Hi all,

I left my K on its sidestand all day today, on a slightly cambered road so it was leant over a bit more than normal. When I came back to it around 8 hours later (so it was stone cold) I noticed the pipes on the side of the engine looked like this, all sucked in:





My firt thought was of course that the pump has filled the engine with water but I cannot see any evidence of this, there are no extra layers in the sight glass, the engine still sounds well oiled and the coolant level still reads maximum. After starting the bike I found that the pipe would actually be sucked in even more if the engine was revved, but after a couple of minutes it filled out back to normal again and it got hot as it should normally. Basically, is this normal? Do they do this? I've never noticed it before..

Not an expert on this part but as far as I know when the system cools it should draw some coolant in from the expansion tank via the pipe that goes from the radiator back to the tank at the battery. The cooling system is pressurised and the only vent point is the radiator cap which controls the pressure by letting the excess coolant out to the expansion bottle and then taking it back as the system cools.
If the pipes are collapsing then it would seem this small pipe is blocked [cheap to replace] or pinched....by side panel cover? or blockage in expansion tank [easy to clean] or fault at the radiator cap [Might just need cleaning but it can fail] or faulty thermostat. But I have seen this happen before too without any bad consequences.

The pipe to the tank is not original, it is now a clear pipe that runs where it likes. It is possible I have trapped it under the tank or something but strange that I've only noticed it today when it hasn't been in bits at all for at least a couple of months.

I'll keep an eye on it though!

If you take your radiator cap off (cold motor) and the pipe returns to normal, then there is a fault with the radiator cap.

Renew the cap seals. You may even have to replace the cap as well.

Keep us informed please.

Is there an easy way to get the cap off or is it the tank off job that it looks to be? I get the feeling I'm being very optimistic..

You can get to the cap without taking the fairing sides off but it does require some delicate wriggling.

I didn't really pay attention to the layout of the rad when the bike was in bits, it worked fine and I had other things to fix. Why do the fairing sides have to come off? Do I have to remove the radiator grille? Please no, it was so difficult lining up the screw holes to put it back together last time.. :pale:

I lengthened the fuel hoses from the petrol tank so it will slide back enough to open the radiator cap. Otherwise its a tank off job. Faulty radiator cap likely.

Good tip about making the hoses longer.

Snoddy...remove the black "infill panels", unclip the two "C" clips that hold the tank down (if fitted!) and see if you can slide the tank backwards enough to get to the cap.

If you haven't fitted longer fuel lines, then you'll have to disconnect the lines and remove the whole tank (don't forget to unplug the tank electrical connection as well...right hand side of the tank).

I've just had a look at it in the dark and cold and the hose is collapsed even more than it was before, it definitely can't suck air or water in. I will begrudgingly investigate this probably over a few evenings next week, it has started just as life got busy  I might manage it tomorrow though. I fitted new fuel hoses a couple of months ago to replace the originals (!) but cut them to the same length so it's definitely tank off.

Thanks for the help so far, everyone!

Slide the tank back and then lift the right side and there is enough room to get the cap off.

Live and learn Rick...I was under the assumption that you had to remove the tank!

So many tips, I'll see how it goes! To be honest the worst part is unclipping the side panels, those lugs are much too tight a fit in my opinion and make the whole process quite nerve wracking.

Coat the lugs with silicone/synthetic grease suitable for rubber and that should make life a bit easier.

The first time I removed a side panel on my bike, I broke the bloody plastic clip!

Top tip, I am using basic brown lithium grease at the moment as it was to hand when the side panels went back on but they just don't want to ever come off again.

blaKey wrote:Live and learn Rick...I was under the assumption that you had to remove the tank!

It is tight but I have done it by just enough to get the cap off and put a small tube into the filler and syphon some coolant into it. You can barely even see the cap when you do it.

It's probably the flat rubber seal on the radiator cap. They seem to swell with age and traces of oil.

 I once cut about ¼" out of one and super-glued the ends together as a temporary fix. 

Here's how to never fear taking your tank off  

https://www.k100-forum.com/t3308-fuel-line-connectors?highlight=fuel+connectors

Paul.

