Latest LTT and 3rd rad test results
Discussion
Just thought I would keep those interested up to date on the latest tests of the third radiator. These tests with a 3.8 997 (as before).
To re-cap - too much radiator cooling capacity (like fitting a third radiator) combined with a thermostat fitted on the inlet to the engine - actually raises the outlet temperature - so if that is something to avoid (which we think it is) the third radiator is only a benefit if the overall capacity of the system is unable to cool the engine in some circumstances (like a very hot day or continuous heavy acceleration at relatively low running speeds).
Driving fast on an open road seems to allow enough air flow to maintain cool temperatures and driving slowly in traffic doesn't usually create excessive heat inside the engine - so both of those scenarios usually don't create an over-heating problem. The main problems we found were stopping for a while after a fast run and then driving off again - as the bank 2 thrust side of the engine picks up a lot of heat and thins the oil film between the piston and the cylinder wall.
The latest tests were conducted in ambient air temperatures of 5 to 7 degrees C, a third radiator fitted and our Hartech LTT.
Temperature sensors were fitted at the inlet and outlet of each bank of the cylinder block and each cylinder head. Speeds were a mixture with heavy acceleration and high top speeds as a part of the tests as well as town driving and followed exactly the same routes and conditions as before except ambient conditions were cooler.
Compared to our recent results conducted at 12 to 18 degrees C( In which the bank 2 outlet temperatures were significantly higher after a period of ticking over or engine stopped) - with all three radiators in use or with one main radiator completely blocked there was no significant difference in any of running temperatures. This indicates that the overall radiator area of two main or one main and one centre radiator are sufficient in those conditions and that they are well balanced to the thermostat and the slight change in coolant flow speed resulting. It suggests the system has been well designed for operating within these lower temperature ranges - sub 12 Degrees C. The temperature rises we recorded with the engine stopped or ticking over when the air temperature was higher - were much greater and therefore fit in with what we might expect from colder ambient conditions and their ability to cool the outside of the engine and exhaust more quickly due to static and low speed air flow over the outside of the hot parts.
With both radiators blocked (running only on the centre radiator working) the temperatures only rose by approximately 20 degrees from around 80 degrees C to 100 degrees C - but remained stable .
It was a surprise to me to see that throughout the tests the main temperature gauge on the dashboard never budged from reading 80 degrees although the coolant temperatures rose from 79 to 103 degrees and the oil temperature rose from 85 to 105 as the radiator area was gradually reduced.
This made me wonder how the temperatures that others report are being read since it is obviously much too unreliable to use anything on the dashboard instrument and we have not found a way to read the ECU temperature on these later cars (as GT4 explained could be done on the earlier 996 examples). I should have taken our diagnostic computer with me to see what engine temperatures the ECU was recording but was probably too anxious to get the tests done and didn't think of it at the time (and will do so on the next tests).
During these tests we fitted a small part of our latest development and that did keep the outlet temperature of bank 2 from rising too much by controlling the influence of the third radiator (more later).
It is unfortunate that we cannot control ambient conditions and I fully accept that we will be more interested in the results on a stinking hot day - but we will have to wait for that for some time it seems. But these tests will be repeated throughout the year and I hope will throw more light on the conditions that lead to bore scoring and meanwhile we will be testing out the full new system very soon and hope it reveals the type of temperature controls in bank 2 we are seeking and that they continue to work as expected as the weather warms up..
Next tests are more complicated but I will keep you informed as promised.
Baz
To re-cap - too much radiator cooling capacity (like fitting a third radiator) combined with a thermostat fitted on the inlet to the engine - actually raises the outlet temperature - so if that is something to avoid (which we think it is) the third radiator is only a benefit if the overall capacity of the system is unable to cool the engine in some circumstances (like a very hot day or continuous heavy acceleration at relatively low running speeds).
Driving fast on an open road seems to allow enough air flow to maintain cool temperatures and driving slowly in traffic doesn't usually create excessive heat inside the engine - so both of those scenarios usually don't create an over-heating problem. The main problems we found were stopping for a while after a fast run and then driving off again - as the bank 2 thrust side of the engine picks up a lot of heat and thins the oil film between the piston and the cylinder wall.
The latest tests were conducted in ambient air temperatures of 5 to 7 degrees C, a third radiator fitted and our Hartech LTT.
