Turbo Coolant Temperature Sensor test?

[Spencan]

After getting into boost I noticed that the temperature gauge would rise. Then cool back down after calm driving. I also noticed that my auxiliary turbo water pump was not turning on when I turned the car off, like it used to. The fans came on but not the water pump.

So I tested relay and the results showed that it is working properly. Then I jumpered the pump from where the relay is located and the pump turned on.

So now I’m wondering if it’s the temperature sensor itself or the wiring. How do I test to find out which one is the culprit?

[cda951]

The auxiliary water pump is only used after engine shutoff to circulate coolant through the turbocharger bearing housing ----it does NOT function while the engine is running. The turbocharger cooling circuit should always have coolant flowing through it (via the main engine water pump) as long as the small turbo thermostat (mounted inside the plastic flange on the water pump) is open. I have never heard of one of these being stuck closed, but it is worth checking.

The term "analog" is thrown around entirely too much within the enthusiast car scene in 2024, but the 944 has a truly analog coolant temperature gauge circuit, the needle shows the temperature in real time, unlike the electronically dampened circuits of newer cars. Where does the gauge needle reside during normal driving, and how much does it rise during periods of boost? And how long of periods of boost cause it to rise? I'd also look into the engine oil cooler and its circuit/thermostat. Again, have never heard of one being stuck closed, but the engine oil and coolant temperatures are very closely correlated during periods of extended boost because both circuits run in close proximity within the turbo bearing housing. This is the reason why many track-focused 951s eliminate the turbo coolant circuit. I chose to keep it because street car, and the second factory 951 oil cooler that I plumbed in series with the original keeps engine oil and coolant temps in check, even during a 100 degrees F track event (which I refuse to do anymore, not much fun!).

But to answer your original question, the auxiliary water pump should run for ~30 seconds after any time the engine is switched off, hot or cold. The turbo temperature SWITCH (NOT a sender!) keeps the pump running for longer if the coolant temperature is above the rating of the switch ( I think about 105 or 110 degrees C, it's stamped on the switch). A K27-sized turbocharger needs a shorter temperature switch, and I at first made the mistake of installing a VDO temperature sender (varies resistance with temperature instead of merely switching a path to ground on or off at a certain temperature), so there was always a path to ground and the auxiliary water pump ran all the time after the engine was shut off . . . .

[Spencan]

Thanks for the overview. The pump doesn’t run after shutdown. So how do I know if it’s the wiring or metal sending unit?

[Spencan]

I reread your post again and I’m not sure what you’re referring to “turbo temperature switch”. Is that what I have in my picture or are you referring to the thermoswitch on radiator? If it’s the one in my picture, do happen to have a part # for the correct one for my K-27?

I recently replaced all of the o-rings and the gasket on my oil cooler (2 weeks ago). What could I have done wrong to possibly cause this condition?

Thanks in advance!

[Tom]

The switch he is referring to is the one in your picture. He called it a switch because it simply connects the wire to ground when hot enough.

Don't worry about getting a new one of the k27. You only need to do that if the original one does not physically fit in the space available. Yours fits based on the picture, so you should be fine.

You 'could' test the switch by putting it in boiling water and confirming that you get continuity between the thermo-switch's threads and the wire -- only when hot. However, if your pump does not come on when you turn off the motor, your symptoms suggest it is the relay rather than (or in addition to) a bad thermo-switch. You said you tested the relay. Did you test the timer circuit, like in this video?

[Spencan]

Tom,

I did not test it like you did in the video and I will check it tomorrow. Thanks again!

I have been down in Monterey all day today. Sitting in traffic for hours, as well as doing a little spirited accelerations. My temporary “overheating” issue has been found. It’s the result of the bigger, faster spooling turbo. I have an EBC that allows me to adjust how strong the boost initially comes on. I had it on 85% when I had first noticed the temp gauge climbing. Today I turned it down to 50% and, even after sitting in traffic for 30 minutes, getting on the boost had little to no noticeable increase in temperature. I did do WOT pull and it did raise to the halfway to 2/3 but not the next mark on the gauge. Normal driving reduced back within 5 minutes or so. I think that I need to find a way to release that heat or improve airflow in that area.

[Tom]

Were you are Werks yesterday? And/or are you still trying to get out?

Test the relay and let us know. Do you see any coolant overflow after heavy boost? How is your radiator? I replaced mine with a brand new factory unit many years ago and it seriously helped keep the temps down. The original was 'fine' and not leaking, but it's cooling capacity was seriously diminished, presumably from corrosion and deposits over the years. My car now stays at the first hash mark and rarely goes over half, and that's with a 3 liter and big turbo.

[Spencan]

It was my first time going down there today, so I just went to exotics on broadway.

I have not lost any coolant during these events. I’m still running the stock radiator and it s been solid in all conditions except high rpm/high boost. That’s when I see the gauge climb. I was in stop and go traffic and just parked for several minutes today and the gauge didn’t budge from the first mark for hours.

I’ll definitely test it the relay and report back soon.

[Spencan]

I tested my relay to make sure that’s sending the signal to open the switch and it did. It lasted for about 24 seconds. So I ordered a new switch.

Also, I meant to write that I was sitting in traffic for several “hours” NOT “minutes” on Saturday. ..

[Tom]

I tested my relay to make sure that’s sending the signal to open the switch and it did. It lasted for about 24 seconds. So I ordered a new switch.

Also, I meant to write that I was sitting in traffic for several “hours” NOT “minutes” on Saturday. ..

I don't think the switch is your problem. The pump should go on whenever you shut off the car, even if the switch is broken or not installed. I'd be more inclined to swap out the relay, even if it passed the (admittedly partial) test in my video. Try this: pull the connector off the thermoswitch and short the harness connector to the intake. The pump should come on for 24 seconds. Does it?