[RexxyPup]

Just when I thought I could relax into just riding again, another minor niggle has appeared.

I pulled to a stop at traffic lights this morning and by the time the lights turned green, I had noticed that my temperature gauge had just hit the edge of the red.

Turned for home, and as long as I kept moving I could keep it in its normal place (1/4" above top of cold). As soon as I stopped it would start rising, and not stop at its normal hot with fan on position (mid way on gauge).

On nursing it home I check that I had enought coolant (It was full when I did the pre-ride checks) and it was at mid point on expansion tank.

With my helmet off it became clear that the fan wasn't running. At least that explained why the temp at the lights.

So I checked the 15A fuse (and the spare that is in the next slot ). The fuse was fine.

Right, so fan not running and fuse okay. My electrics knowledge is not the best (as you know) but if I remember correctly, a fan draws maximum current at start (zero revs) so I am thinking that if the fan were jammed that the fuse would have popped. A quick hand in the right fairing confirms the fan moves freely.

So does the bike have a separate sensor to switch the fan, or is it switched via the ECU? I'm starting to think this is a switch or power issue.

I guess it is main fairings off time. Could someone help on how to troubleshoot this please, as the workshop manual appears to be no help at all on this one!

Second question.... would I have done any lasting damage when the temperature hit the red? What sort of thing should I check for?

Thanks in advance..

A concerned Rexx

 

[rminchi]

2 days and no responses? I'll toss my 2p in. On my '99 there's a temperature sensor on the left side of the radiator that runs the fan. These tend to fail (on all types of vehicles) after several years as well as the wiring going to it. Since I don't have a wiring diagram handy, I don't know if the sensor works on a ground or hot lead. I'd check to see if the sensor is stuck open by checking with a ohm meter both cold and run until it gets hot enough where the fan would normally start. When cold, the meter should not show any reading at all (infinite) and when hot it should show nearly 0 ohms. Getting the fairing off is probably more work than anything.

As far as doing damage, unless you kept running it with the needle in the red for > 5-10 minutes I wouldn't worry about it. Not ideal, but usually not a problem.

Rob

 

[johnny]

Take a pair of jumper wires (one fused for power side) and direct connect fan to batt or acc. If fan runs then your problem may lie with the temp sensor (that tells the fan its time to come on). From there perform test of temp switch.

 

[redmosquito1]

after jumpering to test the fan you could install a switch and hard wire the fan to the battery through a 15A fuse. That way you save a little cash on the sensor and you have full control of when your fan is running and when its not. We did this on my bros gsxr and it works beautifully.

 

[oldndumb]

Forget comparing this system to an old automobile. The fan does not actuate from a temp sensor. The ecu determines when it is supposed to run.

I agree that you can hot wire the fan motor to test it, but I would hesitate to permanently wire it that way, until someone that has a schematic of the ecu can tell you if that is acceptable. It goes without saying that it must be isolated from the ecu before performing the hot wire test.

Before someone mentions the fan relay, that did not come about until 05.

 

[Calliway]

Even if the ecu triggers the fan the ecu still needs to know what temperature the bike is so there must be a coolant temp sensor on the bike. If that is faulty then the fan wouldnt come on but I would also assume that it would trigger an error code as well.


http://www.bikebandit.com/triumph-motorcycle-sprint-st-2002/o/m17603

Line number 11 on the fiche.

 

[oldndumb]

Sure it has a temp sensor. My concern was that the fan is not connected to the sensor. The fan is connected to the ecu, as is the temp sensor.

Out of caution, I would not be wiring the fan direct until someone with knowledge of the ecu circuitry and logic approved it.

I could be totally over cautious on this, but my concern is that a smoked ecu would really upset your day.

I would not be as concerned if we were talking about a +05, since the ecu is triggering the relay and the fan is not connected to the ecu.

 

[Calliway]

Yes youa re absolutley right with your last comment OnD. I to would be concerned about jumping any wires connected to the ecu.

My only point was that if the ecu doesnt know what temp the bike is then it wont switch the fan on and since the source of that data is the coolant temp sensor then it might be a good idea to start there.

It should be a two pronged unit and something that you can test with the bike at running temperature but heed OnD's warning. I might of come across like what he was saying was irrelavent which was not my intention.

 

[oldndumb]

No problem on this end. I understood your point.

Standard method of testing the temp sensor is take resistance readings.

Warm engine - 200-400 Ohms

Cold engine
20 C 2.35-2.65 K Ohms
10 C 3.6-4.0 K Ohms
0 C 5.6-6.25 K Ohms

 

[CLB]

Rexx,

You probably have a fan relay, I can't imagine sourcing fan current from the ECU.

It's probably the same part number as your headlight relay. Pull them both and check.
If they are the same... swap them for a check.

 

[oldndumb]

I didn't imagine it either. It is per the schematic.