Temperature sensors were fitted at the inlet and outlet of each bank of the cylinder block and each cylinder head. Speeds were a mixture with heavy acceleration and high top speeds as a part of the tests as well as town driving and followed exactly the same routes and conditions as before except ambient conditions were cooler.
Compared to our recent results conducted at 12 to 18 degrees C( In which the bank 2 outlet temperatures were significantly higher after a period of ticking over or engine stopped) - with all three radiators in use or with one main radiator completely blocked there was no significant difference in any of running temperatures. This indicates that the overall radiator area of two main or one main and one centre radiator are sufficient in those conditions and that they are well balanced to the thermostat and the slight change in coolant flow speed resulting. It suggests the system has been well designed for operating within these lower temperature ranges - sub 12 Degrees C. The temperature rises we recorded with the engine stopped or ticking over when the air temperature was higher - were much greater and therefore fit in with what we might expect from colder ambient conditions and their ability to cool the outside of the engine and exhaust more quickly due to static and low speed air flow over the outside of the hot parts.
With both radiators blocked (running only on the centre radiator working) the temperatures only rose by approximately 20 degrees from around 80 degrees C to 100 degrees C - but remained stable .
It was a surprise to me to see that throughout the tests the main temperature gauge on the dashboard never budged from reading 80 degrees although the coolant temperatures rose from 79 to 103 degrees and the oil temperature rose from 85 to 105 as the radiator area was gradually reduced.
This made me wonder how the temperatures that others report are being read since it is obviously much too unreliable to use anything on the dashboard instrument and we have not found a way to read the ECU temperature on these later cars (as GT4 explained could be done on the earlier 996 examples). I should have taken our diagnostic computer with me to see what engine temperatures the ECU was recording but was probably too anxious to get the tests done and didn't think of it at the time (and will do so on the next tests).
During these tests we fitted a small part of our latest development and that did keep the outlet temperature of bank 2 from rising too much by controlling the influence of the third radiator (more later).
It is unfortunate that we cannot control ambient conditions and I fully accept that we will be more interested in the results on a stinking hot day - but we will have to wait for that for some time it seems. But these tests will be repeated throughout the year and I hope will throw more light on the conditions that lead to bore scoring and meanwhile we will be testing out the full new system very soon and hope it reveals the type of temperature controls in bank 2 we are seeking and that they continue to work as expected as the weather warms up..
Next tests are more complicated but I will keep you informed as promised.
Baz
The problem is that if none of the test engines fail whether they have standard coolant or special coolant - how do you determine if they are worthwhile using or not.
My own view is that anything that has a higher boiling point (or that will not boil at the temperatures these engines experience even if the coolant pressure falls to ambient) must be better than standard coolant - but whether your engine or car will ever experience conditions for which that safety margin is necessary - I cannot determine.
Most of the improvements we research and apply to our cars relate to standard engines and coolant (in an attempt to help owners avoid bore scoring in the first place) and so far we have not experienced a reason to upgrade - but if a safety margin is of concern there certainly would be no harm in doing so.
Baz
My own view is that anything that has a higher boiling point (or that will not boil at the temperatures these engines experience even if the coolant pressure falls to ambient) must be better than standard coolant - but whether your engine or car will ever experience conditions for which that safety margin is necessary - I cannot determine.
Most of the improvements we research and apply to our cars relate to standard engines and coolant (in an attempt to help owners avoid bore scoring in the first place) and so far we have not experienced a reason to upgrade - but if a safety margin is of concern there certainly would be no harm in doing so.
Baz
Yes except that having the thermostat on the inlet can create a problem. If the ambient is very cold - the thermostat has to close a lot to try and raise the inlet temperature to the thermostat setting and this in turn slows the coolant speed through the engine and lifts the exit temperature.
If you go back many year radiators often had blinds fitted to allow you to close them in a controlled way in cold weather (my Saab 96 had one when I was in my twenties). This is a good idea as it allows the coolant speed to keep up.
The M96 and M97 engines only allow about 10% of the coolant to pass through the cylinder block so slowing it down even more creates the possibility of high temperature rises within the block and with bank 2 thrust face being at the hottest part just before the coolant leaves the cylinder block - temperatures could get quite high there.
This is why we are presently experimenting with ways to control the exit temperatures rather than the inlet temperatures.
