Poor oil consumption, oil in the inlet tract, lacquering of cylinder heads.

[Ascari32]

Continuing with my investigations/modifications to my 3.2 JTS Brera engine, installed in my 159, I posted the following on the UK Forum. I thought it may be of some interest to my Aussie Friends.



A steady state indication of 90deg.C on the temperature gauge runs counter to the evidence gathered relating to this thread's title. Surely, even if there was a twenty - thirty degree error between what the instrument reads and the actual engine temperature, it would not be enough to cause the degradation of the oil, so early and to the extent that it lacquers the heads, inside the cam covers, with burnt oil.

So, having fallen over the issue of late when investigating a failed thermostat, I though I would post my observations on a separate thread in case someone has experience/knowledge of these symptoms and can throw some light on the issue!

How to test the theory, I have yet to figure out? But it is worth investigating as it is issues such as this that dog an otherwise successful exercise wrt my 159 Q4, 3.2 JTS.

Quote:- Failed Thermostat.

"Well, being running with what I believe to be a failed Thermostat! However, it is confusing. In preparation for fitting the one from my spare 159 engine, I check it functions ok in a pan of hot water. It is stamped 82 - 95deg.C., yet all those I have seen on the web quote 105deg.C.

No matter, more than happy to have it fully open at 10 - 23deg.C. sooner. Perhaps the one fitted at the moment is also 105deg.C., given I have said before, the engine is up to temperature withing a couple of miles?

But there again, was this just a little too swift? I just don't know, although it's operation is linked to the NTC sensor in the front head, cylinder two exhaust port and the engine warm - up phase involving the exhaust camshaft timing and the manifold cat temperatures/light - off.

It really is galling, as although there is plenty of theoretical information about such techniques yet nothing definitive from Alfa. It has been forever thus however, Alfa being deliberately vague about the specifics, when these techniques are in common use by "less - august" marques. Maintaining the pretence of purity when they have been shafted by virtually every other car manufacturer on the planet!

I believe someone on this Forum intimated that once the NTC sensor reaches 30deg.C., the ECU will retard the exhaust camshafts from their max warmup advance position. This at least is a clue. Rotation of the inlet camshafts, from their maximum retard starts once the car is in motion.

Fine - so far, even if this in itself is less than exact. So with what I believed was a normally functioning Stat; getting the engine up to temp (90deg.C.), within two miles, it seemed the initial tick - over of 1350 - 1400rpm on start - up from cold, backing off to 850rpm over a period of ~ 60 - 90 seconds, before moving off, is/was correct!

Now however, because the engine temperature is taking an incredibly long time to get to between 70 - 80deg.C., I make the assumption the thermostat has failed.

But what is confounding me is, it takes no time at all for the revs to die back to 850rpm and I can drive away almost immediately, with what appears to be a very cold engine when compared to what it was before. It is as if the NTC sensor is now operating within its range better. After all, why would Alfa employ such a sensitive element with a characteristic that changes uniformly within it's range, if the temperature at its location is almost immediately going to the extreme?

Rectangle Slope Line Font Parallel


Typical characteristic of NTC sensor, for illustration purposes only!

It is reputed that both the temperature sensor in the MAF, works in conjunction with this device to modify the air fuel ratio; plus other sensors but one would assume they all have their role to play and if one is out, then they are all out - if we are talking about precision! If these do indeed control AFR then it should be reflected in a burn closer to Stoichiometric Ideal, thus less sooty deposits on the tail - pipes, which is exactly what has happened with the seemingly "Faulty Thermostat"??

It is only in very slow urban driving that the engine temperature comes up to 90deg.C., whilst in all other conditions it is typically 70 - 80deg.C., although I have not done along run yet. Which seems to make sense given if one looks at the graph above at that temperature the NTC sensor is still within it's linear range. At 105deg.C. however, the slope dy/dt does not change by the same degree between 0 - 100deg.C. And the characteristic withing the ECU which is designed to re - instate a linear response will create errors to the overall.

I am becoming suspicious that either the NTC sensor is the wrong type, if one includes the heat retention factor of the Man - cats, or practically; if it is the correct type, then it's function is redundant beyond cold start - up!!!!!

