Okay guys with the multimeter set to 20k Ohms I get a reading of 2.15~ and that's with the bike cold.....well 80 degree ambient air temp, but the bike is in the garage sitting in the shade water at room temperature. Does that sound correct?

I do not have the specs,,,,someone else here will.
As long as to are right there....you could put a hair dryer to the water outlet where the switch is and see the resistance start to climb as you warm it????

At 20K setting you get 2.15.....that is 2,150 ~. If you were to put the meter on 2K setting.....the decimal on the screen would move 1 place to the left and the reading should still be identical @ 2.15 or 2,150~.
For a moment if we assume from the previous test..that High resistance turns on the fan relay......then if you warm the housing with a heat gun.......that reading should start to head Upward........20.2......2.3.......2.4.
Right now we are just guessing. If you can see one of the six "flats" on the sensor base......there is probably a small 2 or 3 digit number stamped in the side of it......we may get more clues from that......
I haven't been able to find a reliable spec, yet.

Just a side note for future ref.
The temp sensor numbers change at build date 7/93
Up to 7/93 = 13 62 2 242 184
from 7/93 on = 12 61 1 341 602
still no spec yet

Done working on the bike until tomorrow, but I had a spare connector from my old wiring harness so taking the readings off the water temp sensor will be easy. I have a heat gun I wonder if I can get the housing warm enough to trigger the fan?

You would have to get the coolant inside up almost to boiling to get the fan to start (221 deg F?)
I was just hoping you could warm it enough to see if the resistance indeed goes up with heat or perhaps down?
As discovered earlier.....most coolant temp sensors get less resistance the hotter they get.
The fact that your fan started with the plug unconnected would SEEM to indicate that this system works the opposite???
Another thing to consider is that the ECU may start the fan as a fail safe "limp home" method should it see the circuit be completely open as when you pulled the plug.
The fact that it starts with the sensor disconnected does not absolutely indicate that the ecu normally starts the fan at high resistance when connected up.
As soon as you can warm the sensor......we will know more. Still searching for the spec......Regards, ibj...

Yes that is correct I would expect that range and as the bike heats up the resistance drops so if the fan is coming on when you unplug the sensor then the fault is most likely in the Motronic unit. Does the bike start with the sensor unpluged?
I know the 8v will not start but I am not real sure about the 16v Motronic2.1 but I belive it is the same.
The motronic units can be interogated to find out what is wrong. The K11OG forum should have the testing procedure and required led to read it. You can buy a tester for the ABS and Motronic 2.1 from Motobins in UK with all the instructions.
I have one but the postage would be more than they are worth, I paid 28 pounds for mine and its worth its weight in gold.
The instructions on how to make are around the net try K11OG or Motobrick
one thought check the plug itself for damage it is possible that when you removed the plug the wires touched, unlikely but possible. I have been working on machinery of all sorts for most of my life and the plug itself is often overlooked and sometimes by me. You test each way at the plug but not the plug connection.

Thanks Rick I think that's going to be a worthwhile investment since I plan on putting some serious miles on this bike.

So the bike will start with the sensor unplugged. The fan comes on as soon as the ignition is in the on position, and then the load shed relay shuts the fan off while the starter is depressed. The fan comes back on when the bike is running.

The only 3 digit number I think I was able to see was a 280, but I am going to try to clean things up in there a little bit and get a closer look to be sure.

I'm on daddy day care duty right now, but will be heading out with the heat gun shortly to check the resistance when warming up.

I did read through the starting troubleshooting guide on this forum, and there is a chart with specs regarding the temp sensor, but I am not sure if those numbers apply to my bike. By roughly taking a look at the reading I got and trying to intersect the line on the graph, the temperature numbers actually seemed to be reasonable....somewhere around 25-30C?

* Another quick update....I did check resistance on the temp sensor again this morning as it was around 52F last night. I read out 2660 this morning so that also seems to point to lower resistance when the bike heats up. Now I'm puzzled about the unplugged cable and fan coming on as well as the bike allowing a start with the sensor unplugged?

As I mentioned earlier.......I'll guess that when the ecu see's "OPEN" on that reference line....it is fail safed to start the fan??????

Heat gun test complete and as expected I was able to lower the resistance from 2600~ Ohms to 1580~ Ohms within about one minute of applying direct heat from the gun to the sensor housing. I didn't feel terribly comfortable blasting 1000W of heat gun on the components for too long so decided that was plenty to give me an indication of what is supposed to happen.

Still working on getting that 2 to 3 digit number.....wish I had an inspection mirror as I think it is on the flat that is at the rear rather than the front of the bike.

Shorting the connector out also causes the fan circuit to become active.....ibj I think you are probably correct about the fail safe aspect of the Motronic seeing an open circuit.

