Engine oil running cool. 50C on the motorway, 60C on track.
Discussion
I recently installed an oil temp gauge:
Thread here:
http://www.pistonheads.com/gassing/topic.asp?h=0&a...
I think the sensor/gauge are working OK, but the oil temps I'm seeing on my Porsche 968 engine seem low. Takes a good five miles to warm up, 50C at steady 70-75mph cruising, 60C above 75mph and during a track session, rising to 80-85C in traffic. Ambient temps during testing were 12-15C.
The sensor is in the sump plug and has about 2-3mm around the outside of it so oil should be able to circulate. However, the gauge does not protrude outside the sump plug threads, so although oil can get to it, it's not as far as I'd like into the sump. Nothing can be done about this unfortunately.
I suspected a failed oil temp thermostat, but after investigating it seems the default position would be for oil not to reach the oil cooler if this has failed. The undertray was removed so I improvised something at the weekend. It didn't make any difference to the 50C cruising temp.
Somebody suggested running the sensor from an oil cooler line, but if the thermostat diverts oil away from this below the 95C setting, surely it won't read accurate temperature before then.
The car is a 924 Carrera GT replica so has a hood scoop. On the real GT this runs into an intercooler (although on the GTS the intercooler is at the front of the engine so the scoop there is also a dummy) that I suspect is providing a fair amount of cool air when on the move, but I'd be surprised if it can get all the way around the engine to cool the oil.
Water temps are 69C on the move rising to 85C in traffic. The car runs great, I'm just surprised it's running so cool.
Any ideas?
Thread here:
http://www.pistonheads.com/gassing/topic.asp?h=0&a...
I think the sensor/gauge are working OK, but the oil temps I'm seeing on my Porsche 968 engine seem low. Takes a good five miles to warm up, 50C at steady 70-75mph cruising, 60C above 75mph and during a track session, rising to 80-85C in traffic. Ambient temps during testing were 12-15C.
The sensor is in the sump plug and has about 2-3mm around the outside of it so oil should be able to circulate. However, the gauge does not protrude outside the sump plug threads, so although oil can get to it, it's not as far as I'd like into the sump. Nothing can be done about this unfortunately.
I suspected a failed oil temp thermostat, but after investigating it seems the default position would be for oil not to reach the oil cooler if this has failed. The undertray was removed so I improvised something at the weekend. It didn't make any difference to the 50C cruising temp.
Somebody suggested running the sensor from an oil cooler line, but if the thermostat diverts oil away from this below the 95C setting, surely it won't read accurate temperature before then.
The car is a 924 Carrera GT replica so has a hood scoop. On the real GT this runs into an intercooler (although on the GTS the intercooler is at the front of the engine so the scoop there is also a dummy) that I suspect is providing a fair amount of cool air when on the move, but I'd be surprised if it can get all the way around the engine to cool the oil.
Water temps are 69C on the move rising to 85C in traffic. The car runs great, I'm just surprised it's running so cool.
Any ideas?
Pumaracing said:
I'd be very surprised if those were true oil temps. I suggest your sender / gauge is not working properly.
Thanks. The gauge reads room temperature accurately between 10 and 25C, but I wonder if when attached to the car heat sink is causing the sender to read the temperature of the sump, which is being cooled by the airflow, rather than the oil within it. Unless I can buy a longer temperature sender to further protrude into the oil it looks like I'll have to go to a remote oil filter housing via a sandwich plate adapter. A straight sandwich plate adapter alone won't fit.I'll also blank off the cooler, and ordered a new thermostat today just in case. £41+VAT from Porsche.
Back to the drawing board...
Can you not buy a pukka oil temperature sender designed to screw into your sump plug thread like the ones VDO do for VWs? I have no idea what your sump plug thread dimensions are but 14mm x 1.5mm is a common size.
Edit. Ok looked at the previous thread and it seems like your sump plug is a lot bigger than that if it has a 20mm hex socket size.
Edit. Ok looked at the previous thread and it seems like your sump plug is a lot bigger than that if it has a 20mm hex socket size.