Hokay, I actually got to this today. My first step was to remove the right side panel because it was staring at me with the bike on its sidestand and seemed a good place to start. With that off I immediately saw that a cable tie was really pinching the breather tube to the overflow tank. I chopped the tie off but nothing changed, so I went for a run. The water level in the tank stayed the same, well, everything stayed the same. That probably wasn't it.

I then removed the left side panel and knee pads both sides, plucked the tank out from its lugs (no clips here, they seem totally redundant), put a towel over the sharp knee pad holding bracket on the left and moved the tank onto it. You can certainly get to the radiator cap without removing the tank on an RS! I was greeted with this sight:



I wondered if maybe it should be solid water to there and work as a siphon as the tank is way below, so undid the cap (and the pipes expanded back to normal!) and tried to shovel some water in. No dice, the rad at least is still full so the pipes aren't collapsing because they're empty, it's just that the system can't suck inwards. I wondered if maybe there is some BMW voodoo going on with the cap so had a look, it's like this:



Just a space around the outside of the rubber seal of the cap where air/water can be sucked in or pushed out, nothing complicated and a siphon would never be maintained so I guess all is normal. I poked the blowy outy valve in the cap and pulled apart the sucky inny valve so neither is stuck, and then just put the cap back. I sort of warmed it up again afterwards but not properly, though the pipes have not collapsed this time. I will update after a few good rides tomorrow and Saturday.

Current suspect is the cap, it's the only part of the system that can really fail and cause this kind of thing and it could be cleaner inside, perhaps it's just crudded up. Or it has expanded as paul says..

Having read this thread, I checked my LT hoses (on a hot(tish) day, bike on side stand) and was equally surprised that the bottom hose was all sucked in. As with Snod Blatter's, everything else appeared normal and the expansion tank level was where it should be.

Then rode into London yesterday and near disaster! After a nice steady run for the first 110 miles, cranking along at max road speed I then hit the London traffic - 3 lanes of closely packed vehicles without enough room to squeeze through and all crawling painfully slowly along. Very quickly I noticed that my temperature gauge was rising and then realised there was no reassuring roar of the fan cutting in. Temperature quickly in the red and as the traffic wasn't really stopping, no chance to kill the engine when stationary. Interesting the high temp light never came on on the dash (I was waiting for this as my cue to pull over urgently). Next thing I was in a cloud of steam as the system boiled over. Made a quick exit off the road and up onto the pavement (sidewalk) and shut down. Expansion tank was empty, hot fluorescent coolant was dribbling everywhere.

Having allowed the bike to cool a little, I managed to roll back down the pavement and cut into a side road that was clear of traffic. Located a service station nearby (thank goodness for GPS) and managed to ride there to buy some more coolant to refill the expansion tank. Even with the coolant lost, provided I kept moving, the engine temperature stayed normal, but as soon as I had to stop (traffic lights etc.) the temp gauge would immediately start rising. Eventually made it to my destination in North London, 10 miles further on, where the bike was able to sit and cool properly while I socialised and drank coffee/ate cake. 

I topped the expansion tank up a couple of hours later before heading off to see my sister to the south of London. What a nightmare journey that was - riding right through the centre of London in heavy traffic on a Saturday, red traffic lights every few yards. I certainly didn't enjoy the sights, loads of roads were shut off for some unknown reason (particularly around Trafalgar Square) and the GPS was being difficult (kept losing signal - probably due to the lack of clear sky between the buildings). To add to the misery, I had to endure torrential rain (it is Wimbledon and Glastonbury week - so what did I expect from the British weather), but at least that added some external cooling to the engine block! I was relieved to finally get to my sisters, but provided I stopped the engine every time I came to a standstill, the temperature remained OK for the journey (nearly 2 hours to do 22 miles!). Hate to think how many start cycles I put on the starter/battery!

When I arrived, I noted the bottom hose was sucked in but the expansion tank level was half a glass. This morning (I was persuaded to stay over for supper a couple of glasses of wine and a bed for the night - didn't take much persuading), I noted the hose was back to normal, but the expansion tank was empty. Topped up to max and rode home without any problems and no traffic delays, but now the expansion tank is back the the half glass level.

