Engine Noises re Boxster, 996 and 997.
Copied from threads with contribution from Hartech and Pope.
There are several noises that can be heard from these engines that sound very much the same, some of which are caused by serious engine failures while others are relatively innocent. This makes diagnosis very difficult without comparative experience and other tests. To explain the difficulties - typical noises that are serious are caused by the following.
(1) Crankshaft bearing failure – big ends or main bearings (or both) – usually caused by wear on the rear main bearings lowering the oil pressure just there and consequently therefore lowering it to the number 6 or 3 big-end (because the oil flow path travels to the main bearing first and from that up into the crank pin/big end area). The noise from this sounds quite heavy and increases with revs and is present as the revs rise or fall – usually louder if the engine is hot. This failure of the main bearing white metal in the shell is caused by crankshaft bending or flexing increasing the wear loads on the shell due to the large overhang at the rear of the engine to accommodate the two chains being driven there for the camshaft system. If the engine is being used for aggressive braking or is over revved - this can bend the flywheel end of the crankshaft so much the starter ring can touch the bell housing inside. The dual mass flywheel is a contributory factor as it is so heavy and relatively poorly balanced as it wears. For most racing applications in other cars with similar dual mass flywheels – they are removed to reduce the weight and improve balance and replaced with lighter solid flywheels.
(2) A fuel injector can also make a tapping sound that varies with throttle opening (and consequently with revs) but usually if you rev the engine and then shut the throttle – it would not make the sound as the revs fall because there is no fuel passing into the engine (the injector pulse is switched off).
(3) A hydraulic tappet is the usual diagnosis for a slower noise than at crankshaft speed – more of a ticking noise. It often goes away after the engine has warmed up (as the hydraulics pump up the tappet) but can also be caused by the top of the tappet actually wearing away (or even creating a hole through the top) in which case it is permanent. Wear in the tappet housing can also reduce the pumping efficiency of the oil transfer from the oil delivery system into the tappet.
(4) The Cayman S 3.4, 996 3.6 & 997 3.6 &3.8 engines can have damaged pistons and scored bores – usually after about 50 to 70K miles. The damage is like seize marks on one side of the piston (the reasons and causes of which we understand and can remedy during a rebuild) but the extra piston clearance that results allows the piston to tilt as it goes over top dead centre and the edge of the piston just touches the cylinder head making a tapping noise. This can be checked for by putting a camera (often called a bore scope or boroscope) to look inside each cylinder. The damage usually seen on bank 2 (cylinders 4, 5 & 6) on one, two or all three bores (because on this side of the engine the thrust load of the piston is on the hotter part of the bore whereas on bank 1 the thrust side is the coolest) . The damage seems to take some time to gradually get worse and initially can be present with no symptoms at all – gradually increasing oil consumption and slightly less sharp performance (that many owners do not notice until they receive a rebuilt engine back and remember how crisp it used to be).
(5) Valve seats. It is not unknown for a valve seat to become lose in the cylinder head and bend or break a valve – early signs would probably be an intermittent tapping noise.
(6) A worn IMS bearing (or a worn or damaged chain) or swarf trapped in the scavenge recirculation pumps, or damage to the chain sprockets, or wear on the chain slippers, or worn hydraulic tensioners – can all throw up a snatch in the engine system as it revolves which can emanate as a tick.
(7) You will probably be sceptical when I tell you that the exhaust manifold can also make a noise exactly like a metallic tick (but believe me it can and I won the bet to prove it!). The bolts that hold the manifold to the cylinder head rust and the manifold becomes slack. The gasket is a thin metallic one with small ridges to seal against the surfaces, but it doesn’t fit too well leaving a very thin sealing face between the exhaust port and the exhaust gas recirculation (EGR) ports. All heads have the grooves for these ports cast in and some are drilled through to inside the exhaust ports while others are not – and some have an EGR system on the car while others do not. The ports connect all three cylinders on one bank together. When the slack bolts and thin contact result in a gap appearing between the exhaust port and the EGR ports, at tickover – the cycling of the pulses makes a very metallic sounding noise that goes away as the engine falls in revs but is there as it increases from low to medium revs (very like a tappet getting quieter as the revs and oil pressure rise – but tappets would still be noisy as the revs fell – but not the exhaust system as the throttle will be closed and the exhaust pulse of little or no intensity).
