Cooling fan not turning on

[kevinu]

I know this has been discussed previously, but the thread that I have carefully followed ("cooling fan activation time interval") had not been posted in for more than 180 days, so here goes:

I have a 1994 K75.  I live in the desert and it is about to get really hot outside, and I am concerned about the fan and the temp light.  On previous moderately warm rides, I have noticed that the fan never turned on, and I never saw the temperature light come on.  So, I followed a thread or two and did this:

1.  I confirmed that the fan spins freely.  And that the horn works - I believe they are both on the same fuse/circuit

2. I checked the temp sensor by testing the ohms on position 10 of the FI computer.  Results: 2.5 K ohms at cold (~85 degrees F outside), 1.2 K ohms 1.5 hours after engine fully warmed up, and 0.183 ohms immediately after running the engine to hot.  Conclusion: this data seems to indicate the temp sensor is sending the correct signal to the FI computer

3.  I checked the temperature relay switch:
         jumping A2 to 15 turned the fan on.  good.
         jumping A3 to ground turned the temp light on. good.
         jumping E to ground (supposedly simulates an overheat scenario, should turn the fan on). nothing. bad.

4.  So, I purchased a used temp relay switch and tried it, but with exactly the same results.  Fan and light work, but no fan when I jump pin E to ground.

Should I try a new temp relay switch, or is the "pin E to ground" not the correct test for what is wrong?

[herseyb]

The only time my K75 fan came on was in stop and go traffic on the baking hot expressway in 90+ weather outside of st louis.   I usually get heat stroke myself as a warning before the bike overheats.  No point to this anecdote except to say that you might consider fitting a manual override switch while you've got it opened up. 

[Scott_]

+1 on a manual fan switch...

[herseyb]

Just double checking - you're grounding E while the engine is running, ja?

[rbm]

You can EITHER bridge E to ground OR jump A2 to 15. Both end up performing the same function - turning on the fan independently of the actual coolant temperature.  Since you've found that jumping A2 to 15 turns on the fan, then go with that solution.  Later, when the engine is running, try out grounding E and see if that also works for you.

[kevinu]

None of my tests have been done while the engine was running.  Is this necessary?

[herseyb]

yes for the E ground the engine must be running is my understanding.

[Laitch]

. . . I checked the temp sensor by testing the ohms on position 10 of the FI computer.  Results: 2.5 K ohms at cold (~85 degrees F outside), 1.2 K ohms 1.5 hours after engine fully warmed up, and 0.183 ohms immediately after running the engine to hot.  Conclusion: this data seems to indicate the temp sensor is sending the correct signal to the FI computer

Your sensor check determined the temp sender is sending the correct signal to the FI computer so the fuel mixture is ok. What seems to need diagnosis is whether the sensor is also sending a signal to the temperature relay to activate the fan at an appropriate temperature. If the fan and overtemp light are activated by grounding pin E but the fan won't turn on when idling in hot weather, it's likely a new coolant temperature sensor is needed. If you've idled the moto in traffic at 85º and the fan hasn't cycled briefly at least once, there might be a problem.

In 85ºF weather, the fan should probably cycle on in ~15 minutes of idling then cycle off. The on-off cycle will continue until road speed begins. That performance is also based on a radiator filled with appropriate coolant, a radiator cap in good condition, a functioning thermostat and a reservoir at the appropriate level.

The fan system works well during hot weather in my experience but it's a faith-based experience. Those with less faith install a fan switch as has been described elsewhere in the thread. I like the thrill of anticipation. 

[The Mighty Gryphon]

Is the problem in getting 12v to the fan, or is it that the fan isn't running when it gets power?

The Bosch fans on these motos have a lot of issues when they get a lot of miles and years on them.  Normally, the motor seizes and the blades don't turn, then there is the situation where the blades slide forward on the shaft and grind into the radiator core.  In your case you may have melted brush holders that don't let the brushes press onto the commutator.  Eventually the brush wears down to where it no longer makes contact, and the motor won't run.

These fans were manufactured using plastic that didn't handle high temperatures well.  Brush holders and fan blades are a fairly common source of problems.

[rbm]

Is the problem in getting 12v to the fan, or is it that the fan isn't running when it gets power?

From the first post, he says the fan is free to spin and that it comes on when he jumps A2 to 15 on the temp module.  So that indicates the fan is functional and the power and wiring to the temp module and to the fan is also functional. 

He needs to start the engine to supply power to the temp module electronics to test pin "E".

