The only reason for me posting on this subject is to repeat what I have already posted many months ago and to add information for readers to make of what they will.
Firstly can we make the distinction between "bore scoring", "hot seizures" and "cold seizures". and the influence of oil types and viscosities and ambient conditions on all three.
"Bore scoring" occurs when the cylinder block is at the right temperature and so was the piston (so clearances were OK) just before a minute piece of silicon breaks free from its matrix on the cylinder wall and is trapped between the piston face under load and the cylinder wall.
The load/area distinction then puts ALL the load from the piston driving the con-rod and crankshaft (normally shared across a wide area of the piston face at 90 degrees to the gudgeon pin) on the piece of debris where the increased load/over area enables it to penetrate the oil film, digging into the piston surface and rubs up and down the bore before often escaping with the oil.
Like a pot hole on a road (where a piece of stone moves in the direction of travel and tends to elongate the hole) the damage releases more pieces of silicon as it travels up and down the bore and they join in the damaged area elongating the bore score.
Because the load was concentrated where the piece became loose (which is the thrust face because that is where since new the forces have been greater pressing against the cylinder matrix) the other side of the piston and cylinder bore are undamaged.
This type of failure has been seen before in Alusil after several hundreds of thousands of miles (944/968) where the piston coating is hard and much more frequently in Lokasil (because the silicon particles are bigger and less well bonded into the matrix).
Thicker oil can result in a slightly thicker oil film between the piston and cylinder bore and may extend life slightly but eventually cannot protect the damage mode for ever.
"Hot seizing" is where the piston has become bigger than the bore it is supposed to fit into with a clearance because something has increased the temperature dramatically.
This can be because the cylinder block is too hot (due to lack of coolant or coolant flow) or more often because the ignition has made the piston much hotter than normal (fuel, weak mixture, too much ignition advance etc).
Although cylinder bores should be round, pistons are not and are oval (because the thrust load is at 90 degrees to the gudgeon pin) so the part of the piston that grows too big is usually across the thrust face and because the piston is now bigger than the cylinder bore, it squeezes out any oil film resulting in metal to metal contact which raises the temperature locally so much there is temporary melting and setting of parts of the piston thrust face and it digs into the cylinder bore where the friction further increases localised temperatures and the result is a cylinder bore and piston scored (or seized) on both thrust and the opposite face. Once again although better oils may delay failure it usually will not prevent the failure.
"Cold seizures" in our terminology at Hartech (others may use other descriptions) is where the piston thrust face may or may not have been damaged but some part of the piston no longer has clearance between itself and a part of the cylinder bore - not because there is a fault in the settings, fuel or coolant - but simply because the piston has heated up and expanded before the cylinder bore and reduced the clearance to zero resulting in too much pressure between the piston area that fails and the cylinder bore and causing the rubbing described in a "hot seizure", the rapid temperature rise resulting and the localised scoring.
Pistons skirts do distort but so does the local area of the cylinder bore subjected to this rapid local temperature rise. Since the temperature rise is on the surface of the cylinder bore - just there the surface expands into the bore resulting in creating another localised high pressure are that may be in the thrust face area or not.
"Cold seizing" does not mean the engine is not warm or the coolant is not at normal running temperature it means that the heat has not yet soaked fully into every area of the cylinder block distorting some areas and reducing expansion generally elsewhere.
Pistons heat up and expand quicker than cylinder bores.
The dimensional changes during thermal expansion take place as a proportion of the size so bigger pistons expand more than smaller ones. As a result cold seizures are more common in big diesel engines (on ships) than small petrol engines but when they do occur in petrol engines they are more likely to have relatively big pistons commonly in the 100mm diameter size much more than say 70 or 80 mm. If you had an engine with a piston 10 times bigger than a 9A1 piston (i.e. 1020mm) you could not run it with a piston clearance of 10 times as much because the weight of the piston would slap against the cylinder wall and squeeze out the oil film as the engine ran - so cold seizures are more of a problem.
When metals are formed from heat (castings, extrusions, hot rolling etc) the inside of any large cross sections remain hotter while the outside cools first. The outside freezes as it cools while the inside continues to shrink in size as it cools later and this always leaves a stress pulling inwards towards the centre of a thick section.
If you bore a hole in a hot rolled piece of bar and then gently heat it - when it has cooled again, the centre will usually be smaller than it was when you bored it as a result.
Similarly in castings you have the highest temperature gradient between hot (when it was molten) and cool (when it was all at ambient) so all castings retain internal stresses pulling inwards at the thickest sections.
Machining the castings releases stresses that were retained in the areas you have just machined away so if they are machined almost to finished size and then undergo a stress releasing process and then finally machined - it minimises the problem and retains the best shape stability but this is time consuming and costly.
In volume production it is difficult to build into the process stress releasing features and often they are ignored. I don't know by what process the 9A1 cylinder blocks are manufactured regarding stress relieving (does anyone else?).