(8) SOMETHING ELSE. We have received two cars with a noise that sounds like it has been caused by some of the above faults and religiously gone through every option and replacement – only to find the noise is still there despite none of the above being the cause – a mystery that we have not yet solved. In both cases they were tiptronics and the noise was there when hot and went away on revving the car even slightly. As the revs fell it gradually re-appeared and sometimes was heard on and off at random. Interestingly the noise also went away instantly if we engaged “R” or “D” even though the engine revs and the oil pressure remained exactly the same. This means that in those cases the noise was not caused by anything related to oil pressure, crankshaft shells, valves, valve seats, tappets, pistons, chains, tensioners etc (or any of the above causes) so what is it caused by? So far we do not know for sure but we do have a theory – that may or may not prove to be right. The sound is always coming from bank 2 oil pump end of the engine usually in the head area (which is why it is often mistaken for some failing there) – but we (and others it seems) have changed everything there (including cams and checked cam housings etc etc) checked or changed little ends (the list is endless) and the evidence seems to point to it being caused by something else entirely. The clue came when the noise stopped as soon as we engaged a gear in the tiptronics gearbox because that should have little to do with the engine after all - except – that is – that it alters the load or weight/mass of the crankshaft and as such increases its damping effect. Engines have flywheels designed to damp out the fluctuations that occur in the crankshaft speed between firing impulses. The crankshaft slightly speeds up and slows down as the engine changes form one cylinder firing and the next one compressing the mixture – so changing from positive to negative impulses. They are designed so the result does not become a shake or a snatch that makes an unsettling noise – but the larger the weight the slower the car accelerates and the more fuel it uses – so manufacturers fit the lightest they can get away with. Whenever an engine is used for racing – the lighter the flywheel the faster it accelerates so weight is usually taken off – but this often results in crankshaft snatch at low revs causing various noises (especially if the camshaft is chain driven). Early engines have 5 chains and later ones have 3 – and as the chains wear and the sprockets wear (together will the rest of the engine) eventually there could be enough slack to allow the snatch to sound like a tapping noise (as the reversal of the crankshaft speed pulls the chains tighter on one side and then the other against the tensioner which at tickover has relatively low oil pressure and tension). But the sound always seems to come from cylinder 4 area (but this is where the chain driving bank two camshaft sits). The firing sequence is 1 6 2 4 3 5 and looking from the top from the rear towards the front (of a 911) on the LHS are cylinders 1, 2 and 3 and on the RHS 4, 5, and 6. This means that usually a cylinder fires one close to it in the next sequence. Cylinder 6 next fires cylinder 2 (next opposite), 2 to 4 is the same, 4 to 3 is at opposite ends of the crankshaft but 3 to 5 is again next door to one another opposite as is 5 to 1, but then 1 to 6 is at opposite ends of the crankshaft again. This means there are two phases in which the loads are transferred from one end of the crankshaft to the other – one transferring drive from bank 2 (cyl 4) to bank 1 (cyl 3) and one transferring it from bank 1 (cyl 1) to bank 2 (cyl 6). When this first reverse happens (firing from cyl 4 to 3) the cylinder 3 is at the end of the engine where the crankshaft drives 2 chains (and the intermediate shaft is not required to do anything) as it drives bank 1 camshaft directly and so there is plenty of oil splash from the chains onto the bank 1 camshaft chain drive and no IMS shaft twist to consider. But in comparison when the drive changes from cyl 1 to cyl 6 (at the other extreme end of the crankshaft) the intermediate shaft also has to transfer the bank 2 camshaft drive to the other end of the engine and then up through only one chain (less lubricated than the two bank 1 chains). This means that specifically – the cylinder firing order from bank 1 to 6 would potentially create the greatest theoretical twist/snatch of all the other cylinders onto the least lubricated chain that is running right next to cylinder 4 (bank 2) where the noise usually comes from. It seems possible then that as all the engine components gradually wear and all the parts that contribute towards the mass of the system get loser – that this snatch creates a noise – only at the lowest of engine revs and that this goes away when the torque converter is engaged by selecting “D” or “R” (because this alters and increases the mass and damping effect of the crankshaft – just as if it had suddenly become heavier). If so a similar affect may occur with a worn dual mass flywheel. Similar noises have been heard for years in 3.2 911’s (from the gearbox at low revs) from racing 944’s and 968’s (with lighter flywheels) and from many other racing cars with lighter crankshafts. If this theory eventually proves right and this noise is merely a snatch noise slowly creeping up on older engines at tickover – it has no detrimental effect and it is pointless wasting huge amounts of money trying to get rid of it as it does no harm and you would have to replace lots of parts that are working perfectly OK with new ones for no reason other than a slightly stiffer fit or movement to quieten a noise only present on tickover that is otherwise doing no harm. However – in view of the seriousness of the other 7 potential faults that can cause a similar noise (listed above) they may all need to be properly investigated. Baz/Hartech
(9) GT3 start-up rattle
Aside from chain tensioner and chain drive noises on start-up (similar to the other versions of the engine type) the GT3 engine uses 'Variocam' - a system that enables active adjustment of valve overlap by changing the position of the Inlet camshaft in relation to the Crankshaft and Exhaust camshaft. The electro-hydraulic system on the inlet camshaft sprocket has oil pressure fed to both sides of the sprocket element to fix it in place; adjustment is effected by changing the pressure on either side of the sprocket to rotate the element (left or right).