But what is appearing to be the case is, it is involved in determining AFR along with the temperature sensor of the MAF, and other devices. In fact, I am persuaded the NTC sensor is working correctly for an engine which would be devoid of Manifold - cats, i.e. such as a phase II GTV, yet is incorrect when the heat source of the man - cats in included! Because, the NTC sensor would always be outside its linear range - choosing to fit a thermostat which operates at 105deg.C. on the feed to the radiator, virtually guarantees the sensor at cylinder number 2's exhaust port - even though it is in the water jacket, cannot be accurate!

And if it is inaccurate, then indicated engine temperatures above 90deg.C. would be inaccurate, which would manifest itself in elevated engine temperatures, of cylinder heads and cooking/burning/of oil and lacquering of cylinder heads under the cam covers. I do not see how this would not also apply to Phase III cars with Busso engines which also suffer from overheating.

Whatever the truth is; and Alfa will not help in determining this, my sports cat is keeping emissions within limits, the exhaust pipes are shedding the heavy soot deposits, progression from tick - over to 4000+ rpm is seamless with no indication of transition from lean - burn: @1500rpm, and @ ~ 2500rpm the camshafts start to do their job.

If indeed the thermostat was faulty and has now failed fully open, then that only seems to have enabled the NTC sensor to work properly and appears to have benefited the engine overall. Oil consumption has reduced - have to check to what extent over the next couple of thousand miles - and a certain temperamentality at the low end had disappeared.

Unless there is some significant degradation of performance from how the car performs at the moment, I shall not be replacing the thermostat. A cooler running engine begets lower oil temperatures, which was one of my target at the start of this project. I shall of course monitor just how well this all pans out but if oil decay is seen as much less than hitherto when the original 159 engine was installed and I continue to get mid 20's consumption along with silky engine performance, then I can ask no more!! "



Quote from MACE Engineering:-

"This performance 82°C thermostat suits Holden Buick LN3 L27, Ecotec L36 and L67 Supercharged 3.8L V6 engines.



Key Benefits:

Drops engine temperature by 10°C.
Prevents overheating and improves performance.
DIY installation.
Super-affordable performance modification.
Up to 30% higher coolant flow in comparison to standard replacement items, due to larger cross sectional area and more stream lined features.


Why

Heat is the single biggest killer of performance and can actually damage your engine if its exposed to extreme temperatures for a extended time frames. Avoiding heat in its various forms will not only help boost your engine's performance but also ensure that power levels are consistent and reliable.

A cheap way to reduce coolant temperatures in these engines is by installing one of our 82°C thermostats.



What

Factory-supplied thermostats begin to open at 92°C. This is far too high for a performance application. At this temperature, your engine could begin to suffer from pre-ignition (detonation) or at the very least the ECU will retard the ignition timing to prevent detonation, robbing you of precious horsepower.



How

Our 82°C thermostat opens much earlier than the factory item, allowing coolant to flow through the engine from the radiator much sooner. This limits the peak engine temperature by 10°C, preventing overheating.

With your engine running consistently cooler, your horsepower levels across the rev range will be more consistent and reliable which is especially good for boosted applications or even when towing on hot summer days.

From investigations, the 3.2 JTS Brera, 159 & Spider are fitted with thermostats which operate at 105deg.C.

The SAAB Z28net engine which is a turbo, operates at 82deg.C. Being a turbo its C/R is less than the 3.2 JTS. However, with forced induction, the effective cylinder pressure is at least the same; most likely greater. Equally the Vauxhall 2.8T Insignia. Neither of these are fitted with manifold cats!!!!!!!!!

When the manifold cats on my engine were replaced with Autodelta Free Flow Headers and the valve timing changed to 23.5deg. Positive Valve Overlap, a 100cpi cat and Supersprint exhaust system, the time it took the engine to reach optimum should have been much greater. Yet within two miles it was up to temperature. Although oil consumption reduced, it I still experienced sooty tail pipes and the garage said The engine was burning oil. It certainly appeared to be the case.

However, at some point the thermostat must have failed as Adam said it took a very long time to get the engine up to temperature for the emissions tests.