You have demonstrated that the sensor is working.......I doubt if you need to go further there, except for DeOxing it ang getting the cable plugged in.
As a next step after its connected.....I would locate those 2 wires at the cpu plug and see if they still read at that end.
We have proved (I think) that the ecu has the ability to switch on the fan relay.
I think that once we have verifiable battery voltage (new batt?) we are close to starting it up for a test.
As I'm not up to date on the piping arrangement, I can't say exactly, but on my bike, I think I would remove the fill cap and squeeze the hoses to work any air out, then run a test with that cap off.
Perhaps you can borrow an actual mercury thermometer to insert in the coolant there?
Since the fan is supposed to start at 221 deg F.......you may not be able to get it to run with the cap off, but it's close, because with a 50/50 mix the boil point at sea level might be above 221??? Research shows that an exact 50/50 mix would result in a 226.8 boil.
Not much wiggle room between 221 & 227 so maybe the cap should be On for the test.
Another possibility for your mystery failure might be that the cap fails to hold pressure. Inspect it carefully and clean out any gunk......then when you do install it screw it on tight.
You may be able to find someone near you that has a radiator pressure tester that has an adapter to fit it?

ibj I think the cap and the coolant overflow hose will be my next two points of scrutiny....combined with a new battery of course.....as Paul mentioned earlier, a crack in my coolant overflow hose or as you say a radiator cap that isn't sealing correctly would result in coolant filling the overflow tank, but not returning to the radiator as the bike cools?

Sprayed CRC QD Elecronic Cleaner on the connections and cleaned the contacts as well. Everything is plugged back in and I feel a little more competent regarding my fan circuit.

@ibjman wrote:At 20K setting you get 2.15.....that is 2,150 ~. If you were to put the meter on 2K setting.....the decimal on the screen would move 1 place to the left and the reading should still be identical @ 2.15 or 2,150~.
For a moment if we assume from the previous test..that High resistance turns on the fan relay......then if you warm the housing with a heat gun.......that reading should start to head Upward........20.2......2.3.......2.4.
Right now we are just guessing. If you can see one of the six "flats" on the sensor base......there is probably a small 2 or 3 digit number stamped in the side of it......we may get more clues from that......
I haven't been able to find a reliable spec, yet.

wrong ... if the reading was 2.somthing k on the 20 k scale and then changed to 2 k scale the meter would read open ....or infinity till it gets under the full scale range of the scale of the setting ...

many have made this mistake and declared things faulty

till they changed to the correct range after some advice .


auto ranging meters show differently and each has their own merits and deficiencies , pending what you are looking for .

just a heads up in the use of a digital meter with a manual scale adjustment

The 2k scale did show infinity as you mentioned charlie which is why I ended up using the 20k scale to get a reading that was usable. Thanks for the clarification!

good work cafe ....

yes, just to clarify

I stand corrected Charlie. Thank you for that.

OOOOPS......I just went back & checked again....
I was switching between 20K and 200K to get that result.
But I posted between 20k and 2K didn't I????
My apologies to everyone!
Regards, Ibj...

Bit late but the temperature sensor response curve is illustrated in this diagram.



Resistance is inversely proportional to temperature and non-linear (logarithmic).

My thoughts on the fan on if it's unpluged is that the system goes into limp home mode and as a result it starts the fan just in case the temp is what is causing the problem
Well you have shown that the switching circuit for the fan is OK. Usually when the coolant comes out in the quantities you describe it is overheating that causes it and not trapped air which makes it overflow and gurgle and bubble etc. unless an airlock has caused it to boil by suddenly getting coolant to the hotspot.
I would be inclined to go for a short ride to get it up to working temp then let it idle while watching the temp rise with a infrared thermometer on the left tank of the radiator and see if the fan comes on at the right time.
I have had that type of boiling experience with cars a few times and nothing was ever found to be wrong and was put down to air locks.

Sounds good.....I actually just got done letting it idle after a quick ride up and down the street. I idled the bike for about 10 minutes after my ride, but without the aid of an infrared thermometer and knowing what parameters I would be testing with it, I felt that it was probably not a good idea to push the bike too much. Coolant definitely entered the expansion tank and I'm currently watching the level drop again as it draws the coolant back into the radiator while the engine cools down.

Thermometer on order and then more troubleshooting! It felt great to even put that half a mile on it as always.

Time for one of these and thank you guys!

Gents I got the boil over again in the driveway without the fan kicking on. I could clearly hear the coolant boiling for about 30 seconds or so while it spewed out of the tank, but no sign of life at all from the fan. I'm thinking it might be time to think about changing the sensor? I still don't have the temp gun to confirm the overheat, but based on observation it seems as though he bike is not getting the signal to kick on the fan.

