Hi guys,

Recently my bike has been flashing the overheat light after about 10 or 12 minutes of freeway driving. The odd thing is sometimes it will go off, and sometimes it will flash back on. Sometimes it will be on and then I go up a hill and halfway through it goes off. Sometimes when I pull in the clutch quickly it goes off as well and if I'm in first gear and linger in the sweet spot where the transmission is about to take over and accelerate the bike, the light will also go off. I've heard that on a downhill run sometimes the light will go off because water goes through the thermostat if it's not functioning correctly, but this isn't the case.

I just tested my temperature sensor with a multi-meter in hot water like Crazy Frog did in "oh dear and now temperature issues". This is the chart in fahrenheit if anyone is interested.


I checked the temperature conversions and it seems just about the same, so this means my temp sensor is working. This makes sense because I assume it was working before as the temp light never came on. What doesn't make sense is that the temp gauge which is connected directly to the water pump drain plug is working just fine. It gets up to normal and stays there as it should. I haven't tested the gauge with hot water yet so we can put that on the list of things to check, but it doesn't seem like it would fail overnight. I have the right amount of coolant in the bike, I tested the thermostat in hot water and it opens at 85 degrees celsius where it should. The overflow tank rises when the bike gets hot. The oil is at the right level and I opened the oil drain plug to let some out and didn't notice a sheen in it (I assume since oil floats on water, the water would be at the bottom of the oil sump.) I've gone through all the connections on my bike and sprayed with Deoxit D series and gold series, and have scraped oxidation off and sprayed all the grounds I know about. I have burped the radiator cap many times to try and eliminate air bubbles. The radiator fins are in pretty good shape. The fan is working, but it usually spins slowly and kind of in spurts until after about 8 to 10 minutes then comes on full blast and creates quite a cold draft (and yes its blowing the right way). I haven't tested the RPMs on it and don't know how, but I assume it's working.

So given all this information, and assuming the temperature gauge works (for now), I wondered what you guys think about this: The temperature gauge lead is connected directly to the water pump which is before the coolant goes into the engine to cool it, and says the temperature is normal, the temperature sensor is on the pipe coming directly out of the engine, which works as it should and is saying the temperature is too high. I have deduced a list of possibilities. Maybe the water pump is partially blocked and doesn't get water through to the engine in the right amount and so as a result the water on the temp sensor is hotter. Maybe the water pump isn't working at all, but I would think this isn't the case since water seems to be circulating. Maybe it's a plain old error in the temperature sensor relay. Maybe the coolant temperature light is not working (although that doesn't make sense because it comes on after a certain amount of time pretty consistently). If something was clogged in the engine part of the cooling system I think this could cause it but I don't know how I would snake something in there to check.

I have probably given you everything I know and exhausted any readers who have made it this far so I will stop. I am just hoping this causes a lightbulb for someone and could offer some tips about what to do next, and if this involves anything like taking apart the water pump, being so kind as to include any gaskets, washers, or other parts you're supposed to replace each time. I will be happy to take pictures of anything anyone asks me too (on the bike that is ) if anyone thinks that will help.

Are you aware that the sensor behind the radiator is for the temp monitioing by the ECU and the fan switch.
There are 2 seperate sensors and work between the pin and earth not between the pins. The graph I am seeing is just about the correct curve for the ECU but the top end resistance is too high. Have you just tested the ECU sensor or both in series by testing across the two pins.
Also when I started reading the post I immediately thought this is a wiring problem but on reading on futher wasn't so sure. Look for chafed wires bad connections at the relay plug and the multipin connector on the instruments. When you check try to pull the connector off the individual wires they probably won't come but a damaged wire will allow the plastic insulation to stretch a bit.

yes i agree kf

the terminals for he temp sensor ..(which is connected to the water stub behind the radiator ) connect up around all the other connections under the tank

freelander ...the best place to check the sensor is the pin 10 of he injector control connection under the seat .....