Look at item 38 and you will see it connected direct to the ecu.

 

[redmosquito1]

OnD, how would shorting the fan switch burn up the ECU?

My thoughts are that if the fan is off then its basically an open circuit. So by cutting the wires in the clear, or disconnecting the harness, the ECU should see the same thing, fan not running. If your logic on burning the ECU up by shorting it to the battery is correct then the bike in its current state would toast the ECU. No Fan is no fan no matter how it got there. (as long as no path to ECU should be good.)

The problem is most likely in what ever turns the fan on, not the fan itself. Therefore the ECU isn't calling for the fan to come on and wiring it directly to the battery should have no affect on the ECU as it cant turn it on so it shouldn't be able to see it on either.

My electrical circuitry knowledge is not overly proficient as most of what I do is based on high voltage circuit breaker control wiring and transformer compensation schemes so I am most likely far off in what I'm thinking. But for the most part what I suggested is to simply bypass the ECU in its entirety giving full control to the operator. We do this a lot at work when control switches break or a relays decides to take a poop, but I love solid state stuff cause they need me to change them out, and I don't have to do the work only make sure its done safely and right. I love my job.

Either way thats my 2 cents.

Oh and last paragraph not trying to sound like the guys that say "I'm an engineer so I must be correct" I'm far from an engineer
and even farther from truly understanding auto ecu's and the such. (plus they always mess up easy stuff by over thinking it. no offense)

 

[Calliway]

Red,
There is no fan switch. The ECU is the fan switch. What OnD was saying is that the Sprint is like the old cars where you had a thermo fan switch that heated up, closed and turned on the fan.

In the Sprints case, you have a coolant temperature sensor that sends the current temperature of the bike to the ecu and the ecu then sends the signal to the fan to turn on.

Oh and ecu's can fry from static a static shock. Very little is needed to cook them.

 

[CLB]

Fair enough. Looks like it's a low side switch in the ECU.

http://www.t595.net/messageboard/thread/51548.aspx

Item 38?
I looked all through my HP130nr schematic and can't locate it.

I didn't imagine it either. It is per the schematic.

Look at item 38 and you will see it connected direct to the ecu.

 

[redmosquito1]

Cal I follow how the fan turns on. The temp sensor hits a certain level of resistance allowing more/less current to flow, the ecu sees that and does some cool solid state switching and allows full current flow to the fan. So the "fan switch" is either the ECU or temp sensor, depending on how you look at it. I see it as the temp sensor because depending on it position/resistance depends on whether the ecu turns the fan on or not.

My point was that it'd be difficult to toast the ECU from back fed current via the fan directly hooked to the battery if the fan is isolated from the ECU itself. So whether the fan was running or not the ECU shouldn't be able to see it. All that would be happening is there'd be potential across the 2 wires that power the fan. Perhaps if the ECU looks to see if the fan was running by analyzing the current used and voltage drop of the fan and it not seeing what it should then perhaps I could see the ECU having an issue there. But that should be easily solved by wiring in a resistor across the 2 wires that has an aproximate value of the fan. But then again thats a ton of pointless work when it could simply be a bad temp sensor.

 

[DEcosse]

Forget comparing this system to an old automobile. The fan does not actuate from a temp sensor. The ecu determines when it is supposed to run.....

I know you know this OnD but just to avoid confusion - it does of course actuate off the temp sensor but indirectly through the ECM after it processes the sensor data. I'm certain that's what you meant though.

The other part about the relay - actually the older (pre 02) S3's & Daytona's used to operate the fan through a relay - I would perceive the Sprints of that era might be similar? That generation (S3/Big D) had permanent ground on the fan and the power switched by a relay which was actuating on a ground from the ECM - Phew!

Now the pre-04 that Rexxy has - again I am assuming same as the S3/Daytona (since i have no schematic for the Sprint) - you would be correct in that would be permanent power on the fan (via ignition, MPR & the fuse) & ground supplied directly from the ECM (switched when trigger temp exceeded)
You can identify which wire has the power, then ground the other directly to frame - that will validate the fan operation (or not!)

Alternatively you could take a 100 ohm resistor, pull off the connector to the temp sender & connect the resistor across the connector pins, turn on ignition - you should see temp gauge in the red and the fan should run. That would test your entire fan circuit including ECM. (100 ohm should be equivelant to ~ 125C)
Or - if you got a 5kohm potentiometer you could connect that and start at 5kohms (which would be close to freezing! 2.5K is equivelant to ambient 20 deg) then turn resistance down toward zero - when you hit the crossover on the ECM, the fan should kick on.