Since having the car back the tail pipes continue to show improvement as the sooty deposits clear from the tips. The engine runs cool and pulls away cleanly from start - up, without fail. Progression is seamless and there is no indication of misfire/slight hiccups when accelerating hard.

It is only suspicion however, only time will tell. But for what it is worth, my coolant level hasn't dropped, my oil level unmoved and the heat under the bonnet with a hot engine is much less intense/fierce. If a thermostat is chosen such that the cylinder head water is raised to 105deg.C. what then is the core temperature of the Alloy under the cam covers??? From evidence, clearly it is excessive and given that all elements that could retain heat have been replaced with elements that, according to the laws of physics will shed it.

I do not know of any other possible culprit involved in elevated engine temperatures, after all others have been eliminated. I am currently trying to locate a replacement thermostat which will fit. I may need to modify either one from MACE, Vauxhall or SAAB.

[Colin Edwards]

Sounds like your oil is vaporising.  Seem to recall the 3.2JTS directs a spray of oil up under the pistons to cool the crowns. 
Maybe you cam mods or cam timing mods or both have reduced the EGR so the combustion temperature is now way higher?  What are the NOX emissions like?
Maybe you need to use a less volatile oil?  Something with a lower NOACK % than what you are currently using?
Cant see engine coolant issues causing oil to vaporise.  The hottest engine component the oil comes into contact with should be the underside of the piston crown.  Its this process that really hurts the oil.
Maybe higher combustion temperatures are the cause of your altered coolant temperature gradient? 

[Ascari32]

The latest versions of spray nozzles have very well machined snouts, to deliberately keep the oil from them, in a tight jet formation, to lubricate the gudgeon/wrist pins and direct it towards the ports through the piston and into the oil ring groove.

Early versions, which my Brera engine was fitted, had badly formed snouts, with burrs and little by way of jet pattern shaping. This indeed, did cause a spray, very similar to one putting a finger over the end of a garden hose. This resulted in creating a oil mist, which was immediately whipped - up into windage by the rotating journals. So very little of the oil was directed towards the underside of the piston - this theory was tested using my old nozzles.

Quote my UK forum post: -


The work undertaken to rebuild and install my ex - Brera 3.2 JTS engine into my 159 has been a very long exercise. I do not believe there is any aspect of it, short of investigating the ECU maps, that have not been thoroughly studied with references to a multitude of sources and expert help from many individuals.

The engine which came from the 159 was in a dreadful state with obvious indications of elevated temperatures over an extended period. The temperature from underneath the bonnet had long been a target for investigation and the removal of the manifold cats a priority. In conjunction with this; after some research, was the matter of valve timing which Alfa describe loosely as playing a role in maintaining Light - off temperatures to maintain low emission. Late opening inlet valves, after the exhaust valves had closed would limit cross - flow from inlet to exhaust, thus reducing the potential for upper cylinder cooling: static timing. So the second target was to replace the Alfa Camshafts and the chance opportunity to buy a set of Colombo Bariani camshafts enabled me to alter the 2.5deg Negative Valve Overlap to 23.5deg. Positive Valve Overlap.

Thus the removal of the Man - cats and the massive cooling effect which 23.5deg. crossflow should have resulted in a tangible reduction in engine core temperature. Whilst there was some improvement and oil consumption did reduce by a fair amount, coolant temperature was still too high, prompting the all too frequent top - ups.

Along the way, coil packs suspected of failure were replaced with less than conclusive results and still the occasional mis - fire. A whole gambit of problems, some transitory, others more protracted - such as MAF element mismatch with the new engine characteristics, excessively sooty exhausts, prompting suggestions of burning oil, the issues seemed endless. It is true, overall the performance of the engine was substantial when compared to the original, but all these issues only served to suggest there was something fundamentally amiss!

Finally, the car being due its M.O.T., took an inordinately long time to get up to working temperature. So much so that I asked the garage to fit a completely original new Bosch MAF unit. From then on the engine settled down although a persistent EML and Motor Fail Indication remained. It was at this point I decided the next step would be re - mapping and I asked that the Cat - Foolers be removed so the system became "Stock" for the remap.

Up to this point, my exhaust tips were heavily coated with soot. On removal of the Cat - Foolers, they slowly emerged as smoky - slate grey. It became obvious the reason the engine took so long to come up to temperature was due to a failed thermostat [ failed open]. This remains the case.

