Water temp sensor fault

Danroyd

Member
I got a consult cable to see why my car fans weren't switching on when the car got hot and i can see now its tripping out it gos from 80degrees to 0 then back up continuously, its a new sensor , ive checked the continuity on the 2 wires back to the ecu and there good so now in stumped has anyone got any idea's? How do the ground wires going to the ecu work?
 

Mr B

Member
Have you checked the 5v reference signal at sensor harness to the sensor is correct
have you check earth continuity from sensor harness to engine ground
Have you checked/cleaned ecu harness plug/pins, sensor harness plug/pins
Have you pulled sensor out and tested it in heating water with multimeter on sensor, just because it new sensor doesn't mean it good sensor (aftermarket sensor quality can be awful)
What values ecu get with sensor disconnected and disconnected with pins bridged (fans should run bridged)
Also have you checked fan operation either by switching fans via diag software or manually by grounding pin 9 on ecu .
FSM sensor info in link below:
https://drive.google.com/file/d/0B7Ub2apq5-pfMDRwZHhuRktnLVk/view
 

keastygtir

Well-Known Member
Have you checked the 5v reference signal at sensor harness to the sensor is correct
have you check earth continuity from sensor harness to engine ground
Have you checked/cleaned ecu harness plug/pins, sensor harness plug/pins
Have you pulled sensor out and tested it in heating water with multimeter on sensor, just because it new sensor doesn't mean it good sensor (aftermarket sensor quality can be awful)
What values ecu get with sensor disconnected and disconnected with pins bridged (fans should run bridged)
Also have you checked fan operation either by switching fans via diag software or manually by grounding pin 9 on ecu .
FSM sensor info in link below:
https://drive.google.com/file/d/0B7Ub2apq5-pfMDRwZHhuRktnLVk/view
What he said!
 

Mr B

Member
reference voltage seems good .
earth has continuity but not sure what average resistance value is with ign key off so 12ohm could be classed as suspect at moment.
I would assume the 30 to 80 sensor disconnected slow temp increase is ecu limp mode value for water temp, does it stay at 80 once reaches it ? (if is constantly cycling 30 to 80 disconnected would assume earthing/circuit issue)
What happens to temp value if bridge the pins in sensor harness and do fans kick on (if fans not known working test circuit by grounding pin 9 via a back probe, 100ohm resistor good way bridging/testing.
Have you done a basic test of the coolant sensor resistance at room temp and in water being heated, if value flickering or way off FSM spec then that the fault. if Sensor tests A1 with no doubt then it earth wiring related by looks of things, if you got issues with more sensors etc it could all be related to same poor earthing perhaps .
Doing all tests above and reviewing date should eliminate more areas and give direction to what at fault or how test it further .
 

Danroyd

Member
Yip with it disconected it raises to 80 and stays on 80, with the two wires bridged the fans come on thats how im running it atm, yea i had same issue with the old temp sensor but ill test the new one anyway, so where does the ecu get the earth from? Am i best to try re-earth the ecu or just earth the sensor sttaight off the engine next to the sensor?
 

Mr B

Member
ECU has a couple of main grounds (pin 6 & 13) but if had main ground issues at ecu would expect lot of issues not just ECT sensor.
Worth checking grounds though just be sure .
You can't just re-earth the sensor as the earth return at pin 21 is the analogue data input for the ecu (ecu is reading voltage difference in what is basically a potential divider circuit and pin 21 is the tap off point for that measurement)
If ECT sensor test fine which likely will if you on second sensor yet same issue persists, this would make the ECT return wiring suspect to pin 21 & 29 (29 is shared earth with tps switch, does tps function properly/test to FSM spec ?, if disconnect tps any difference to ECT reading ?)
You could run through this wiring and connector points or do a quick test with resistors or ECT sensor wired direct to ecu pins 18 21 & 29 via back probes and see if ecu can actually read a good signal via displaying a realistic temp to resistance you implementing in the circuit, if so you know wiring bad and can dig into it further, if still doing much same it could be ecu circuit gone bad.
My hunch mainly as most common will be towards wiring but it wouldn't be first time I heard/seen ecu fault on ECT circuit .
 

Danroyd

Member
Ok ill check the wiring and all the earths of the ecu and see if thats the problem , thank you for your help il get back to you on what i find
 

Danroyd

Member
ECU has a couple of main grounds (pin 6 & 13) but if had main ground issues at ecu would expect lot of issues not just ECT sensor.
Worth checking grounds though just be sure .
You can't just re-earth the sensor as the earth return at pin 21 is the analogue data input for the ecu (ecu is reading voltage difference in what is basically a potential divider circuit and pin 21 is the tap off point for that measurement)
If ECT sensor test fine which likely will if you on second sensor yet same issue persists, this would make the ECT return wiring suspect to pin 21 & 29 (29 is shared earth with tps switch, does tps function properly/test to FSM spec ?, if disconnect tps any difference to ECT reading ?)
You could run through this wiring and connector points or do a quick test with resistors or ECT sensor wired direct to ecu pins 18 21 & 29 via back probes and see if ecu can actually read a good signal via displaying a realistic temp to resistance you implementing in the circuit, if so you know wiring bad and can dig into it further, if still doing much same it could be ecu circuit gone bad.
My hunch mainly as most common will be towards wiring but it wouldn't be first time I heard/seen ecu fault on ECT circuit .
Ok so i checked continuity on the earths and they seem good , and i went a brought a new oem nissan coolant sensor and its reading fine gets up to 80degrees but the computers not switching the fan on? It switches on at 80 right?
 

Mr B

Member
fan switching around 100 . either wait for fans cut in/out with car running (take while and get hotter than you think) or test circuitry by simulating 100deg+ with a resistor bridging sensor connector, variable resistor pots can be had for £2 and few of them along with some standard resistors make excellent diagnostic aid for car sensors and circuitry .
 

Danroyd

Member
200ohm to low 100ohm , variable resistor more usable tool but basic resistors will do job .
Ok i put a 180 in and it read 101 degrees on my computer and the fans were on , so i was just going to let it get to that on the computer but on my gauge it looked like it was overheating at 85 degrees so i dont know what to do could my gauge be playing up?20170811_180657.jpg
 

Mr B

Member
Gauge or sender for gauge out of tolerance I would assume, test sender or buy quality new one (lot of R's get wrong/cheap senders fitted so always keep that in mind), as it ntc sender you could calibrate it to some extent by putting a resistor inline on the single wire. they read about half for normal running and little over to almost 3/4 max for normal hot and fans just switching on then fall back to around half .
 

Mr B

Member
I think the fans come on at 94 or 95ish.
To be totally exact it has variations based on vehicle speed.
below 20mph fans switched on @ 100degC
above 20mph and to 80mph fans switched on @ 96degC
above 80mph fans switched on @ 105degC

* values above based on aircon "off"
* for stationary testing ecu fan ground switching circuit _ simulate 100degC+ 105degC+ input values
 

Danroyd

Member
Ok so i got a new oem temp gauge sensor , and got and installed another temp gauge but its still showing the car to be overheating when the fans kick in and ive got a consult cable an the fans are switching on at 96 degrees. Where is the gauge sitting normally at idle when the fans kick in on a gtir?
 

Mr B

Member
My needle is a little under 3/4 to bang on 3/4 at idle and fans just cutting in & half or fraction over in general running, most I seen about same with very minor difference.
Does yours display high even in general running rather than half . Could do with thermal probe or infra red heat gun get good idea real temps around sensor and cooling system, What voltage is the gauge sensor wire .
Don't rely totally on what you see in consult as ecu calculating that so some board issues could effect tolerance so you going need confirm the coolant temps ideally .
 
Top