Don't panic. The cooling system in these cars is simple. Did you change your temp sensors at the engine block? They go bad. Bleed the air out of the head? Also check the water pump is not locked up and belt tension is correct. Most remote possibility is a bad head gasket allowing air to enter the system and disrupt flow of coolant. If all this checks out it could be a lack of airflow over the radiator and condenser.
Cooling fan(s) question. I have an ‘86 turbo and slowly troubleshooting the cooling fans. I completely rewired the resistor wiring from under the dash back to the fuse box. The coolant temp sensor in the radiator is new along with an F9 relay (with diagnostics) so I can manually activate the low/high speeds at the push of a button (vs a jumper on the plug). Here’s what working: High speed fans kick on based on temp and manually Low speed fan runs for a few minutes after shutdown and then turns off BOTH fans never kick on at low speed though. The pass side one does as the temp rises but the driver side does not. I know the driver side fan itself is good because it operates on high speed. I can see the low speed light come on the relay as the temp rises so that tells me the temp sensor doing what it’s supposed to. Should BOTH fans be coming on LOW speed during normal operation vs just the one? The result is that the temp creeps up until the light comes on. The fans will drop the gauge back down but the light didn’t reset and turn off (maybe that’s a separate issue). The gauge hit the top line for maybe 30 seconds and then starting dropping down but the temp light and warning light stayed on. I even shutoff the car and restarted thinking it would need to reset. Because the one fan is coming on based on temp, can I rule out a bleeding issue? thanks Bob
@@bobd9868 There are a few things that are out of place here to me, first off the way the temp switch light works is similar to how the thermo fan switch works, at a certain temperature it grounds the circuit to illuminate the warnings on the dash, but below that threshold the light should come off. So there does seem to be an issue at the very least with your coolant temp sensor or instrument cluster. Second of all unless you vacuum bleed the system I would never rule out a bleeding issue causing weirdness with your dash readings. These things are absolute PAINS to bleed properly. Not saying that is your issue, but I am saying you definitely shouldn’t rule it out. Now as for your cooling fan situation, here’s how my fans have always operated between both my 83 and my 86 NA stock fans and aftermarket fans. Above first mark, low speed fans both sides should come on. At or above the thin white mark at the top of the dash is when the high speed fans should kick on and bring the temp back down rather rapidly. Then the fans will continue to cycle. Check your thermo fan switch and make sure you didn’t plug it in upside down or anything so that it’s giving power the wrong direction. Also you might want to double check you’ve ordered the correct temperature range thermofan switch, if it is not paired with your thermostat properly it can cause issues and even cause the car to overheat because the fans are cooling the fluid in the radiator and the thermostat is not opening even though the block is at operating temp. I’d double check your gauge readings with IR temps to see if when the light comes on the head really is at 100c Lastly, I’ll recommend you to Clark’s garage for the fans trouble shooting, there could be something i am missing since your car is a turbo so I would verify with them as a source. I hope this helps at all and if you continue to have questions let me know
Clarks has some good procedures on testing the gauge, you should also be able to test the sensor as well with those directions, your IR temp readings are going to be close enough, so unless the IR gauge is way off your car is running hot, and hitting 100c while in the garage is off as well...even in an AZ summer. So it looks like you are definitely chasing 2 issues, over-heating and a bad gauge. I've had my 944 sit in the garage running for 40+ minutes in tropical heat (close to 100 at the radiator), and the car easily handled it. Somebody mentioned a head gasket, that is immediately what I thought, mostly because I had a car that did the same thing with over-heating. No oil/coolant mix, and I wasn't really losing any coolant, but the car simply overheated. Doing a simple compression test can help, or use one of the headgasket test kits (they use a chemical reaction and you hold it on the radiator or overflow in the car).
I really hope it’s not a head gasket but I probably should run a compression test just for peace of mind. I might save that for after I replace the radiator if the issue continues.