On the 9x6 and 9x7 models with second generation Variocam the sprocket is mechanically fixed until oil pressure builds and takes over sprocket control after start-up and it is from here that the noise emanates - In the GT3, the sprocket element isn't fixed so as the engine is cranked the sprocket is free move in its designed adjustment range (stop to stop / rotationally left to right) - as the cam loads against the next set of valves to open the camshaft moves to its 'left stop' (potentially with a 'clack'), as the engine continues to rotate the camshaft reaches peak valve lift and the lobes begin to ramp back off of the valves; at this point, because the sprocket isn't at full tension / pressure, the camshaft is helped round by the valve spring pressure and can rotate to its 'right stop' (potentially with a 'clack' - see where this is going??)
As oil pressure builds (almost instantly) the 'clack, clack' sound is lessened to eventual 'silence' (maybe not the right word!!)
Depending on the oil temp, how long the engine has been left to stand, the position of the cams when the engine was last stopped etc the noise is more or less apparent - in no way does the noise suggest a fault, more just a characteristic of the system.
I have seen one case on a Mk1 GT3 where the oil valve in the camshaft housing worked loose which allowed faster pressure loss / loss of oil retention once the engine was stopped (the valve could not have failed completely but the relative movement around where the valve should have been fixed was enough to make the above noises more prevalent).
There are several noises that can be heard from these engines that sound very much the same, some of which are caused by serious engine failures while others are relatively innocent. This makes diagnosis very difficult without comparative experience and other tests. To explain the difficulties - typical noises that are serious are caused by the following.
(1) Crankshaft bearing failure – big ends or main bearings (or both) – usually caused by wear on the rear main bearings lowering the oil pressure just there and consequently therefore lowering it to the number 6 or 3 big-end (because the oil flow path travels to the main bearing first and from that up into the crank pin/big end area). The noise from this sounds quite heavy and increases with revs and is present as the revs rise or fall – usually louder if the engine is hot. This failure of the main bearing white metal in the shell is caused by crankshaft bending or flexing increasing the wear loads on the shell due to the large overhang at the rear of the engine to accommodate the two chains being driven there for the camshaft system. If the engine is being used for aggressive braking or is over revved - this can bend the flywheel end of the crankshaft so much the starter ring can touch the bell housing inside. The dual mass flywheel is a contributory factor as it is so heavy and relatively poorly balanced as it wears. For most racing applications in other cars with similar dual mass flywheels – they are removed to reduce the weight and improve balance and replaced with lighter solid flywheels.
(2) A fuel injector can also make a tapping sound that varies with throttle opening (and consequently with revs) but usually if you rev the engine and then shut the throttle – it would not make the sound as the revs fall because there is no fuel passing into the engine (the injector pulse is switched off).
(3) A hydraulic tappet is the usual diagnosis for a slower noise than at crankshaft speed – more of a ticking noise. It often goes away after the engine has warmed up (as the hydraulics pump up the tappet) but can also be caused by the top of the tappet actually wearing away (or even creating a hole through the top) in which case it is permanent. Wear in the tappet housing can also reduce the pumping efficiency of the oil transfer from the oil delivery system into the tappet.