Now to tear down the bike and get the fan working again (it work beautifully in Dublin traffic a couple of months back), restore the coolant level (I don't have any leaks, so think the emptying expansion tank is just it doing its job of topping up the radiator every now and then), and most importantly dry out/flush all the wiring under the fuel tank out with WD40 electrical contact cleaner - I haven't looked yet, but I am pretty sure that when the system boiled over, it vented through the filler cap so expect to find sticky greenness over everything which won't be good for the electrics at all.

Moral of the story - if something don't look right, it probably isn't! 
Pics to follow, if there is anything worth photographing.

Have a good look if the hose between the radiator cap and the overflow tank isn't squeezed somewhere. If not get yourself another (new?) radiator cap. It seems the valves in it are not working properly.

After a couple of motorway runs mine hasn't sucked the pipes in again but then the coolant doesn't get as hot there than in the city. I will report back after getting it really hot some time.

Nice to have someone else on the case but I bet Paul doesn't appreciate his new helping role! 

Snod Blatter wrote:
Nice to have someone else on the case but I bet Paul doesn't appreciate his new helping role! 

Well said Mr. Blatter! - I'm delighted to "help"/share your problem of course....but you're right, I didn't anticipate getting quite so involved myself.

Update on where I'm at...

Side fairings off, tank off ...it is the only way to get at the stupid rad filling cap on the LT (without extended fuel hoses and thin hands). Good news - the fan works when given direct 12V supply. Changed the temperature sensor unit (thank you donor bike) but no way of testing it and will put the rad cap from donor bike on too, while I'm in there. Again no easy way of testing the cap without the proper test unit.

All hoses, including skinny line to/from expansion tank in good condition and not crimped. No evidence of coolant in wiring loom, but given all electrical connectors a good dose of electrical contact cleaner while they're exposed.

Wiring in an auxiliary power supply and switch to allow manual operation of the fan if required (handily, there is a spare/dummy switch on the dash...and I've taken the hazard light switch off the donor bike and fitted it). This has to be a better bet than stop starting the bike all the time if the temperature sensor goes duff.

Rad refilled, tank and fairing going back on the ono and fingers crossed....

If you just want to prove the fan works, have a look at:

 http://www.ibmwr.org/ktech/fan-diagnosis.shtml

Scroll down to the Fan Motor Test section.

Mike

mike d wrote:If you just want to prove the fan works, have a look at:

 http://www.ibmwr.org/ktech/fan-diagnosis.shtml

Scroll down to the Fan Motor Test section.

Mike

Thanks Mike,

I checked the fan as soon as I got the tank off and could get to the connector - applied 12V to the fan motor and it ran.

Tried running with new temperature sensor unit (from donor bike) and watched temperature soar into the red - proved that the excess coolant does go back to the expansion tank! As it cools down over then next few hours, I'll check and see if the coolant goes back to the rad (thereby proving the rad cap is working).

Whist I have a fan override switch now, I do want the system working properly - next component for investigation is the sensor in the coolant stub pipe. Again donor bike should come in handy for the spare...

PaulLipscomb wrote: Tried running with new temperature sensor unit (from donor bike) and watched temperature soar into the red
- next component for investigation is the sensor in the coolant stub pipe.

Which sensor did you first change? The correct one is the one at the coolant stub.....it's a dual element sensor both for the fan and the injection system.

To check the sensor....remove the connector at injection ECU and the fan relay
in the relay Box....measure the resistance to ground at both, you should get around the same values......
At injection connector between contact #10 & 13, counting from the rear end...also the empty slots.
At the relay socket between contact E and 31....

Measure when the bike is both cold and and at operating temp.
If you measure around the same values......the problem is in the relay or the related wiring......much easier to deal with than changing the sensor.

You find a temperature/resistance table in the troubleshooting diagram at the portal.

Good Luck!

Inge K. wrote:

PaulLipscomb wrote: Tried running with new temperature sensor unit (from donor bike) and watched temperature soar into the red
- next component for investigation is the sensor in the coolant stub pipe.

Which sensor did you first change? The correct one is the one at the coolant stub.....it's a dual element sensor both for the fan and the injection system.

