bazhart
Well-known member
- Joined
- 20 May 2009
- Messages
- 1,343
It is several years since we last conducted the engine running temperature tests that resulted in the provision of the "low temperature thermostat" and we thought we should revisit the tests to be sure our advice provided since is still valid and to check several questions that readers have asked since about how to minimise the risk of bore scoring. Our original tests were carried out at this time of year and we know that in the Summer the temperatures are a little higher and the difference between std and our low temperature thermostat is slightly reduced by about 2 degrees.
In a standard 3.8 997 (with no Hartech machining mods to the coolant flow in the blocks i.e. as your standard car would be) but with 6 temperature sensors fitted to the engine and simply with first a standard and then a Hartech Low Temperature Thermostat we have just conducted back to back tests on the same day, routes and speeds etc and recorded temperatures driving, stationary, after ticking over for 2 and 5 minutes, switched off and on again after 5 and 30 minutes and at town and fast road speeds with the following results.
(1) Despite the temperatures recorded varying from 79 to 104 the dash board temperature gauge read 80 for every reading - confirming that the instrument in our test car has a buffer zone of about 25 degrees during which it does not change nor reflect what is going on in the engine or the coolant. This may lead to some who have fitted such a thermostat to question if it was doing its job! which it probably will be (and certainly will be if it is one of ours) .
(2) With the standard thermostat - on average the thrust side of bank 2 ran 5 degrees hotter than the thrust side of bank 1 @ 93 and 98 respectively. In contrast- with the Hartech thermostat - the difference was only 1.6 degrees @ 83.8 and 85.4 and this shows that the thrust side of bank 2 is now running cooler than the original thrust side of bank 1 (where we do not experience bore scoring failures).
(3) The oil temperature was on average 9.9 degrees cooler with the Hartech Thermostat (100 and 90.1).
(4) Whether driving slowly in traffic or fast on an open road the temperature differences between bank 1 and bank 2 were not significantly high (because at slow speeds there is little temperature generated anyway and at high speeds both the engine revs and air speed through the radiator are higher and overall cooling very efficient).
(5) The major difference was obvious when the car was either stopped but running on tickover for a few minutes or stopped for a few minutes with the engine switched off and started again. In both cases bank 2 thrust side increased temperatures much higher than bank 1 (because the contra-flow works on bank 1 and maintains a more even temperature gradient inside the engine whereas bank 2 exists with a higher temperature range that soaks out when the coolant flow reduces or is stopped) . On tickover or with the engine stopped - the oil around the piston in bank 1 and bank 2 would naturally gravitate to the bottom this would leave the piston in bank 2 thrust side (the top) with less oil on the thrust face for a restart and much hotter and thinner to resist the high torque loads possible driving off again - while bank 1 piston would be cooler and carry more oil on its thrust face (the bottom) for the re-start. This confirms our advice to drive off modestly after a period stopped at say lights or for a brief engine "off and on" stop before opening the engine up again. It also confirms that high speed driving at higher revs should not be a problem (as it has not been also in our race car engines).
(6) With both thermostats bank 2 cylinder head ran slightly cooler than bank 1 (96.7 to 98.1 std and 87.2 to 89.2 Hartech).
(7) The engine under test is fitted with pistons with yet another pair of different coatings and when it is stripped for the first time to check wear rates - we will modify the coolant flow to reproduce our usual rebuild changes and test out the results of those again too.
(8) In general the fact that the thermostat is fitted on engine entry did not de-stabilise the temperatures but of course resulted in much higher internal running temperatures and thinner oil film strength than was the case when thermostats were fitted to the outlet. Assuming the same thermostat was fitted either before or after the cylinder block - the difference in internal cylinder temperatures would be about 8 or 10 degrees. The fact that the std thermostat is rated higher than those tradditionally used in more conventional engine coolant flow layouts would also mean that the internal engine temperatures std are probably in the region of 10 to 15 degrees hotter than they used to be (and all to assist the emission reductions required to export cars to major outlets).
Similarly with the more traditional layout the higher internal coolant temperatures were maintained the same as the engine usage was increased through faster more aggressive driving whereas with the thermostat on the inlet - as the engine is worked harder - so too does the internal temperature range increase.
I am not sure if it is an advantage to use a thicker oil if a lower temperature thermostat has been fitted (as the running temperature reductions will inevitably result in the same oil holding a higher viscosity) but personally - once any new parts have been run in - I still think an upgrade (as Wizard proclaims( is a benefit likely to extend the engines life).
It must be remembered that fitting a low temperature thermostat will not repair any existing wear rates that might have worn down the piston coasting somewhat already and therefore although running the engine at a cooler temperature may have avoided bore scoring all together if adopted when the car was new - with scored bores on bank 2 having occurred @ 9000 miles (but more usually about 30 to 60K) and as yet no evidence of scoring on bank 1 even up to 100K - it does seem that the temperature difference between the banks is what promotes early bank 2 failures and therefore although reducing those temperatures in bank 2 should extend piston life significantly, it is impossible to establish how worn they may already be at the time of fitting the low temperature thermostat and therefore being unable to predict how long the benefit will last for afterwards.
Although the low temperature thermostat is a low cost way of improving reliability - we are still working on hopefully even better solutions both for retro-fitting new designed parts to existing cars and during rebuilds. While fitting 6 Nikasil cylinders removes the problem all together - it is still expensive and so despite trying several different coatings unsuccessfully to date - we are still working on finding a reliable piston coating that may enable us to overbore scored cylinders that combined with changing the coolant and cylinder block temperatures - may eventually reduce rebuild costs with much improved long term reliability to "Nikasil" levels.