At the same time the EML & Motor Fail indications cleared. It seemed the YouTube video about Fuel Trims described accurately what I hoped would happen. By reducing the amount of oxygen sensed by the Lambdas, the ECU backed - off the amount of fuel injected.

However, coolant level and engine oil have remained at their correct levels. We travelled from Suffolk to Cheshire, 205 miles, door to door in 3.5 hours, rarely coming below 70mph with the engine remaining cool @ 70 - 80deg.C., only creeping up to just below 90deg. when stuck in town centre traffic. I met a Forum member in Waitrose car park to give him my Ascaris and thought it would be a good time to check both the coolant and oil levels after the none stop drive. He agreed they were both at optimum!

Considerable research has been ongoing to try and discover the correct replacement thermostat. However, my old 159 engine's stat is stamped 82 - 95deg.C. All those listed as replacements are 105deg.C.

Assuming an operating range of 82 - 95, the mean would be 88.5deg.C. - nominally mid - way 50% open. All the information I can find out about the 105deg.C. replacement indicates this is the temperature at which it begins to open!!!! Assuming the same range, this would make it 105 - 118deg.C.!!!!!! In my opinion, this seems a little too high.

If however, I am to accept this to be correct, why then are so many other variants of this engine running thermostats whose value is 82 - 95deg.??

The engine, under normal operating conditions should not be so hot as to burn oil and create lacquered residue virtually all over the engines cylinder heads, cam covers, camshafts and create sufficient Positive Crankcase Pressure as to force oil into the inlet tract, up stream of the MAF?

The valve timing of the 3.2 JTS engine is intended to maintain light - off of the man - cats, not heat the cylinder heads! "Unless there is insufficient coolant flow around the engine."

This observation is true regardless of whether it it an 82deg. or 105deg. stat fitted, given the one from my now defunct 159 engine is stamped 82 - 95deg.C., yet this engine too cooked!

On close inspection of the Thermostat housing; into which the thermostat element it is fitted, it seems there is considerable restriction to any flow, past the open thermostat, due to it being quite small in diameter and has very little distance to travel between fully closed and fully open.

It is therefore my conclusion that both the size of the thermostat, the restriction it creates within the alloy body to which it is fitted, and the limited distance it can travel between open and closed is the root cause of the symptoms all 3.2JTS engines display. Overheating the Engine.

I continue to run with the failed (open) thermostat and both oil a coolant levels have remained substantially the same at over 1000 miles of driving, six hundred plus on motorways continuously driving for over 200 miles at speeds in excess of 70 miles per hour for extended periods.

Testing a good thermostat in the 3.2 JTS thermostat housing, even the 82 deg. version, opens no more than 0.6cm.~ when fully up to temperature, which greatly restricts coolant flow. I have therefore come to the conclusion that the thermostat will be permanently removed and control of temperature will be left entirely to the radiator fans, something some modern cars rely upon anyway.

Foot - note.

Close coupled manifold cats would maintain light - off in any case, with engines as powerful as the 3.2 JTS, so there is no need for such a complex valve timing arrangement such as Alfa's. The combination of both the manifold cats and the sporadic function of modulating the exhaust camshafts seems to ensure that a lot of the heat stored by the man - cats, is reflected back into the cylinder heads, to the extent that it can cause coil packs to melt, on top of inciting low rpm knock!

When this happens, and the anti - knock devices operate, the ECU adjusts the timing to prevent damage. This action in turn will strip torque from the engine, which could well be another reason why the 3.2 JTS is gutless at low revs.

Either way, one cannot continue to find fault with components, without questioning the way Alfa chose to employ them. Not enough was understood by Alfa of the mutual effect adding so many emission control techniques to this engine, or the consequences for its performance.

My modified engine sailed through M.O.T. with emissions well within limits and for a 16 year old car, that is remarkable; albeit, the engine has been rebuilt only just over 11,000 miles ago. It is fitted with a full Supersprint Exhaust system, with 100cpi sports cat, Autodelta free - flow headers and Colombo Bariani Special camshafts, the valve timing of which creates 23.5deg. of Positive Valve Overlap, beginning a 0.5deg. BTDC. I know of no engine with such a huge valve overlap, that ticks over at 850 rpm - rock solid.