(4) The Cayman S 3.4, 996 3.6 & 997 3.6 &3.8 engines can have damaged pistons and scored bores – usually after about 50 to 70K miles. The damage is like seize marks on one side of the piston (the reasons and causes of which we understand and can remedy during a rebuild) but the extra piston clearance that results allows the piston to tilt as it goes over top dead centre and the edge of the piston just touches the cylinder head making a tapping noise. This can be checked for by putting a camera (often called a bore scope or boroscope) to look inside each cylinder. The damage usually seen on bank 2 (cylinders 4, 5 & 6) on one, two or all three bores (because on this side of the engine the thrust load of the piston is on the hotter part of the bore whereas on bank 1 the thrust side is the coolest) . The damage seems to take some time to gradually get worse and initially can be present with no symptoms at all – gradually increasing oil consumption and slightly less sharp performance (that many owners do not notice until they receive a rebuilt engine back and remember how crisp it used to be).
(5) Valve seats. It is not unknown for a valve seat to become lose in the cylinder head and bend or break a valve – early signs would probably be an intermittent tapping noise.
(6) A worn IMS bearing (or a worn or damaged chain) or swarf trapped in the scavenge recirculation pumps, or damage to the chain sprockets, or wear on the chain slippers, or worn hydraulic tensioners – can all throw up a snatch in the engine system as it revolves which can emanate as a tick.
(7) You will probably be sceptical when I tell you that the exhaust manifold can also make a noise exactly like a metallic tick (but believe me it can and I won the bet to prove it!). The bolts that hold the manifold to the cylinder head rust and the manifold becomes slack. The gasket is a thin metallic one with small ridges to seal against the surfaces, but it doesn’t fit too well leaving a very thin sealing face between the exhaust port and the exhaust gas recirculation (EGR) ports. All heads have the grooves for these ports cast in and some are drilled through to inside the exhaust ports while others are not – and some have an EGR system on the car while others do not. The ports connect all three cylinders on one bank together. When the slack bolts and thin contact result in a gap appearing between the exhaust port and the EGR ports, at tickover – the cycling of the pulses makes a very metallic sounding noise that goes away as the engine falls in revs but is there as it increases from low to medium revs (very like a tappet getting quieter as the revs and oil pressure rise – but tappets would still be noisy as the revs fell – but not the exhaust system as the throttle will be closed and the exhaust pulse of little or no intensity).
(8) SOMETHING ELSE. We have received two cars with a noise that sounds like it has been caused by some of the above faults and religiously gone through every option and replacement – only to find the noise is still there despite none of the above being the cause – a mystery that we have not yet solved. In both cases they were tiptronics and the noise was there when hot and went away on revving the car even slightly. As the revs fell it gradually re-appeared and sometimes was heard on and off at random. Interestingly the noise also went away instantly if we engaged “R” or “D” even though the engine revs and the oil pressure remained exactly the same. This means that in those cases the noise was not caused by anything related to oil pressure, crankshaft shells, valves, valve seats, tappets, pistons, chains, tensioners etc (or any of the above causes) so what is it caused by? So far we do not know for sure but we do have a theory – that may or may not prove to be right. The sound is always coming from bank 2 oil pump end of the engine usually in the head area (which is why it is often mistaken for some failing there) – but we (and others it seems) have changed everything there (including cams and checked cam housings etc etc) checked or changed little ends (the list is endless) and the evidence seems to point to it being caused by something else entirely. The clue came when the noise stopped as soon as we engaged a gear in the tiptronics gearbox because that should have little to do with the engine after all - except – that is – that it alters the load or weight/mass of the crankshaft and as such increases its damping effect. Engines have flywheels designed to damp out the fluctuations that occur in the crankshaft speed between firing impulses. The crankshaft slightly speeds up and slows down as the engine changes form one cylinder firing and the next one compressing the mixture – so changing from positive to negative impulses. They are designed so the result does not become a shake or a snatch that makes an unsettling noise – but the larger the weight the slower the car accelerates and the more fuel it uses – so manufacturers fit the lightest they can get away with. Whenever an engine is used for racing – the lighter the flywheel the faster it accelerates so weight is usually taken off – but this often results in crankshaft snatch at low revs causing various noises (especially if the camshaft is chain driven). Early engines have 5 chains and later ones have 3 – and as the chains wear and the sprockets wear (together will the rest of the engine) eventually there could be enough slack to allow the snatch to sound