[kevinu]

The fan motor runs fine - I have a temporary hot wire through a switch to turn it off and on. 

But I did not have the motor running when I grounded pin E to simulate an overheat event - several posts seem to  indicate this is necessary.  Is it?

[Laitch]

- several posts seem to  indicate this is necessary.  Is it?

What would it take to convince you?

[rbm]

Yes, the engine must be running.  Why?

The temp module gets power into pin 9 from fuse 6, which is fed from the FI relay.  The FI relay closes and supplies power to fuse 6 only when (1) the ignition is on, (2) the kill switch is closed and (3) the starter button is pressed.  ECU pin #7 sends ground to the coil of the FI relay as soon as the starter switch is depressed. The ground will stay on after the starter switch is released as long as the ECU receives a signal from the Hall sensors (i.e. engine is running). So, until the starter button is pressed, no power is present on fuse 6 and therefore no power is available to the temp module to register the coolant temperature.

Start the bike and test again.

[kevinu]

So, the reason I did not have the engine running the first time I did the temp relay test was that I had the gas tank completely removed.  I did not realize then that you could access the relay with it in place.  Now I know you can just lift the gas tank up enough to get to the relay. 

This morning, I got my 16 year old wiz kid son to help me (he's good with arduinos and things electrical) and we accessed the relay, started the motor, jumped pin E to ground, and saw 12 volts on the fan input and the temp light came on.

So, maybe I never had a problem in the first place - maybe I never had it hot enough to trigger the fan.  My question now is: Do the temp light and the fan both come on at the same time?  I thought I had read somewhere that the temp light came on later (lower ohms from the temp sensor), but we saw the temp light on when the fan was getting 12 volts with the pin E jumped to ground.

Thanks to all for the help so far.  I hate not understanding problems (or lack of).

[Laitch]

  My question now is: Do the temp light and the fan both come on at the same time?

The fan is activated to prevent the engine from overheating and keep it at an optimal running temperature.. The overheat lamp is lit when the engine overheats. The fan starts first. As was indicated in post #7, the fan should be triggered if the engine is idling long enough. If the fan doesn't activate but the overheat lamp is lit then a new coolant temperature sensor is likely needed, based on your description of your test results.

See if idling the engine for 15 minutes or so will trigger the fan, or the overheat lamp.

[rbm]

Remember that you are simulating a very very hot coolant temperature condition by dragging pin "E" to ground potential.  The temp sensor in the stand pipe has a logarithmic response curve.  The temp module is programmed to activate the fan at a preset temperature, and to activate the overheat indicator at a different warmer temperature.  So, as Laitch points out above, the fan and indicator activate at different times under normal conditions.

[Billk]

Wish there would have been some visuals with this discussion like pictures or diagrams

[Laitch]

Wish there would have been some visuals with this discussion like pictures or diagrams

If you had done a search using the guidelines on this site with something like fan switch diagram site:motobrick.com as your Google search term instead of wishing, you'd have found this discussion in the third hit. As it turns out, your wish came true anyway.

You young whippersnappers should get with the program. 

[rbm]

Wish there would have been some visuals with this discussion like pictures or diagrams

This is a schematic of the essential parts for the temperature relay:
http://www.motobrick.com/index.php?topic=8295.0

The temp sensor is a variable resistance depending on the refrigerant temperature.  When the refrigerant is really cold, the resistance is high and the fan is off.  When the refrigerant  gets very hot, the resistance is near zero, just like a short to ground. So pin E goes to ground and turns on pin A2, where power for the fan is supplied from Pin 15 (theres an error on the diagram; it's marked 45 on the diagram but 15 on the actual relay).  Power for the whole module goes to pin 9 and ground for the module goes to pin 31. if the the sensor tells the temp module that the temperature is really hot, the module turns on the temp light on through Pin A3.

[k]

In my case, the fan was broken, and I had to order one from euromotoelectrics.com

[BrickDad]

Www.tills.de also has SPAL fans to replace bad OEM fans on K bikes.

[stokester]

Www.tills.de also has SPAL fans to replace bad OEM fans on K bikes.

Yes, mounted in an aluminum bracket with the proper connector, new bushings and mounting bolts.

I replaced the fan on my '93 75k K75S just because I had read about the failures of the Bosch units and did not want to be stranded somewhere.  I purchased one from tills.de for about $120 and it was delivered in about a week.  A quick spin of the original spin revealed a squeaking so it was not far from failing.