I'm always glad to duplicate a problem, because then the logic says it should be fixable. I'm thinking sensor right now, but I think I need more reference points before I decide to conclude that this is the problem.

Hi there, you were on the driveway, so, the boil over occurred while running, not at a stop or traffic jam. Under that condition you should be far from fan's cut in's temperature and the boil over may be happening due to a lack of pressure. It could be a radiator's cap failure. I would check the cap's gaskets.
Best regards
JAP

I agree, The pressure test on the cap is needed + I still think we need to test the sensor readings at the ecm plug to verify that both the connections all the way to the sensor are sound.
Another thing that could be tried although I'm not sure if it would prove/disprove anything is to do it again tomorrow knowing now about how long it takes from cold to boil. Run a few minutes short of that and then while still running, artificially power up the fan to see if it will continue to run without boiling.
(This too shall pass.....soon).
Can you take the cap down to the local radiator repair shop or auto parts store to test it????

after filling the radiator leave the cap off
let the bike run for a little while to settle out the cavitation in the pump and get some flow happening ....else you may just create a bubble down there .
after you have let it idle for a couple of minutes ...tip the bike to the left with the filler cap off
and refill the fluids again ...right up to the filler cap
the top of the radiator doesnt let all the air escape fully because of how the filler tube enters .....
i wouldnt be surprised that you could get another half pint in there using this proceedure ....and eliminate the air bubble ....air expands really well when heated
just a thought ...many of us do this process ,.but maybe you are unaware of it ?

Thanks I was unaware of the tipping the bike left trick to help fill the radiator and eliminate air. I will do that as well as having my radiator cap pressure tested. That's also a very good point.

What I didn't do yesterday when I had the connector shorted out is attempt to start the bike. The bike did start with the fan on and the circuit open, but I think we have established that as a fail safe. With the connector shorted however, the bike should not start if I understand the circuit correctly?

* ijb I am running an errand this morning so the radiator cap will be going with me. It looked okay to me visually, but that's not a 100% accurate assessment :-)

It's not important to me, but I'm not sure what you meant? The terminology between..... shorted & open.
Originally I thought you said you had the connector disconnected and the fan started?
Now I'm understanding that you had it "shorted" and the fan starts. Which is it?

I did both. I had the fan disconnected and the fan came on and the bike also started.

Then I shorted the connector to show a 0 resistance condition. The fan also came on in this condition, but I did not attempt to start the bike which I should have done.

Sorry for the ambiguity

Also finally got the part number of the sensor that is on my bike:

0 280 130 032

Looking into the specs for that sensor now.

www.bosch.com.au/content/language1/.../sensors_temperature.pdf‎

In case anyone is interested this data sheet lists the specs for the temp sensors used on my bike and I believe on some of the other K bikes as well.

That is the sensor which the Jetronic using.
It seems like someone allready have changed the sensor and got the wrong one.

@ibjman wrote:I still think we need to test the sensor readings at the ecm plug to verify that both the connections all the way to the sensor are sound.

Finished my continuity checks after running through my guess and check method since I didn't have the wiring harness pin out and I have continuity from both plugs on the temp sensor all the way back to the motronic unit harness plug.

@Inge K. wrote:That is the sensor which the Jetronic using.
It seems like someone allready have changed the sensor and got the wrong one.

Sorry misread on my part.....what I read to be a 032 turned out to be a 055.....thought I had something there......tough to see

I also did think that the reason to the problem was found.................

Did check the Bosch # 280 130 055 in a cross reference list, it's the correct sensor.
The sensor using contact # 21 & 32 at the Motronic connector.

Thanks for all of the replies and help.  I have installed a manual fan switch that grounds the relay directly which should allow the overheat circuit to work regardless of the position of my switch.  This has so far kept my overheat situation under control.

I still plan to further investigate the reason that this switch may not be activating, but in the meantime I have to deal with a gear box leak!  I will post my results when I have time to remove the coolant temp sensor and test it.

Don't remove the coolant sensor to test......instead.....locate the wires to/from the coolant sensor at the fuel injection computer harness plug and test the values there.

ibj I had planned to measure the temp of the water that I was heating and to reference my multi-meter off the values I recorded. Not sure if that's the best way to do it, but that's what I came up with. Figured if I could control the heat (on my wife's stove with her thermometer) then I could accurately measure in resistance what the sensor would be sending the computer. Thoughts?

The bike is in pieces right now so no big rush ;-)

I have done it on the stove but be sure to use pure water as many town water supplies have enough salts and impurities in them to conduct electricity and will give you false readings. If you leave it in the small riser tube and place the bottom of the tube in the water without getting the sensor in the water you can use any water and add salt to raise the boiling point to about 106deg C. You need a lot of salt to get to 106 probably something like 1/2 a cup of salt to 1 pint of water but if you get both the connections in the water you get real crazy readings.