However - if scoring/seizing marks are NOT ON THE THRUST FACE but are further around the circumference of the cylinder bore (and therefore opposite an area of the piston where the clearances are usually smallest) something has distorted the cylinder where the scoring marks are and reduced the clearance resulting from the cylinder no longer being round but having circumferential areas where if you measure across the diameter the bore is smaller than it was when it was originally manufactured.
Those used to 2 strokes decades ago will already know that cylinder stud pressures can distort these areas of cylinder bores and stop them being as round as they were designed to be when the piston ovality was tested and formulated resulting in cold seizures.
IF an engine suffers a typical "cold seizure" (in our terminology), and it is in an area where there is thicker sections of the casting (for studs etc) it is usually cased by distortion.
If an engine suffers a typical "cold seizure" in line with a large section change in the cylinder block casting it is often because over time the continual heating and cooling cycles have minutely released the stress that was retained in the thick section by pulling inwards and changing the shape. The clue is in what part of the piston it affected and where in the cylinder block there are large section changes.
If a failure has an age related stress releasing problem it will not occur when they are new and it will take many cycles to show itself. There will come a time when all the retained stresses are released and the shape is permanent.
If when this happens, there is still sufficient bore clearances no damage will result.
If a bore is gradually distorting and in so doing reducing the bore to piston clearances in a particular area over time, it will reveal itself when the engine has been put under a high load before the block has fully warmed up and expanded - so is more likely in winter than summer.
Unfortunately, because there has been a gradual almost imperceptible change, driving from cold just like the owner has got away with perhaps for years will suddenly catch him out resulting in seizing/scoring.
Recognising the scoring being opposite the crankcase large section where the main bearing shells sit and where the casting is joined to the cylinder bore, combined with the scoring not being at the biggest part of the piston but on either side resulted in us thinking it may be caused by age related stress relieving and so we attempted to measure the failed bore all round and up and down and the undamaged ones next to it.
In every single one of the 5 we measured the scoring was on an end cylinder and the adjacent cylinder had shrunk by about half the amount the failed one had while the furthest away was still round. This would be typical of a casting in which the molten metal was perhaps hotter than usual (or the mould cooler) and the flow cooled one end sooner than the other (although this is just typical engineering opinion trying to link cause and effect).
We have measured cylinders in engines we have rebuilt that had a completely different fault - we have found to all still be perfectly round.
In the early stages of diagnosis there are not enough samples around to be sure about anything and engineers have to draw on their experience and try and connect it to the evidence. Anyone making grand statements about the eventual outcome at this stage are just being dramatic for effect and not being calm and logical (especially if they have not carried out any forensic investigations themselves like we have).
I am not sure what the cause is, I cannot tell if there will be a time or mileage limit to the distortion, how many engines may be affected and if it was a problem that has been recognised in manufacture and eliminated.
I can tell you that the undamaged cylinders we measured next to the damaged ones had piston clearances reduced to under half a thou across the diameters near the crank shell locations and that the upper undamaged areas of the cylinders that had scored/seized had shrunk more than that. I can also confirm that they all failed during the winter.
Although the coolant and oil may well be running at the same temperature summer or winter, the outside of the cylinder block will not and the block will expand more in the summer and quicker.
The difference in ambient temperatures summer to winter may be only 20 or 25 degrees (UK) whereas the difference in cold to hot running temperatures will be 65 to 70 degrees in summer but perhaps 100 degrees in winter - so if the shrinkage is slow over time it is most likely to reveal itself in the winter where the piston will heat up as quickly and expand just as much and as fast summer or winter but the cylinder block will lag further behind it.
If our conclusion (admittedly based on a very small sample but never the less consistent) is right, oil types will make little difference and frankly be a distraction although there is never an argument not to use the best oil you can get hold of nor not to increase the viscosity rating as engines wear and clearances increase as a result.
I have no problem if you disagree with our opinion, it makes no difference to us but I do hope that in the interests of accuracy, and helping those without the benefit of our analysis - anyone posting alternatives can back up their opinion with similarly convincing evidence and actual experiences.
Baz
Porsche 991: 2011-2019
Porsche 997: 2004-2012
Porsche 996: 1997-2005
Porsche 993: 1993-1998
Porsche 964: 1989-1993
Porsche Carrera 3.2: 1983-1989
Porsche Carrera SC: 1977-1983
Porsche 930 Turbo: 1975-1989
Porsche Early 911: 1964-1977
Porsche Carrera GT
Porsche GT1
Porsche 959
Porsche Macan : 2013-
Porsche Panamera : 2009-
Porsche Cayman: 2005-
Porsche Cayenne: 2003-
Porsche Boxster: 1997-
Porsche 968
Porsche 944
Porsche 924
Porsche 914
Porsche 912
Porsche 904
Porsche 550
Porsche 356 Speedster
Porsche 356