Edited by Pumaracing on Monday 28th October 13:46
I wonder if you are measuring the ally sump rather than the oil? it may be worth borrowing a temperature probe to see whats going on ( useful for checking the oil stat too). the only oil sender i have fitted in a 968 was a couple of inches to the right ( in your pic) and higher. obviously the sump had to be drilled.
LarJammer said:
I wonder if you are measuring the ally sump rather than the oil? it may be worth borrowing a temperature probe to see whats going on ( useful for checking the oil stat too). the only oil sender i have fitted in a 968 was a couple of inches to the right ( in your pic) and higher. obviously the sump had to be drilled.
I appreciate the info, thanks. Did you use a 1/8 NPT sender and just screw it into the hole drilled in the sump?The fact that the supposed oil temp is low when the car is moving fast and there's plenty of airflow but hotter when the car is in traffic suggests that the temp you are actually measuring is the sump alloy temp not the oil temp. More protusion into the oil from the sensor tip is required methinks.
Maybe the sump plug can be machined down a bit to let the sensor stick out further.
Maybe the sump plug can be machined down a bit to let the sensor stick out further.
Had the exact same issue with my oil temp when I fitted it... I used the table provided and it read ambient ok but showed 43deg c on motorway and would cheap up in traffic.
Repeated attempts to map it failed and it was quite frustrating and worrying.... I ended up fitting a more expensive bosch unit (full £25) which came with a pre map for my ecu... Now reads accurately and all is well :-)
So my experience of cheap sensors is they are not worth the hassle or worry
Repeated attempts to map it failed and it was quite frustrating and worrying.... I ended up fitting a more expensive bosch unit (full £25) which came with a pre map for my ecu... Now reads accurately and all is well :-)
So my experience of cheap sensors is they are not worth the hassle or worry
This is a SPA unit though so I doubt it's a quality issue. I'll test it at higher temperatures off the car.
As others have said, I think the problem is that the sensor doesn't protrude enough into the oil. There's about 2-3mm around the sensor to the sump plug threads, but the sensor doesn't stick out beyond the thread. Perhaps the circulation isn't good enough for this to offset the cooling effect of the sump. SPA doesn't sell a longer sensor.
Rather than reduce the size of the plug I'll follow LarJammer's idea of drilling into the sump, so the threads of the sender are covered by the sump alloy but the sensor itself has a good flow of oil around it. I've resisted drilling the sump until now but suspect it's going to be the best solution.
I'll post pictures when I am done but it might be a few weeks if I don't find the time this weekend.
As others have said, I think the problem is that the sensor doesn't protrude enough into the oil. There's about 2-3mm around the sensor to the sump plug threads, but the sensor doesn't stick out beyond the thread. Perhaps the circulation isn't good enough for this to offset the cooling effect of the sump. SPA doesn't sell a longer sensor.
Rather than reduce the size of the plug I'll follow LarJammer's idea of drilling into the sump, so the threads of the sender are covered by the sump alloy but the sensor itself has a good flow of oil around it. I've resisted drilling the sump until now but suspect it's going to be the best solution.
I'll post pictures when I am done but it might be a few weeks if I don't find the time this weekend.
Evoluzione said:
As a rough rule, shouldn't it be the case that oil temp is the same as water temp?
Let's say, both optimum at about 85'c ?
In an ideal world oil temp should be around 100c and a minimum of 85c. Too cold is bad for both evaporation of water and other contaminants and also for viscosity and hence engine power. Too hot of course and the viscosity drops like a stone and bearings don't get properly lubricated.Let's say, both optimum at about 85'c ?
When I was building a lot of race engines for the XR2 Challenge (Fiesta 1.6 CVH engine) where stupidly the rules didn't allow oil coolers we found that 130c was like an absolute switch for engines grenading. None of mine happily because I built to very tight bearing clearances and uprated pressure relief valve settings to compensate for the reduced viscosity but I remember one very hot summer day's race at Oulton Park where 8 or 9 of other people's engines blew up one after the other, all at the exact same spot, just as the drivers lifted off at the end of the main straight. As the cylinder pressure load on the piston crown fell when the throttle closed the reciprocating forces on the big end bearings ceased to be cushioned by it and immediate metal to metal contact with resultant cloud of smoke and unhappy wallets.