That is very similar behavior to how my fuel gauge behaved when i first got mine - the needle was untrustworthy but the low fuel light would come on correctly. Have you already cleaned the football shaped grounds inside the gauge cluster? Obviously not the root of your issue, but if those are dirty (specifically the one for that gauge) the needles can act a bit crazy. It might be the case that once cleaned the gauge acts in tandem with the light and you can (more than before) rule out the gauge cluster being a variable.
Also for what it’s worth, the car only overheats when the AC is on, and when I shut the AC off it will cool back down. If I never turn the AC on the car will idle for over 40 minutes and just hang out right above the first line. Last night I cleaned the original sending unit and reinstalled it. Let the car run, much smoother operation of the gauge needle. After about 30 minutes I measured 80C at the lower radiator hose. Does this sound similar to what you saw with your head gasket issue?
I really don't think it is an overheating issue - so you may need not worry about that. 100c at the cylinder head sounds very normal too. I am confused as you are with the warning light - I tried researching a bit, and it is not clear if the red warning light and the coolant temp gauge are two different sensors or the same one? And if they are separate, where is the warning light sensor?
You are correct. One sensor is for the DME to control mixture. This sensor has two spade terminals on it and is for the gauge specifically. One spade controls the needle, and the other controls the warning light.
@@TakeFlightGarage If I had to bet, I'd say it sounds like something's up with the double spade sensor. It doesn't make sense why that light comes when the temp gauge is showing low, and also confirming with an IR gun. Since you had replaced it recently, it could have been DOA.. although, your ac issue may be unrelated
There are technically 2 sensors inside the gauge sensor: one is a resistance vs. temperature sensor (signal goes on one of the two spade terminal). The other inside the same case is an on-off temperature switch (closed when temp passes a certain limit and passes along the second spade terminal).
@@afeudale is there a way to test what temperature triggers the open/ close part of the circuit? If I can determine that for example 90C is causing the warning light to come on, that would demonstrate the needle is in fact correct, but the threshold on the sending unit for the warning light is way too low.
I don’t know what the proper operating temperature of the coolant or cylinder head, but I know that water boils at 100C (212-214F). But depending on your coolant mixture, that will affect your boiling point. Sorry, can’t help you much. Maybe try another sending unit, test two sending units for comparison? By the time you get this message, hop you’ve figured it out.
@@glow9999 I appreciate your feedback. Since this video I decided to completely overhaul the entire cooling system on the car with upgraded modern components, and i actually filmed a series documenting the upgrades on my channel. A year later and the car has been holding up to 115 degree heat
Okay great. The highest I’ve ever seen on that hose even when the block reads 100c i believe is somewhere around 80-90c. So this tells me that when it’s hitting 90 it is getting warm, but I shouldn’t be at risk for melting the engine.
I would trust your handheld gauge for now - I checked over on Clarks Garage and anything sub 100C is fine and actually normal to get that high. I got nervous last year with high temps as well then researched that it's designed this way (to bounce off the high limit and then back down again). Trying to "cool it down" with a different thermostat, etc. was not recommended. You just don't want to go into the red solid zone. Also whipped out the electrical schematics and the gauge sensor is two conductor - one is a variable resistor which shows the temp level and the other conductor is the "alarm" which turns on the temp warning light on the gauge. They operate independently afaik, even though being in the same sensor. Think of the second one as a backup for the first. Two sensors in one. Sooooo, my first guess is the temp sensor you ordered is crap or miscalibrated or wrong part. It might fit in the hole, but have a different temp range vs. resistance. Get a genuine Porsche one from Pelican or 944online. More info on Clark's here: www.clarks-garage.com/shop-manual/elect-19.htm The other temp sensor in the block (again, afaik) does not affect the gauges at all - it goes straight to the DME for it to do it's calculations based on temperature. It might affect how the engine runs when it gets hot by changing certain running parameters, but you won't see it on the gauges. The gauge is controlled by the longer shaft sensor.
