MG TF 160 VVC coolant temperature sensor (ECU) off?

Hi all,

I have a 2004 TF 160 VVC "Sunstorm". Last week, warm day, I drove in town, slow traffic, lots of speed bumps, and after a while the temperature gauge went up to the 3/4 marker. I turned off the engine, opened the boot and steam came out of the pressure cap on the expansion tank. I didn't hear the fans behind the radiator, nor the one in the engine bay kick in. All in all I lost 1 l of coolant. When the engine had cooled down a bit, I drove home. In a traffic jam the temperature gauge again went slowly up, but with heater and blower on full blast I could get it down again to the normal position (just beneath the 1/2 marker).

I searched the forum and found several tips. I checked the fuses, and they were fine. I removed hood and inspection lid, next the plug from the coolant temperature sensor (for the ECU), a black one, started the engine and both engine fan and and radiator fans immediately kicked in. Engine was cold, car had been standing still for days. I had tinkered two small cable terminals isolated with shrink tubing to fit the two prongs inside the sensor, with short cable ends, and hooked up my multi-meter and watched the Ohms meter with the engine running. (I couldn't get the multi-meter pins to fit the prongs ánd see what I was doing…) On the laptop I had Dieters table to check the resistance values. Right after hooking up, engine not running, it read 2.2 kΩ, which corresponded to the outside temperature of about 25 degrees Celsius. After 10 to 15 minutes the Ωs gradually decreased to 170 Ω. The engine at that point was at normal operating temperature (on the temp gauge at least). Both rad fans and engine bay fan were running. I turned off the engine and ignition, removed the multi-meter, hooked up the plug to the ECU, started the engine and both rad fans and engine bay fan remained silent.

According to Dieters table at normal operating temp it should have read 250-300 Ω, instead of 170. This confuses me: if it is a thermistor with Negative Temperature Coefficient, I would have expected the fans to have kicked in earlier instead of not at all when the engine turned hot on my trip through town and in the traffic jam, which they hadn't – although the temp rise in the traffic jam on the way back, could be attributed to the loss of a liter of coolant, I guess. I only topped up the next day, when the engine had cooled down.

Any ideas?

Richard

Forgot to mention that the contact surfaces of plug and sensor were clean as a whistle…

Anyone?

I'm still trying to figure out why the fans didn't kick in when the engine was obviously overheated, with the sensor working as it should, and fuses, fans and ECU apparently okay. (With the plug disconnected, the ECU turned on the fans, so the ECU seems to do that part of its job okay.) I looked up thermistor in Wikipedia, but the description is much too beta for my brain, and read up on common causes of thermistor failure , but I don't recognize any of the symptoms mentioned there: the base resistance was what one would expect, 2.2 kΩ at 20 degrees Celsius, and I found the decrease in resistance value to be gradual, without any strange hickups… The only odd thing was the rather low resistance (170 Ω) at operating temperature… But wouldn't that have prompted the ECU to turn on the fans earlier instead of not at all?

I'm at a loss here…

Richard

I'm not much use on these kind of problems but have you tried a 220 ohm resistor across the plug terminals?
Also probably needs a new pressure cap as it shouldn't have blown with only 3/4 reading on gauge.

A 220 ohm resistor? I don't even know what that is I mean, I googled images of them, but I wouldn't know what to do with them…

I'm going to order a new sensor, MEK000030 if I'm correct, they're only a tenner, and maybe I'll put the old one in a pan of water and keep track of resistance and temperature while I boil it.

You mean to say the pressure cap opened too soon? They should be 1 bar, you said somewhere on MG-rover? I've put a new one ("A-kwaliteit") in my shopping basket.

Richard

From what I remember of my two years of physics in secondary school: lower pressure means lower boiling point, so, if the pressure cap blew off pressure too soon, the coolant may never have reached the temperature (102 degrees C), or the sensor the resistance value, to trigger the ECU to switch on the fans…

Richard

One thing to note is that if the coolant level is very low the two temperature senders may not be immersed in coolant…if that is so, then the temperature senders will not be giving correct information to the ECU or gauge

The coolant has always been on the right level. I check every time. I remove the cap and look inside the tank. I have never had to top up. Except for last week, that is.But that was after I had lost coolant due to the cap blowing off pressure…

I’m busy packing the van for another couple of weeks away but I’ll have a look at the diagrams later If I Get Time. I’ll be taking my netbook with me & I’ll load a load of info on it as we’re off to the land of hills & no signal.
I’m always encouraged when I see someone trying to rationalise their problem prior to turning to the ‘bar & when they have found Dieter’s site & are referencing his data, that implies a questioning mind.
I’ll see what I can find later.
M

Thanks, and have fun in the Land of Hills and No Signal. Is the Airportable of the user name by any chance the airportable aka lightweight in the back in the profile picture?
Richard

Yes that’s the old girl, needing a healthy slice of TLC which my son has promised it, however he’s got another shipment of disturbed nights to negotiate as my daughter in law will be dropping another engineer in a month.
The old thing is too heavy for me to do it justice, although once I get it out of the drive & on the open road I can shed thirty years. Not as good as the forty five shed for the MG.
I’ve not forgotten the diagram, I’ve been working on a circuit which has just dumped a transistor.
M

The diagram didn’t give much away other than the air temperature sensor & the water sensor both feed into the engine control module & if the electrons inside that unit agree it bloody hot they will turn the fans on.
But you knew that already & so I’ve brought nothing new to the party, sorry.
M