like a tapping noise (as the reversal of the crankshaft speed pulls the chains tighter on one side and then the other against the tensioner which at tickover has relatively low oil pressure and tension). But the sound always seems to come from cylinder 4 area (but this is where the chain driving bank two camshaft sits). The firing sequence is 1 6 2 4 3 5 and looking from the top from the rear towards the front (of a 911) on the LHS are cylinders 1, 2 and 3 and on the RHS 4, 5, and 6. This means that usually a cylinder fires one close to it in the next sequence. Cylinder 6 next fires cylinder 2 (next opposite), 2 to 4 is the same, 4 to 3 is at opposite ends of the crankshaft but 3 to 5 is again next door to one another opposite as is 5 to 1, but then 1 to 6 is at opposite ends of the crankshaft again. This means there are two phases in which the loads are transferred from one end of the crankshaft to the other – one transferring drive from bank 2 (cyl 4) to bank 1 (cyl 3) and one transferring it from bank 1 (cyl 1) to bank 2 (cyl 6). When this first reverse happens (firing from cyl 4 to 3) the cylinder 3 is at the end of the engine where the crankshaft drives 2 chains (and the intermediate shaft is not required to do anything) as it drives bank 1 camshaft directly and so there is plenty of oil splash from the chains onto the bank 1 camshaft chain drive and no IMS shaft twist to consider. But in comparison when the drive changes from cyl 1 to cyl 6 (at the other extreme end of the crankshaft) the intermediate shaft also has to transfer the bank 2 camshaft drive to the other end of the engine and then up through only one chain (less lubricated than the two bank 1 chains). This means that specifically – the cylinder firing order from bank 1 to 6 would potentially create the greatest theoretical twist/snatch of all the other cylinders onto the least lubricated chain that is running right next to cylinder 4 (bank 2) where the noise usually comes from. It seems possible then that as all the engine components gradually wear and all the parts that contribute towards the mass of the system get loser – that this snatch creates a noise – only at the lowest of engine revs and that this goes away when the torque converter is engaged by selecting “D” or “R” (because this alters and increases the mass and damping effect of the crankshaft – just as if it had suddenly become heavier). If so a similar affect may occur with a worn dual mass flywheel. Similar noises have been heard for years in 3.2 911’s (from the gearbox at low revs) from racing 944’s and 968’s (with lighter flywheels) and from many other racing cars with lighter crankshafts. If this theory eventually proves right and this noise is merely a snatch noise slowly creeping up on older engines at tickover – it has no detrimental effect and it is pointless wasting huge amounts of money trying to get rid of it as it does no harm and you would have to replace lots of parts that are working perfectly OK with new ones for no reason other than a slightly stiffer fit or movement to quieten a noise only present on tickover that is otherwise doing no harm. However – in view of the seriousness of the other 7 potential faults that can cause a similar noise (listed above) they may all need to be properly investigated. Baz/Hartech
(9) GT3 start-up rattle
Aside from chain tensioner and chain drive noises on start-up (similar to the other versions of the engine type) the GT3 engine uses 'Variocam' - a system that enables active adjustment of valve overlap by changing the position of the Inlet camshaft in relation to the Crankshaft and Exhaust camshaft. The electro-hydraulic system on the inlet camshaft sprocket has oil pressure fed to both sides of the sprocket element to fix it in place; adjustment is effected by changing the pressure on either side of the sprocket to rotate the element (left or right).
On the 9x6 and 9x7 models with second generation Variocam the sprocket is mechanically fixed until oil pressure builds and takes over sprocket control after start-up and it is from here that the noise emanates - In the GT3, the sprocket element isn't fixed so as the engine is cranked the sprocket is free move in its designed adjustment range (stop to stop / rotationally left to right) - as the cam loads against the next set of valves to open the camshaft moves to its 'left stop' (potentially with a 'clack'), as the engine continues to rotate the camshaft reaches peak valve lift and the lobes begin to ramp back off of the valves; at this point, because the sprocket isn't at full tension / pressure, the camshaft is helped round by the valve spring pressure and can rotate to its 'right stop' (potentially with a 'clack' - see where this is going??)
As oil pressure builds (almost instantly) the 'clack, clack' sound is lessened to eventual 'silence' (maybe not the right word!!)
Depending on the oil temp, how long the engine has been left to stand, the position of the cams when the engine was last stopped etc the noise is more or less apparent - in no way does the noise suggest a fault, more just a characteristic of the system.
I have seen one case on a Mk1 GT3 where the oil valve in the camshaft housing worked loose which allowed faster pressure loss / loss of oil retention once the engine was stopped (the valve could not have failed completely but the relative movement around where the valve should have been fixed was enough to make the above noises more prevalent).