As to ideal water temp, within limits colder is slightly better for power as it increases charge density and hence air/fuel mass in the cylinders. For a race engine 80c is nice. For a road engine the OE manufacturer has already fitted the correct thermostat for you somewhere in the 82c to 88c range.
So generally oil temp will be 10c to 15c hotter than water temp with a max of 110c. Any more than this and an oil cooler is needed.
Pumaracing said:
In an ideal world oil temp should be around 100c and a minimum of 85c. Too cold is bad for both evaporation of water and other contaminants and also for viscosity and hence engine power. Too hot of course and the viscosity drops like a stone and bearings don't get properly lubricated.
When I was building a lot of race engines for the XR2 Challenge (Fiesta 1.6 CVH engine) where stupidly the rules didn't allow oil coolers we found that 130c was like an absolute switch for engines grenading. None of mine happily because I built to very tight bearing clearances and uprated pressure relief valve settings to compensate for the reduced viscosity but I remember one very hot summer day's race at Oulton Park where 8 or 9 of other people's engines blew up one after the other, all at the exact same spot, just as the drivers lifted off at the end of the main straight. As the cylinder pressure load on the piston crown fell when the throttle closed the reciprocating forces on the big end bearings ceased to be cushioned by it and immediate metal to metal contact with resultant cloud of smoke and unhappy wallets.
As to ideal water temp, within limits colder is slightly better for power as it increases charge density and hence air/fuel mass in the cylinders. For a race engine 80c is nice. For a road engine the OE manufacturer has already fitted the correct thermostat for you somewhere in the 82c to 88c range.
So generally oil temp will be 10c to 15c hotter than water temp with a max of 110c. Any more than this and an oil cooler is needed.
Thanks for the input. Often in Winter it's difficult to get the oil temp up to where you need it to be with a conventional oil cooler so we have combined oil and water coolers made for the integrales, the oil cooling matrix runs in the end tank of the water rad, so the two temps are linked - it cools oil in summer and heats it in Winter.When I was building a lot of race engines for the XR2 Challenge (Fiesta 1.6 CVH engine) where stupidly the rules didn't allow oil coolers we found that 130c was like an absolute switch for engines grenading. None of mine happily because I built to very tight bearing clearances and uprated pressure relief valve settings to compensate for the reduced viscosity but I remember one very hot summer day's race at Oulton Park where 8 or 9 of other people's engines blew up one after the other, all at the exact same spot, just as the drivers lifted off at the end of the main straight. As the cylinder pressure load on the piston crown fell when the throttle closed the reciprocating forces on the big end bearings ceased to be cushioned by it and immediate metal to metal contact with resultant cloud of smoke and unhappy wallets.
As to ideal water temp, within limits colder is slightly better for power as it increases charge density and hence air/fuel mass in the cylinders. For a race engine 80c is nice. For a road engine the OE manufacturer has already fitted the correct thermostat for you somewhere in the 82c to 88c range.
So generally oil temp will be 10c to 15c hotter than water temp with a max of 110c. Any more than this and an oil cooler is needed.
When we built 'Project Sausage' I sourced an off-the-shelf (combo) one in the US for very little money, it was destined for the front of one of those Ford/GM pick-up trucks they like over there. Being what the project was all about I just cut the car 'till it fitted.
LarJammer said:
I wonder if you are measuring the ally sump rather than the oil? it may be worth borrowing a temperature probe to see whats going on ( useful for checking the oil stat too). the only oil sender i have fitted in a 968 was a couple of inches to the right ( in your pic) and higher. obviously the sump had to be drilled.
100% this. Had it on my caterham, might also be the fitting going into the sump sinking heat from the sender tip, on mine it extended right out into the airflow so I wrapped in a bike inner tube and a few rounds of insulating tape and it helped massively indicated temps came up by about 20degCGassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff