gardener

New Member
Yes I know this is a common problem and there are several other posts on the subject, BUT I am pretty sure that my problem is a dodgy sensor and am looking for some additional info to confirm this before starting the process to import a new sensor from the UK (which is a real schlepp)

Background: the vehicle (1998 4.6 HSE) has been standing for some time (2-3 weeks), has been started occasionally but not driven. Suddenly, just after Christmas, the gearbox overheat message appears. The gearbox is clearly not overheating as it is stone cold, not having run. I have tried charging the battery, adding a second battery in parallel (in case of fading battery causing the error). I have driven the vehicle 10-15km to see if the message disappeared when the vehicle warmed and the battery perked up. I have also checked the wiring to the sensor and there is no obvious break/short. When C164 is disconnected pin 1 in the feed side gives a short to earth (which it should) and pin 2 does not register on the Ohmmeter, which would indicate to me that there is no wiring short to earth in the feed to the connector. The sensor itself is giving a resistance of 0.9 Ohms when cold, and this does not change when placed in a kettle which is boiled.

The circuit diagram in the electrical troubleshooting guide would suggest that the sensor thermo-switch is open when cold and closed when hot/overheated, which would be consistent with the very low resistance on the approximating a closed circuit and therefore triggering the alarm message. However, to my mind this would also mean that if the sensor was removed, thus leaving an open circuit, this should approximate to the open switch in normal conditions. But the overheat message still appears with the sensor removed!

So, my questions are: does the circuit need some (large)resistance from the sensor to indicate normal temperature? What is the normal resistance from the sensor when cold? And also when hot? And can I confirm that there is nothing else wrong by placing a suitable resistor in circuit to ensure the message goes away (before going to the trouble of importing a replacement)

Any suggestions welcome
 
The sensor is a switch, it closes at 95C +/- 3C. On the diesel it operates a relay to drive a cooling fan, I can't find any mention of this on the V8 but if it does drive a relay your fault may also be either the relay or the fuse box. Spend some time looking at the circuits in RAVE.
 
Believe me, I have spent an inordinate amount of time in the last week or so poring over RAVE ETM and trawling the various fora for info.!

If the sensor is a straight on/off switch, if it is taken out of circuit should this not then simulate an "off" position and the message disappear? I have taken the switch out and the message remains, which leads me to think that it is a heat sensitive resistive device which is triggering the message when it reaches a particular resistance, corresponding with the out of range temp reading. On an old thread I found there was the suggestion (unconfirmed ) that the resistance when cold was 1.45Kohm but I cannot find any mention of a resitance when hot.

If a straight on/off switch, which is off when cold, then doesn't a virtually zero resistance or closed circuit reading on the bench when cold indicate a duff sensor?

I am pretty sure there is no cooling fan on the gearbox oil cooler on the petrol models; as far as I can see the diesel cooler is mounted lower in the engine compartment and therefore does not have the same degree of ram air cooling as the petrol - hence the fan.

Another problem I have is that the boiling point of water here is only 95degC so my test in the kettle might not mean anything!
 
You are right there is no cooling fan on the V8, I did not say there was but there may be a relay as there is on the diesel, if there is that could be where the fault lies. It is definitely a switch as far as I can see, can't see why it would be a thermistor as it only triggers a gearbox overheat warning and limp mode.
 
Just had another look at RAVE, section E2 warnings and indicators, page 134, it clearly shows the auto box temperature sensor as a switch which feeds direct to the BECM, there is a similar switch for the transfer box.
 
I agree that RAVE shows the sensor as a switch, which is open when normal and closed when fault condition occours. My concern with this is that with no sensor in circuit ie open circuit ie same as open switch, why does the message appear? I can only think that there is more to the little plastic sensor than a simple heat sensitive switch, that there is some reading it provides to the BECM to confirm it is in circuit.

I have re-checked and there is no relay in the main box that relates to the gearbox overheat circuit - my understanding is that the relay in this circuit is for the diesel only, to switch the fan.

Interestingly enough, I had an opportunity yesterday to get my hands on another P38 which also gives a closed circuit resistance across the sensor leads.

I am planning next to check the wiring continuity between the sensor connector C164 and the BECM connection C112
 
It sounds like you have a problem within the BECM, the sensor is definitely a simple switch. One lead to the sensor has ground on it, the other should have either +5 or +12 volts on it (not sure which voltage) from a pull up resistor in the BECM.
 
I have finally got back to this problem after several weeks. And cracked it! I managed to trace the fault back to a break in the wire between the sensor connector (C164) and the plug into the BECM (C112). Some small furry animal escaped from the cave and decided to nibble on the wiring harness, in the corner just below the radiator header tank. Spliced a fresh wire across the break and the overheat message was cleared!

One thing that this does show is that the sensor is not a simple on/off device; there is clearly a residual measurement taken when cold which the BECM software uses to recognise that there is a sensor in circuit. A simple on/off device would not return an error when taken out of circuit (as can be demonstrated by disconnecting C164, the plug to the sensor)

What would be useful for diagnosing similar problems in future is knowing the resistance or other pertinent measurable data about the sensor when a) cold and b) hot, to determine if indeed it is the sensor or some other problem. Anybody got this info?
 
I have finally got back to this problem after several weeks. And cracked it! I managed to trace the fault back to a break in the wire between the sensor connector (C164) and the plug into the BECM (C112). Some small furry animal escaped from the cave and decided to nibble on the wiring harness, in the corner just below the radiator header tank. Spliced a fresh wire across the break and the overheat message was cleared!

One thing that this does show is that the sensor is not a simple on/off device; there is clearly a residual measurement taken when cold which the BECM software uses to recognise that there is a sensor in circuit. A simple on/off device would not return an error when taken out of circuit (as can be demonstrated by disconnecting C164, the plug to the sensor)

What would be useful for diagnosing similar problems in future is knowing the resistance or other pertinent measurable data about the sensor when a) cold and b) hot, to determine if indeed it is the sensor or some other problem. Anybody got this info?

Depends on if sensor is normal make or break. If it's normal make, overheat break BECM will think no circuit is overheat break and throw the fault up.
 
Depends how the sensor is plumbed in. If it was set that off threw up a fault and on turned it off again, it'd throw a wobbler if the sensor failed, preventing the risk of overheating the gearbox if it was wired the other way round.
 
Granted, but (stated elsewhere in several posts) overheat state closes the circuit to earth, activating error message. So, closed activates error, open activates error, range of values between open and closed, corresponding to gearbox oil temperature in normal range, gives no error. Question is, what is this range of values? (in order to test the sensor)
 
Granted, but (stated elsewhere in several posts) overheat state closes the circuit to earth, activating error message. So, closed activates error, open activates error, range of values between open and closed, corresponding to gearbox oil temperature in normal range, gives no error. Question is, what is this range of values? (in order to test the sensor)

Overheat sensor is a make or break switch according to RAVE. It is not indicated as having a variable resistance. And as you say would seem to be open circuit normal, made circuit hot. So can't explain how repairing wire solved your problem.
 
Now that I have a properly functioning circuit again I can confirm that the sensor works like a switch which is closed = normal, open = error (not the other way round as shown by RAVE! and stated elsewhere). This would make sense so that when the sensor fails ie is no longer closed, an error condition is indicated prompting investigation.

So a simple way of checking for a duff sensor is checking that there is little or no resistance across it (my original reading was 0.9ohm). Or connecting pin 2 of C164 to earth. If no error condition is indicated when this is bridged then the circuit is ok and the sensor faulty (assuming an error condition indicated on a cold gearbox!!)
 
Now that I have a properly functioning circuit again I can confirm that the sensor works like a switch which is closed = normal, open = error (not the other way round as shown by RAVE! and stated elsewhere). This would make sense so that when the sensor fails ie is no longer closed, an error condition is indicated prompting investigation.

So a simple way of checking for a duff sensor is checking that there is little or no resistance across it (my original reading was 0.9ohm). Or connecting pin 2 of C164 to earth. If no error condition is indicated when this is bridged then the circuit is ok and the sensor faulty (assuming an error condition indicated on a cold gearbox!!)

Just as i postulated earlier then, which Rave contradicted. :):)
 

Similar threads