Thanks for all your input. I did initial testing for gauge readings but haven’t recorded resistance readings or checked the dash yet. Here are the results: rennlist.com/forums/944-and-944s-forum/1357143-now-iim-really-confused.html
@@TakeFlightGarage I took a quick look and it seems that both old and new sensors read the same on the gauge for the same water temperatures, right? If so, it's possible that your gauge isn't working properly - either bad grounds or bad gauge itself. A way to test the gauge is to use a variable resistor (potentiometer). You can use a 1kohm one. Disconnect the sensor. Connect the middle leg of the potentiometer to ground and one of the other two legs to the terminal that would go from the sensor to the gauge. Sweep through the resistance and watch how the gauge moves. I believe Clarks has this procedure as well. Then when the gauge is at 80c, measure the potentiometer resistance, likewise for 90c, etc. They should match with the values on Clarks site. That would confirm if the gauge works (which again, if it's not, could be grounds or gauge itself).
@@afeudale I didn’t, I couldn’t get it to trigger with my quick tests. The gauges did read differently though. The sending unit I purchased wouldn’t reach the first mark until the water was 90c or above. It was reading 10 degrees too low and then would stay there. The original VDO gauge was off as well but only by about 5 degrees, and the needle moves properly when the car overheats with that sending unit in. I decided to stop wasting time with the sending unit I purchased and returned it because it was clearly defective. I spent an hour cleaning the contacts and the body of the sending unit and reinstalled it into the car. With it reinstalled and the car running again I ran it for another 30 minute idle test. The temp rose normally until it rested just above the above the first line on the dash and it just stayed planted there. I didn’t bother turning the air conditioning on, I just let the car sit and repeatedly took IR temps at the outlet from the engine block and the lower radiator hose leading into the water pump. The temps stayed about 10 degrees apart. As the upper outlet reached 90c the lower rad hose hovered around 80-81. Tomorrow if I have a chance I’ll take resistance readings, and then let the car run with the AC on until it overheats and then take resistance readings again.
Don't panic. The cooling system in these cars is simple. Did you change your temp sensors at the engine block? They go bad. Bleed the air out of the head? Also check the water pump is not locked up and belt tension is correct. Most remote possibility is a bad head gasket allowing air to enter the system and disrupt flow of coolant. If all this checks out it could be a lack of airflow over the radiator and condenser.
Cooling fan(s) question. I have an ‘86 turbo and slowly troubleshooting the cooling fans. I completely rewired the resistor wiring from under the dash back to the fuse box. The coolant temp sensor in the radiator is new along with an F9 relay (with diagnostics) so I can manually activate the low/high speeds at the push of a button (vs a jumper on the plug).
Here’s what working:
High speed fans kick on based on temp and manually
Low speed fan runs for a few minutes after shutdown and then turns off
BOTH fans never kick on at low speed though. The pass side one does as the temp rises but the driver side does not. I know the driver side fan itself is good because it operates on high speed. I can see the low speed light come on the relay as the temp rises so that tells me the temp sensor doing what it’s supposed to.
Should BOTH fans be coming on LOW speed during normal operation vs just the one?
The result is that the temp creeps up until the light comes on. The fans will drop the gauge back down but the light didn’t reset and turn off (maybe that’s a separate issue).
The gauge hit the top line for maybe 30 seconds and then starting dropping down but the temp light and warning light stayed on. I even shutoff the car and restarted thinking it would need to reset.
Because the one fan is coming on based on temp, can I rule out a bleeding issue?
thanks
Bob
@@bobd9868 There are a few things that are out of place here to me, first off the way the temp switch light works is similar to how the thermo fan switch works, at a certain temperature it grounds the circuit to illuminate the warnings on the dash, but below that threshold the light should come off. So there does seem to be an issue at the very least with your coolant temp sensor or instrument cluster.
Second of all unless you vacuum bleed the system I would never rule out a bleeding issue causing weirdness with your dash readings. These things are absolute PAINS to bleed properly. Not saying that is your issue, but I am saying you definitely shouldn’t rule it out.
Now as for your cooling fan situation, here’s how my fans have always operated between both my 83 and my 86 NA stock fans and aftermarket fans.
Above first mark, low speed fans both sides should come on.
At or above the thin white mark at the top of the dash is when the high speed fans should kick on and bring the temp back down rather rapidly. Then the fans will continue to cycle.