Finally I am not sure why there seems to be some form of competition to demonstrate where readers obtained their own lower temperature thermostats from and at what price - ours is just £35 + Vat and these test results only apply to our thermostat and do not necessarily apply to any others.
Have a great New Year - and "KEEP COOL"
Baz
In a standard 3.8 997 (with no Hartech machining mods to the coolant flow in the blocks i.e. as your standard car would be) but with 6 temperature sensors fitted to the engine and simply with first a standard and then a Hartech Low Temperature Thermostat we have just conducted back to back tests on the same day, routes and speeds etc and recorded temperatures driving, stationary, after ticking over for 2 and 5 minutes, switched off and on again after 5 and 30 minutes and at town and fast road speeds with the following results.
(1) Despite the temperatures recorded varying from 79 to 104 the dash board temperature gauge read 80 for every reading - confirming that the instrument in our test car has a buffer zone of about 25 degrees during which it does not change nor reflect what is going on in the engine or the coolant. This may lead to some who have fitted such a thermostat to question if it was doing its job! which it probably will be (and certainly will be if it is one of ours) .
(2) With the standard thermostat - on average the thrust side of bank 2 ran 5 degrees hotter than the thrust side of bank 1 @ 93 and 98 respectively. In contrast- with the Hartech thermostat - the difference was only 1.6 degrees @ 83.8 and 85.4 and this shows that the thrust side of bank 2 is now running cooler than the original thrust side of bank 1 (where we do not experience bore scoring failures).
(3) The oil temperature was on average 9.9 degrees cooler with the Hartech Thermostat (100 and 90.1).
(4) Whether driving slowly in traffic or fast on an open road the temperature differences between bank 1 and bank 2 were not significantly high (because at slow speeds there is little temperature generated anyway and at high speeds both the engine revs and air speed through the radiator are higher and overall cooling very efficient).
(5) The major difference was obvious when the car was either stopped but running on tickover for a few minutes or stopped for a few minutes with the engine switched off and started again. In both cases bank 2 thrust side increased temperatures much higher than bank 1 (because the contra-flow works on bank 1 and maintains a more even temperature gradient inside the engine whereas bank 2 exists with a higher temperature range that soaks out when the coolant flow reduces or is stopped) . On tickover or with the engine stopped - the oil around the piston in bank 1 and bank 2 would naturally gravitate to the bottom this would leave the piston in bank 2 thrust side (the top) with less oil on the thrust face for a restart and much hotter and thinner to resist the high torque loads possible driving off again - while bank 1 piston would be cooler and carry more oil on its thrust face (the bottom) for the re-start. This confirms our advice to drive off modestly after a period stopped at say lights or for a brief engine "off and on" stop before opening the engine up again. It also confirms that high speed driving at higher revs should not be a problem (as it has not been also in our race car engines).
(6) With both thermostats bank 2 cylinder head ran slightly cooler than bank 1 (96.7 to 98.1 std and 87.2 to 89.2 Hartech).
(7) The engine under test is fitted with pistons with yet another pair of different coatings and when it is stripped for the first time to check wear rates - we will modify the coolant flow to reproduce our usual rebuild changes and test out the results of those again too.
(8) In general the fact that the thermostat is fitted on engine entry did not de-stabilise the temperatures but of course resulted in much higher internal running temperatures and thinner oil film strength than was the case when thermostats were fitted to the outlet. Assuming the same thermostat was fitted either before or after the cylinder block - the difference in internal cylinder temperatures would be about 8 or 10 degrees. The fact that the std thermostat is rated higher than those tradditionally used in more conventional engine coolant flow layouts would also mean that the internal engine temperatures std are probably in the region of 10 to 15 degrees hotter than they used to be (and all to assist the emission reductions required to export cars to major outlets).
Similarly with the more traditional layout the higher internal coolant temperatures were maintained the same as the engine usage was increased through faster more aggressive driving whereas with the thermostat on the inlet - as the engine is worked harder - so too does the internal temperature range increase.
I am not sure if it is an advantage to use a thicker oil if a lower temperature thermostat has been fitted (as the running temperature reductions will inevitably result in the same oil holding a higher viscosity) but personally - once any new parts have been run in - I still think an upgrade (as Wizard proclaims( is a benefit likely to extend the engines life).
It must be remembered that fitting a low temperature thermostat will not repair any existing wear rates that might have worn down the piston coasting somewhat already and therefore although running the engine at a cooler temperature may have avoided bore scoring all together if adopted when the car was new - with scored bores on bank 2 having occurred @ 9000 miles (but more usually about 30 to 60K) and as yet no evidence of scoring on bank 1 even up to 100K - it does seem that the temperature difference between the banks is what promotes early bank 2 failures and therefore although reducing those temperatures in bank 2 should extend piston life significantly, it is impossible to establish how worn they may already be at the time of fitting the low temperature thermostat and therefore being unable to predict how long the benefit will last for afterwards.
Although the low temperature thermostat is a low cost way of improving reliability - we are still working on hopefully even better solutions both for retro-fitting new designed parts to existing cars and during rebuilds. While fitting 6 Nikasil cylinders removes the problem all together - it is still expensive and so despite trying several different coatings unsuccessfully to date - we are still working on finding a reliable piston coating that may enable us to overbore scored cylinders that combined with changing the coolant and cylinder block temperatures - may eventually reduce rebuild costs with much improved long term reliability to "Nikasil" levels.
Finally I am not sure why there seems to be some form of competition to demonstrate where readers obtained their own lower temperature thermostats from and at what price - ours is just £35 + Vat and these test results only apply to our thermostat and do not necessarily apply to any others.
Have a great New Year - and "KEEP COOL"
Baz