i had found previously that the temp sensor was intermittent contact between itself and the frame(stub) by simply by removing it ....plugging the stub where the sensor screws in to stop water l;eaking out )...and cleaning it up produced positive results ..

the connection under the cap 2 x off (hard to get off ) needs to be cleaned with deoxit also. then clean up the brass threads and the surface of the thread bind coupling ....and replace the washer that is the seal between the two ....(mine was corroded)

after wards i got good results ...but never had quite the same issues as you ...but the fan wouldnt come on all the time anyhow ....

i would be looking for a chaffed wire ar kf suggested as as the resistance goes low is when all the issues arise ...might be good to check the temp relay first ...if there was corrosion in there it could give internmittent and false readings .....resistances in parallel ..that sort of thing ....

one part of the sensor goes to the controler under the seat ...and the other half goes to the fan relay thingy



good luck

When you guys say ECU do you mean the fuel injector control unit right? I have not test the pins on it because I'm not sure what you do. I've read about you guys "jumping" the connection, does this mean testing continuity with a multi-meter or does it mean using an alligator clip or probe connected to the battery?

I did follow almost all of the wires looking for melted insulation, burns and chafing, and have found none.

Charlie, I have already used deoxit on the cap and the problem persisted. And as for the intermittent contact issue, I'm not really very knowledgeable with electronics but it seems to me the ecu must be looking for low resistance which gives it the cue to turn on the fan and then turn on the overheat light, so if there was intermittent contact wouldn't that be infinite resistance when the circuit was open? That wouldn't just trick the control unit into thinking the engine was cold? I can hear my fan running when I ride so I know that whatever is telling my fan to go is working.

I cleaned the connections at the temperature relay and on the bulb monitoring unit. It didn't look like moisture or corrosion had gotten into the little electrical box of goodies.

Also my ohmmeter is a little hard to read at 35-50 mark so thats probably why the resistance is a little bit too high.

The sensor for the ECU (fuel injection computer) is one of the 2 sensors and the fan the other. the ECU does not control the fan the temperature control relay does that and the light on the panel the ECU only reads the temp to control the mixture. A chafed wire could be shorting to earth which the relay sees as high temp and so brings the light and the fan on. The sensor on the pump only sends to the gauge which is independant of the temp relay. If the gauge says the temp is OK then I would be looking at the temp relay and wiring assoc with it.

yes free the ecu under the seat ( more correctly the injection control unit )

it is a convienient place to check as you are measuring everything in the path to the end ....just get the red probe of your multi meter stuck in there ...the blades of the connection will give to allow it and then spring back ...lthe black leg should go to a reliable earth

......incidentally if this was the issue on this particular leg of the temp sensor .....the bike would run leaner than normal....

what im saying is, that this could be a guide .....all you have to do is get it up to temperature ...than the resistances should be under 2000 ohms ...(closer to 300 - 100 actually )

from what kf is saying and i also agree, the problem exists in the other leg ,.....the one that goes to the temperature relay ..

use the same teqnique of probing the right connection there and a good earth

the pinouts are in the fuel injection problem solving links in the portal page ...from memory

i think we have seen a couple off issues wih this unit before ......but like always ...the truth is in the testing first .... i try to prove the fault before i progress .

just a thought ......has someone wired in a relay activation switch before ......which has now been removed ...and the cable end might be shorting to ground intermitently ?

I don't think the fan should come on at all until triggered by the temp
sensor, that makes me think it's an electrical wiring thingo as well. I don't normally make a lot of sense but have had to spend some quality time on my knees trying to fix my cooling problems. (and I don't mean praying).
RT

G'day All,recently had a similar problem with the coolant warning light staying on all the time.The problem turned out to be a faulty heat sensor relay which was probably caused by a seized fan.Charlie's advice was followed and all readings were spot on.Managed to borrow another heat sensor relay,plugged it in and that solved the issue.Also replaced the fan.Charlies advice is invaluable in matters all electric.Cheers Tom Grabau

.