Check your thermo fan switch and make sure you didn’t plug it in upside down or anything so that it’s giving power the wrong direction. Also you might want to double check you’ve ordered the correct temperature range thermofan switch, if it is not paired with your thermostat properly it can cause issues and even cause the car to overheat because the fans are cooling the fluid in the radiator and the thermostat is not opening even though the block is at operating temp.
I’d double check your gauge readings with IR temps to see if when the light comes on the head really is at 100c
Lastly, I’ll recommend you to Clark’s garage for the fans trouble shooting, there could be something i am missing since your car is a turbo so I would verify with them as a source.
I hope this helps at all and if you continue to have questions let me know
Clarks has some good procedures on testing the gauge, you should also be able to test the sensor as well with those directions, your IR temp readings are going to be close enough, so unless the IR gauge is way off your car is running hot, and hitting 100c while in the garage is off as well...even in an AZ summer. So it looks like you are definitely chasing 2 issues, over-heating and a bad gauge. I've had my 944 sit in the garage running for 40+ minutes in tropical heat (close to 100 at the radiator), and the car easily handled it.
Somebody mentioned a head gasket, that is immediately what I thought, mostly because I had a car that did the same thing with over-heating. No oil/coolant mix, and I wasn't really losing any coolant, but the car simply overheated. Doing a simple compression test can help, or use one of the headgasket test kits (they use a chemical reaction and you hold it on the radiator or overflow in the car).
I really hope it’s not a head gasket but I probably should run a compression test just for peace of mind. I might save that for after I replace the radiator if the issue continues.
That is very similar behavior to how my fuel gauge behaved when i first got mine - the needle was untrustworthy but the low fuel light would come on correctly. Have you already cleaned the football shaped grounds inside the gauge cluster? Obviously not the root of your issue, but if those are dirty (specifically the one for that gauge) the needles can act a bit crazy. It might be the case that once cleaned the gauge acts in tandem with the light and you can (more than before) rule out the gauge cluster being a variable.
Noted I should probably pull the cluster again and clean those grounds
Also for what it’s worth, the car only overheats when the AC is on, and when I shut the AC off it will cool back down. If I never turn the AC on the car will idle for over 40 minutes and just hang out right above the first line. Last night I cleaned the original sending unit and reinstalled it. Let the car run, much smoother operation of the gauge needle. After about 30 minutes I measured 80C at the lower radiator hose. Does this sound similar to what you saw with your head gasket issue?
I really don't think it is an overheating issue - so you may need not worry about that. 100c at the cylinder head sounds very normal too. I am confused as you are with the warning light - I tried researching a bit, and it is not clear if the red warning light and the coolant temp gauge are two different sensors or the same one? And if they are separate, where is the warning light sensor?
You are correct. One sensor is for the DME to control mixture. This sensor has two spade terminals on it and is for the gauge specifically. One spade controls the needle, and the other controls the warning light.
@@TakeFlightGarage If I had to bet, I'd say it sounds like something's up with the double spade sensor. It doesn't make sense why that light comes when the temp gauge is showing low, and also confirming with an IR gun. Since you had replaced it recently, it could have been DOA.. although, your ac issue may be unrelated
@@Seagrams I have similar suspicions myself.
There are technically 2 sensors inside the gauge sensor: one is a resistance vs. temperature sensor (signal goes on one of the two spade terminal). The other inside the same case is an on-off temperature switch (closed when temp passes a certain limit and passes along the second spade terminal).
@@afeudale is there a way to test what temperature triggers the open/ close part of the circuit? If I can determine that for example 90C is causing the warning light to come on, that would demonstrate the needle is in fact correct, but the threshold on the sending unit for the warning light is way too low.
I don’t know what the proper operating temperature of the coolant or cylinder head, but I know that water boils at 100C (212-214F). But depending on your coolant mixture, that will affect your boiling point. Sorry, can’t help you much. Maybe try another sending unit, test two sending units for comparison? By the time you get this message, hop you’ve figured it out.