The duration of both inlet an exhaust cams is 278deg., which prompted one highly respected American Tuner to comment:

"Those angles are very similar to my Ferrari Racer. One thing is certain - they ain't no Granny Cams".

PS. The complete removal of the thermostat element from the engine housing will have a further benefit. Freeing up coolant flow, will also reduce loading on the mechanical water pump. That too should reduce loading on the crankshaft and should further help low rpm torque by a finite amount.

PPS. The reported failings of the 3.2 JTS and the suggested reasons for failure are legendary. Both "Expert" Alfa establishments and owners have created a catalogue of reasons for failures, very few of which have been borne out in repeatability testing. In many cases, proposed solutions are so far off the mark, they read like fairy stories.

The reality is, this is a very sophisticated engine, made fragile by the degree to which Alfa went to, to achieve low emissions, such that it was awarded the highest category by legislators. The cost of trying to maintain this engine, as years go by, will guarantee it virtually disappears completely. Unless it is better understood! There is little evidence to suggest that is going to happen any time soon. 

[Ascari32]

Sounds like your oil is vaporising.  Seem to recall the 3.2JTS directs a spray of oil up under the pistons to cool the crowns. 
Maybe you cam mods or cam timing mods or both have reduced the EGR so the combustion temperature is now way higher?  What are the NOX emissions like?
Maybe you need to use a less volatile oil?  Something with a lower NOACK % than what you are currently using?
Cant see engine coolant issues causing oil to vaporise.  The hottest engine component the oil comes into contact with should be the underside of the piston crown.  Its this process that really hurts the oil.
Maybe higher combustion temperatures are the cause of your altered coolant temperature gradient?

"Maybe you cam mods or cam timing mods or both have reduced the EGR so the combustion temperature is now way higher?  What are the NOX emissions like?"

- The Colombo Bariani camshafts have removed the EGR function the Alfa Camshafts performed hitherto. Alfas exhaust valves closing at 9deg. ATDC and their inlets opening at 11.5deg. ATDC - Static Timing Angles.

Colombo Bariani Special cams - inlet opens at 0.5deg BTDC and exhaust closes at 23deg. ATDC. - Again Static Timing Angles.

However, the 23.5deg. of cross-flow produces a tremendous amount of upper cylinder, valve, piston and cylinder head cooling. It also maintains higher inlet air density than hitherto and produces a greater "cylinder fill", the MAF registering a greater mass air flow and consequently, stoichiometrically is more accurate, producing greater torque at low rpm.

The absence of manifold cats ensures the hot exhaust gases move away from the cylinder heads at a much greater velocity than hitherto.

There is a popular misconception with respect to just how hot these engines ought to run at. For the vast majority of their duty cycle, they are not running at W.O.T. At legal speeds in Britain, the engine is running at ~ 2500 rpm. There is no question in my mind, these excess temperatures are a function of the engines inability to shed heat and the principle vehicle for that is via the coolant system.

"Maybe you need to use a less volatile oil?" - I use the oil recommended by Alfa Romeo and one presumes they know the reason why they chose it!

"The hottest engine component the oil comes into contact with should be the underside of the piston crown." - that is very transitory. Bearing faces are where the oil temperature becomes most quickly elevated.

It is true however, the poor characteristic of my original nozzles produced more of a fine mist, which combined with blow - by HC's and Water Vapour, causing considerable emulsification of the oil, elevated moisture and air entrainment. These elements precipitated out of the oil wherever there was a change in flow direction or a restriction - of which this engine has four major ones - five if one includes the rear bank cylinder head.

This entrainment of HC's, Water Vapour and Air is a major factor in crankshaft main bearing failure and that of big - ends. This is why it is imperative to regularly change the oil and filter at low mileage intervals.

Poor design of the spray jet nozzles and the pressure the internal valve operates at are also major factors in how well this engine survives. The valves in the spray nozzles can jam either open or closed. Open and the block drains down when switched off - particularly when oil is hot, and closed when HC crud builds up and blocks it, running the gudgeon/wrist pin dry.