Also - Rexxy - I know you have Tuneboy - have you checked to see what your fan settings are? I know the massive2 settings are to not come on till 99C - that might already be higher than your red bar. I don't think you have an actual temp readout on the ST (the RS does with the digital instrument) - however the ECM will still switch on the actual temp.
Try lowering the temp trigger on the Tuneboy & reload - it's under Thermofan under EDIT

edit - I recall now Rexxy that the std setting in the file was actually 103 and I lowered it to 99

 

[DEcosse]

.... But then again thats a ton of pointless work when it could simply be a bad temp sensor.

Can't be bad temp sensor Red - same sensor provides the info that is then processed to feed the temp gauge & the fan switch.
Temp gauge is reading in the red therefor not a suspect. (Even if it was 'off' and sending the gauge artificially high, that same artificial high would trip the fan switch)

As far as wiring a switch - a ground switched line to the fan 'low' side would be OK - even if the ECM was still connected - although not necessary. The ECM O/P's are short circuit protected anyway. Takes a lot to blow these things up!
I think you meant to just remove the ECM wire and have a permanent manual switch? I'd have to ask why that would be necessary - the ECM does its job just fine - when working - so the object should be to fix what's wrong.

Most likely candidates still remain
1) Power side - the Fuse (need to be sure you got the right one) and wiring - check for power at the fan itself)
2) The fan itself - ground the 'low' side of the fan (having verified you have power on the other leg) & verify operation.

Is NOT the sensor - by fact of the gauge reading;
Doubtful the ECM is faulty - although the set point may not be being reached by the value in the Tuneboy.

 

[RexxyPup]

A big thank you, and what next!

thanks for all the responses.

Calliway:
Unfortunately the BikeBandit link reverts to the top level of my model of ST, and I haven't been able to find a fiche with an item 11 that makes sense. Could you tell me which fiche you were on please?

I took the panels off last night, but didn't really get the time to look for the sensor. Any clues as to where it is?

Red:
Sorry, there is no way I would put a manual switch in.... I can't be trusted to monitor my temps with the frequency the bike demands and.... it would make more mess in the cockpit . I need to get the ECU back to its daytime job!

DEcosse:
Excellent summary and help for troubleshooting. Thank you.

At the moment I am running the standard Daytona Tune as I haven't yet taken it apart to remove the SAI. Riding Massive/SAI is not a combo to be recommended, particularly at low speed. I agree totally, the temp gauge is working thus the sensor must be working. The fan has worked flawlessly up to this point, so the ECU has been happily triggering it. I think this must concentrate my testing on the power to the fan, the fan itself and then the switched line. Looking at my fuse box with the two 30A fuses to the left, the fuses I checked are the two bottom right. The right most one was the spare, the next one in was the fan according to the Box lid. Both looked well, but I will continuity test them.

All others:
thank you for a. responding and bringing this thread back into the light ( particularly rminchi - that 2p may just save me £££s), and b. giving me the background on how the system is put together.

Based on the assumption that the temperature gauge is registering correctly, I think my next steps are as follows..

Switch on the ignition and put a multi-meter on the fan harness connectors to determine which is powered, and which is the switched earth.

If I can't find a 12v line, then I need to trace the power back to the fuse box, checking the fuses on the way.

If I have a 12v line, then I ground the other connection back to the frame.

If the fan doesn't run, then it's the fan. If it does run then it's a continuity check on the switched wire back to the ECU.

If that lot doesn’t answer the question, then I will have to start on the sensor side, but I think that my problem will be one of the following (in order of likelihood)

1. No power to fan
2. no switch signal to fan
3. fan dead

Thank you all for getting me to this stage, and please feel free to ‘help’ me back to the true path of electrical troubleshooting if I have strayed in the above!!

Rexx

 

[Calliway]

Under the Sprint fiches this is th elink.... sorry for the dud before.

http://www.bikebandit.com/triumph-m...04-engine-management-system/o/m17655sch569304

This is for Sprints though and you are using a Daytona engine are you not?

So number 9 on this fiche:

http://www.bikebandit.com/triumph-motorcycle-daytona-2004-engine-management-system/o/m17619sch567056

 

[oldndumb]

Rexxy'p, your trouble shooting plan sounds good and is in agreement with DEcosse's logic.

The schematic I have shows the fan power wire to be Brown-Pink. I hope that helps, but the Triumph documentation has been wrong/confusing before. Case in point, the schematic and the relay description/functions do not show a fan relay, yet the DTC listing includes a P1552 fault which is "Cooling fan relayshort/open circuit???

It also shows the two grounds from the ecu to be Black-Yellow. Which brings up a related question someone might answer for me. Why does it have two ground wire connections coming from two different ecu pins?

Redskeeter, my cautions were based on a concern that the ecu could be damaged. I left it up to an expert to say definitively whether it would or not. Since I do not know, I always err on the side of caution when it comes to the ecu. Either I am not brave enough, or I am cheap. It is apparent I have no understanding of the ecu logic.