@@glow9999 I appreciate your feedback. Since this video I decided to completely overhaul the entire cooling system on the car with upgraded modern components, and i actually filmed a series documenting the upgrades on my channel. A year later and the car has been holding up to 115 degree heat
Can you take a temp reading at the coolant hose on top going to the t-stat?
Isn’t the Tstat in the water pump housing on the lower hose?
On the upper radiator hose I see about 80c and on the power hose I measure about 70c
@@TakeFlightGarage the lower hose should give you a real world water temp which is what the gauge is reading.
Okay great. The highest I’ve ever seen on that hose even when the block reads 100c i believe is somewhere around 80-90c. So this tells me that when it’s hitting 90 it is getting warm, but I shouldn’t be at risk for melting the engine.
@@TakeFlightGarage what part of Phoenix are you in?
I would trust your handheld gauge for now - I checked over on Clarks Garage and anything sub 100C is fine and actually normal to get that high. I got nervous last year with high temps as well then researched that it's designed this way (to bounce off the high limit and then back down again). Trying to "cool it down" with a different thermostat, etc. was not recommended. You just don't want to go into the red solid zone. Also whipped out the electrical schematics and the gauge sensor is two conductor - one is a variable resistor which shows the temp level and the other conductor is the "alarm" which turns on the temp warning light on the gauge. They operate independently afaik, even though being in the same sensor. Think of the second one as a backup for the first. Two sensors in one. Sooooo, my first guess is the temp sensor you ordered is crap or miscalibrated or wrong part. It might fit in the hole, but have a different temp range vs. resistance. Get a genuine Porsche one from Pelican or 944online. More info on Clark's here: www.clarks-garage.com/shop-manual/elect-19.htm
The other temp sensor in the block (again, afaik) does not affect the gauges at all - it goes straight to the DME for it to do it's calculations based on temperature. It might affect how the engine runs when it gets hot by changing certain running parameters, but you won't see it on the gauges. The gauge is controlled by the longer shaft sensor.
Thanks for all your input. I did initial testing for gauge readings but haven’t recorded resistance readings or checked the dash yet. Here are the results:
rennlist.com/forums/944-and-944s-forum/1357143-now-iim-really-confused.html
@@TakeFlightGarage I took a quick look and it seems that both old and new sensors read the same on the gauge for the same water temperatures, right? If so, it's possible that your gauge isn't working properly - either bad grounds or bad gauge itself. A way to test the gauge is to use a variable resistor (potentiometer). You can use a 1kohm one. Disconnect the sensor. Connect the middle leg of the potentiometer to ground and one of the other two legs to the terminal that would go from the sensor to the gauge. Sweep through the resistance and watch how the gauge moves. I believe Clarks has this procedure as well. Then when the gauge is at 80c, measure the potentiometer resistance, likewise for 90c, etc. They should match with the values on Clarks site. That would confirm if the gauge works (which again, if it's not, could be grounds or gauge itself).
@@TakeFlightGarage also, did you find out what temperature makes it "alarm" the warning light?
@@afeudale I didn’t, I couldn’t get it to trigger with my quick tests. The gauges did read differently though. The sending unit I purchased wouldn’t reach the first mark until the water was 90c or above. It was reading 10 degrees too low and then would stay there. The original VDO gauge was off as well but only by about 5 degrees, and the needle moves properly when the car overheats with that sending unit in. I decided to stop wasting time with the sending unit I purchased and returned it because it was clearly defective. I spent an hour cleaning the contacts and the body of the sending unit and reinstalled it into the car. With it reinstalled and the car running again I ran it for another 30 minute idle test. The temp rose normally until it rested just above the above the first line on the dash and it just stayed planted there. I didn’t bother turning the air conditioning on, I just let the car sit and repeatedly took IR temps at the outlet from the engine block and the lower radiator hose leading into the water pump. The temps stayed about 10 degrees apart. As the upper outlet reached 90c the lower rad hose hovered around 80-81. Tomorrow if I have a chance I’ll take resistance readings, and then let the car run with the AC on until it overheats and then take resistance readings again.