More later.
Baz
If you go back many year radiators often had blinds fitted to allow you to close them in a controlled way in cold weather (my Saab 96 had one when I was in my twenties). This is a good idea as it allows the coolant speed to keep up.
The M96 and M97 engines only allow about 10% of the coolant to pass through the cylinder block so slowing it down even more creates the possibility of high temperature rises within the block and with bank 2 thrust face being at the hottest part just before the coolant leaves the cylinder block - temperatures could get quite high there.
This is why we are presently experimenting with ways to control the exit temperatures rather than the inlet temperatures.
More later.
Baz
What a great question! if you asked it in the first 2 or three years of Gen 1 production you would probably answer that there is not a problem - so to some extent we may have to wait and see - although they seem really good so far.
Pistons are tapered at the top to allow them to expand with heat generated without getting too big for the bore.
Many years ago the problem emerged with powerful Sports cars - that a piston had to have enough clearance to expand when flat out producing lots of power and heat yet not be too slack when it was just running around at 30mph with very little comparative heat generated. It was essential to have enough piston clearance flat out and this became too much at low speed for a car used for both purposes (unlike a race car).
The idea emerged to try and use an aluminium cylinder block so the expansion rates were similar to the piston and to try and keep clearances closer in all conditions.
But aluminium pistons would not run in aluminium bores - so one of many attempts to find a solution became Alusil in which silicon particles were mixed in with the molten aluminium and cast into a block.
The hard spots of silicon were distributed very evenly and securely and after boring to size the aluminium was initially chemically etched back a small amount (leading to a dull finish) and later rubbed away by a type of exposing hone - creating minute spaces for oil to stick and be somewhat similar to the porous cast iron that had been used successfully for so long but had too low an expansion rate to work well in high performance engines.
This was used with great success in the 924S. 944 and 968 engines. However the small silicon particles were still hard and so a thin ferrous piston coating was also used (often with a very thin tin flash) to resist wear.
Nikasil was also used in air cooled engines but was different as it electroplated a sort of thin shim into the bore that held itself together without the destruction of minute particles to interfere with the piston and so worked without any piston coating. It was expensive but ideal for individual cylinders (which all the air-cooled 911's used) easy to apply and hone to size (and was also used in liners for the GT3 and turbo models) where the manufacturers could pass on the extra costs in the price of the car.
Small manufacturers of special pistons seemed unable to reproduce the ferrous coating so they tried several new plastic coating (Molykote, Zylan etc) which lasted quite well in racing engines (as they only do about 1,500 miles/season).
The problem with Alusil was that the silicon was everywhere in the cylinder block and made machining very expensive (although created a fabulously rigid strong block).
So the idea came to locate silicon only near the cylinder block (or "local silicon or LOKASIL") - but to do it a composite tube was cast and held in the casting mould into which pressurised alloy was squeezed and set - creating a porous local silicon mix in an easy to machine block with a hard porous surface - (utopia)?
The early engines were used with the original ferrous piston coating previously used in the 924/44/68 range and worked OK but then apparently the piston coating process became unpopular for environmental reasons and was replaced with the type of plastic piston coating previously used reasonably successfully in special small volume pistons in Alusil.
The publicity about Lokasil described it as exactly the same as Alusil in use but we think it is quite different. Firstly it is less stiff and contributes less to cylinder stiffness and hence the "opendeck" cylinders go oval. Secondly we believe the silicon becomes detached at a more frequent rate.
I think Lokasil may have been OK if it were possible to coat the pistons as before (something we actively seek more later), and/or if the coolant flow was directed at the hot thrust faces and with greater flow rate, and/or if the running temperature was lowered (LTT), or perhaps some mixture of both or all three.
Because they didn't solve the Lokasil problems Porsche went back to the previously successful Alusil in the Gen 2 engines but this time with plastic not a ferrous coating. This is my only reservation because although we accept Alusil is much better than Lokasil - we do not yet know how well it will last with Alusil (my guess is for a perfectly acceptable mileage or age).
Whereas the engines of older Porsche models lasted much longer than they could reasonably be expected to (and therefore cost too much to manufacture) and the in between production of Lokasil did not last long enough generally - I expect the new Gen 2 engines to sit somewhere in between and therefore acceptable in the modern World of fashion becoming more important and longevity becoming less of a sales issue.