To check the sensor....remove the connector at injection ECU and the fan relay
in the relay Box....measure the resistance to ground at both, you should get around the same values......
At injection connector between contact #10 & 13, counting from the rear end...also the empty slots.
At the relay socket between contact E and 31....

Measure when the bike is both cold and and at operating temp.
If you measure around the same values......the problem is in the relay or the related wiring......much easier to deal with than changing the sensor.

You find a temperature/resistance table in the troubleshooting diagram at the portal.

Good Luck!

Thanks Inge K.

Your "Good Luck" did not fill me with confidence! I have done some more checking in line with your recommendations and checking with CFs excellent trouble shooting guide (beyond excellent really).

Ran engine until red lined on temperature gauge and then removed tank (very carefully to avoid splashing fuel on hot engine and/or burning myself on everything hot). Removed temperature sensor relay unit to access pins E & 31 on socket and removed plug from ECU (under seat) to access pins 10 & 13 as instructed.

Results as follows: Temperature sensor unit (in electrical gubbins box under fuel tank) read 12 MOhms. ECU plug read 400 Ohms.

By way of comparison, checked cold readings on donor bike (poor bike won't ever be warm again, but parts ride on) and temperature sensor relay read 2.75 k Ohms and ECU plug read 2.8 kOhms. Won't be able to check "running" bike until tomorrow when she has cooled down again.

I guess, the long and short of this is that the temperature control element (fan control and high temp warning light) in the sensor on the coolant stub has failed and it will be a radiator off job to change it.

PaulLipscomb wrote:I guess, the long and short of this is that the temperature control element (fan control and high temp warning light) in the sensor on the coolant stub has failed and it will be a radiator off job to change it.

For us simpletons does this have anything to do with the pipes being sucked in? Or is it why yours managed to boil over and nothing to do with the suckage? Or is it all interlinked?

Snod Blatter wrote:

PaulLipscomb wrote:I guess, the long and short of this is that the temperature control element (fan control and high temp warning light) in the sensor on the coolant stub has failed and it will be a radiator off job to change it.

For us simpletons does this have anything to do with the pipes being sucked in? Or is it why yours managed to boil over and nothing to do with the suckage? Or is it all interlinked?

Not sure whether the two problems were related or just a coincidence. I noticed the sucked in pipe first, then had the overheat incident in London which made the sucked in pipe thing worse. I am starting to think it was just a coincidence though.

Having spent half the day experimenting, I am now assured that the radiator cap was working properly and discharging or sucking from the expansion tank as it should. When sitting at the road side, enveloped in a cloud of steam and looking at the fluorescent green coolant covering everything, it appeared as if it must have come from the radiator cap (which you can't see with tank & fairing on). Having recreated the boil over this afternoon, with fairing off, it was immediately obvious that the boil over causes the expansion tank level to rise until it sprays out of the tit on the expansion tank cap (straight up wards) - with the side fairing and trim in place, this jet of hot, pressurised coolant would spray all over the place.

Of course, may be I did have a sticky radiator cap and by good fortune (!) when I boiled the coolant due to the fan relay defect, I just gave the radiator cap a boiling hot coolant enema which cleared out gunk? Probably will stay a mystery but I don't have sucked in pipes any more!

Thanks for the report, if I'd known this thread was going to get so serious I would've given it a more grown up title  

The level in my expansion tank has never moved as far as I know, does yours move before it gets boiling hot? Is there an air gap at the top of the breather tube or is it "bled" all the way? I still don't see how it could ever suck in extra water from the tank, the bottom pipe is so soft I'm not at all surprised it collapses first.

Snod Blatter wrote:Thanks for the report, if I'd known this thread was going to get so serious I would've given it a more grown up title  

The level in my expansion tank has never moved as far as I know, does yours move before it gets boiling hot? Is there an air gap at the top of the breather tube or is it "bled" all the way? I still don't see how it could ever suck in extra water from the tank, the bottom pipe is so soft I'm not at all surprised it collapses first.

Sorry, didn't mean to get all deep and techie - your title was perfect as it was before i came along and spoiled/confused it.

