from a 1.8 ! Hhhmm... somebodies been messing around with this car probably a back street bodger.

So thinking over the road test I did with the correct ECU from the breakers installed, I didn't thrash test it, I just tried to replicate the flat spots which were unchanged. So I got the ECU and put it back in again. The result is that the combination of fuel pump and correct ECU seem to have completely cured the reluctance to rev. A top speed test is yet to be completed to confirm full power.

So that just leaves the hesitations and flat spots 3-4k ?

Interestingly Alfa quoted one part number 4673 9436 for a pre July 99 model (which is what I have fitted now) and another 4676 8470 for post July 99. Does anybody know what the July 99 mods were and what improvements were made (if any) ?

For reference, the numbers for the 'wrong one' fitted to car is 4673 9438 (Bosch 1037359650) which according to

Untitled Document

is for a 156 1.8 TS ?

Any thoughts / confirmations on this ECU conundrum ?

So it seems I now just have my flatspots and hesitations. The only possibilities I have left are - the variator (which having read this and other Alfa and Fiat forums I doubt affects performance it just seems to make the engine noisier with age) or the other possibility is the July '99 ECU ...

Do all early 2.0 TS's 'jerk' at 3-3.5k and the sycophantic journos never mention it ? These symptoms are classic signs of crap fuelling curves which still escape the factories, recently Triumphs TT600 was almost unrideable in 2000 it took them nearly 2 years of re-maps to sort it properly.

Originally Posted by rt05492

Do all early 2.0 TS's 'jerk' at 3-3.5k

nope.

most problems that begin at that rpm are from the MAF. disconnect it and see if there is any improvement above 3500rpm.

I agree with jug on this. But to expand on a previous post, I think if the crank sensor is defective, it is giving the ECM improbable signals. If so, the ignition timing is out which explains the knock sensor code (timing all over the place). It could be ditto for the cam sensor, or it could be purely down to the crank sensor. It is a shame that the codes weren't checked prior to the belt, as I'd have routinely changed the cam sensor to be sure.

Thanks guys, you can imagine that never having had the car working properly means I have no datum to aim for.

I did see your previous comments on the MAF but it was one of the things I changed first for a genuine Bosch item (or a very good quality copy ?). But as a double check I swapped the new one back for the old one and the car would hardly run with the idle bouncing up and down all over the place - I took that as an indication that the MAF is OK ...

most problems that begin at that rpm are from the MAF. disconnect it and see if there is any improvement above 3500rpm.

By "disconnect it", do mean the electrical connector only or leave the connector plugged in but take the MAF out of the intake pipe ? I have tried disconnecting the electrical connector but the car would not even pull away so I gave up trying to force it ?

It could be ditto for the cam sensor, or it could be purely down to the crank sensor. It is a shame that the codes weren't checked prior to the belt, as I'd have routinely changed the cam sensor to be sure.

I notice there are about 8 torx bolts holding down what I would call a black plastic 'rocker cover' on top of the head. If I remove this cover will I be able to inspect the cam sensor and at least work out whether I have a chance of changing it myself ?

Now I have the correct ECU installed I may get the codes read again. There seem to be hand held readers on the net for as little as £25 ie less than the garage will charge. Anybody know of one that will definitely work with the Bosch 1.5.5 system that mine appears to have ?

disconnect its wiring connector. the idle will be rough as hell and erratic and it will struggle until you hit 3000rpm, but then it will take off.

if it is faster at higher rpm than it is with MAF connected then MAF has a problem.

Originally Posted by rt05492

I did see your previous comments on the MAF but it was one of the things I changed first for a genuine Bosch item (or a very good quality copy ?).

NTK make a lambda sensor for the 156 and that's useless as it's 4 ohms not 8, so pattern parts may not always be the same.

Originally Posted by rt05492

I notice there are about 8 torx bolts holding down what I would call a black plastic 'rocker cover' on top of the head. If I remove this cover will I be able to inspect the cam sensor and at least work out whether I have a chance of changing it myself ?

I think it still requires the timing belt to come off and that is a big job.

Originally Posted by rt05492

Now I have the correct ECU installed I may get the codes read again. There seem to be hand held readers on the net for as little as £25 ie less than the garage will charge. Anybody know of one that will definitely work with the Bosch 1.5.5 system that mine appears to have ?

Unlikely as most are ODB-II and EOBD compatible .. go for AlfaDiag (alfadiag.net) and get a VAG-COM KKL USB cable from ebay (even the ones from HK work!). You'll need a laptop with a USB port of course. The early 156 wasn't subject to the standards imposed by the EU from 2001, so Alfa/Fiat used a proprietary system based on OBD, but not quite the same as anyone else.

You've gone far beyond anything I had to when mine played up. That was a straight actuator problem in the end after replacing the Oil Separator valve that was missing on mine ... led to a 1400 rpm idle speed!

The only think you don't mention (at least I can't see) is whether the garages have tried resetting the idle actuator (idle speed sensor) using a diagnostic system. This is where AlfaDiag paid for itself for me. I couldn't get the throttle to remain steady at 2500-3000 rpm ... the fast idle speed for the MOT emissions test ... it was only after I'd tried cleaning the throttle and attaching the actuator back AFTER the throttle cable, that I tried this and the idle remained steady and the jerkiness went. For all the time and money you have spend, the €59.95 for the software and around £10 quid for an interface cable are well worth it. It pays for itself with two uses ...

Yes I can see where Alfadiag comes into it's own I may have to invest in a 2nd hand laptop just to sort this car out.

My local Bosch dealer and the stealer in Cumbria got slightly different codes so a post ECU re-read may well be worth it.