While we too have used the best of all Nikasil solution in our rebuild options - we are still trying to find ways to make lokasil work or last long enough to be acceptable. This may prove fruitless and a waste of money or it may result in a breakthrough of sorts (which we obviously think it will or we would not pursue it).
I hope this answers all your thoughts!
Baz
.
Pistons are tapered at the top to allow them to expand with heat generated without getting too big for the bore.
Many years ago the problem emerged with powerful Sports cars - that a piston had to have enough clearance to expand when flat out producing lots of power and heat yet not be too slack when it was just running around at 30mph with very little comparative heat generated. It was essential to have enough piston clearance flat out and this became too much at low speed for a car used for both purposes (unlike a race car).
The idea emerged to try and use an aluminium cylinder block so the expansion rates were similar to the piston and to try and keep clearances closer in all conditions.
But aluminium pistons would not run in aluminium bores - so one of many attempts to find a solution became Alusil in which silicon particles were mixed in with the molten aluminium and cast into a block.
The hard spots of silicon were distributed very evenly and securely and after boring to size the aluminium was initially chemically etched back a small amount (leading to a dull finish) and later rubbed away by a type of exposing hone - creating minute spaces for oil to stick and be somewhat similar to the porous cast iron that had been used successfully for so long but had too low an expansion rate to work well in high performance engines.
This was used with great success in the 924S. 944 and 968 engines. However the small silicon particles were still hard and so a thin ferrous piston coating was also used (often with a very thin tin flash) to resist wear.
Nikasil was also used in air cooled engines but was different as it electroplated a sort of thin shim into the bore that held itself together without the destruction of minute particles to interfere with the piston and so worked without any piston coating. It was expensive but ideal for individual cylinders (which all the air-cooled 911's used) easy to apply and hone to size (and was also used in liners for the GT3 and turbo models) where the manufacturers could pass on the extra costs in the price of the car.
Small manufacturers of special pistons seemed unable to reproduce the ferrous coating so they tried several new plastic coating (Molykote, Zylan etc) which lasted quite well in racing engines (as they only do about 1,500 miles/season).
The problem with Alusil was that the silicon was everywhere in the cylinder block and made machining very expensive (although created a fabulously rigid strong block).
So the idea came to locate silicon only near the cylinder block (or "local silicon or LOKASIL") - but to do it a composite tube was cast and held in the casting mould into which pressurised alloy was squeezed and set - creating a porous local silicon mix in an easy to machine block with a hard porous surface - (utopia)?
The early engines were used with the original ferrous piston coating previously used in the 924/44/68 range and worked OK but then apparently the piston coating process became unpopular for environmental reasons and was replaced with the type of plastic piston coating previously used reasonably successfully in special small volume pistons in Alusil.
The publicity about Lokasil described it as exactly the same as Alusil in use but we think it is quite different. Firstly it is less stiff and contributes less to cylinder stiffness and hence the "opendeck" cylinders go oval. Secondly we believe the silicon becomes detached at a more frequent rate.
I think Lokasil may have been OK if it were possible to coat the pistons as before (something we actively seek more later), and/or if the coolant flow was directed at the hot thrust faces and with greater flow rate, and/or if the running temperature was lowered (LTT), or perhaps some mixture of both or all three.
Because they didn't solve the Lokasil problems Porsche went back to the previously successful Alusil in the Gen 2 engines but this time with plastic not a ferrous coating. This is my only reservation because although we accept Alusil is much better than Lokasil - we do not yet know how well it will last with Alusil (my guess is for a perfectly acceptable mileage or age).
Whereas the engines of older Porsche models lasted much longer than they could reasonably be expected to (and therefore cost too much to manufacture) and the in between production of Lokasil did not last long enough generally - I expect the new Gen 2 engines to sit somewhere in between and therefore acceptable in the modern World of fashion becoming more important and longevity becoming less of a sales issue.
While we too have used the best of all Nikasil solution in our rebuild options - we are still trying to find ways to make lokasil work or last long enough to be acceptable. This may prove fruitless and a waste of money or it may result in a breakthrough of sorts (which we obviously think it will or we would not pursue it).
I hope this answers all your thoughts!
Baz
.
Gassing Station | Porsche General | Top of Page | What's New | My Stuff