My expansion tank level had never moved (not that I had noticed anyway) and like you, I couldn't figure out how it was meant to supply coolant when it was so low down below the radiator filler cap. If it wants more coolant though, it just sucks it right on up (the hose connects at the bottom of the expansion tank so is always flooded and the vent at the top of the expansion tank allows air to be be drawn in so that it doesn't pull a vacuum in the expansion tank). If the coolant gets too hot, the volume expands and has no place to go except down the skinny tube and into the expansion tank, forcing air (and eventually coolant!) out of the vent at the top. 

Ordinarily, with the radiator and pipes all filled correctly, the bike running at normal temperature and the fan doing it's thing to prevent overheating, I suspect there is very little movement of coolant either to or from the expansion tank, so we just don't notice it - possibly a reason why the radiator cap gets sticky. The bottom hose is really soft, the biggest bore too and on the suction side of the pump - as you say, hardly surprising that it collapses first.

PaulLipscomb wrote:
Results as follows: Temperature sensor unit (in electrical gubbins box under fuel tank) read 12 MOhms. ECU plug read 400 Ohms.

Looks like you have found the reason to your problem.
Before removing the sensor also measure at the sensor itself, between each
terminal and ground.
Just to be shure that you don't have bad connectors/wiring....one Connector
at the sensor, next one between the frame harness and engine harness
(left, rear under the tank).

About the coolant hose being sucked flat, check the gasket with the biggest
diameter (as sidecar Paul suggest) at the inner diameter it should be a small
opening between the gasket and the valve housing.
It's three ports around the valve housing, if the gasket swells it blocking these
ports.....letting coolant out, but not that easily in as the gasket being sucket in
to the ports.

It's a frustrating game this one! Cleaned up the water temperature sensor connectors under the fuel tank and swung through with the meter with the engine cold this morning and both read about 2.7kOhms. Thought I'd solved the problem . 

Last night, I successfully removed the water temperature sensor from the donor bike and checked it's readings to prove that it is a good spare (if needed). It was also a useful exercise for tracing the sensor wiring back to the under tank connection.

The water sensor is not so easy to get at (behind the radiator) and I don't have time to do it now on the running bike, but based on the good readings it looked like the cleaning of the under tank connectors had done the trick. So back on with the fuel tank and the fairing. Only once it was back together did I realise that I had put the original fan relay back into the circuit rather than the one from the donor bike. Idiot! However, full of confidence, I thought that maybe the fault was not on the relay anyway (the high resistance readings I had got last night do suggest poor electrical contacts rather than a failed sensor - which probably would be either direct short circuit or open circuit). I can't fully check the wiring from the water temperature sensor until I get the radiator off and get at the sensor and the resistance reading did suggest that the sensor and wiring were now good.

Anyway, just run the beast up....bit of a panic when it cranked heartily and wouldn't start, until I realised that I had forgotten to reconnect the fuel pump/fuel level plug. Idiot x2. Thankfully this plug is very accessible and when connected, the bike fired up immediately. Ran up to temperature...and beyond. Bugger! Still no automatic fan cut in, so maybe the fan relay is duff after all. Still at least I proved the manual fan switch works on demand, so I have temperature control. As I mentioned earlier in the thread, on the open road and moving, the temperature behaves with the airflow across the radiator doing it's thing - it's only when in slow moving city traffic or stationary that the temperature goes too high. With the manual switch, I can now easily adopt "city mode" cooling.

I will get round to fixing the fan relay and/or the water temperature sensor at some point - but could not face tearing the bike down again today. Besides, it would be a waste of new antifreeze/coolant!!

Inge K. wrote:About the coolant hose being sucked flat, check the gasket with the biggest diameter (as sidecar Paul suggest) at the inner diameter it should be a small opening between the gasket and the valve housing. It's three ports around the valve housing, if the gasket swells it blocking these ports.....letting coolant out, but not that easily in as the gasket being sucket in to the ports.

Sorry, I'm not really following this.. Do you mean the seals around the cap? Or around the valves in the cap? Where are these three ports? 

I'm trying to take some pics of my spare radiator cap to show the 3 ports that Inge/SCP are talking about...In the meantime, if you look inside the radiator cap, you should see two flat rubber gaskets (seals). 

• The smaller of the two has a stepped profile that keeps it in place on the end of the hard plastic central boss of the cap and seals against the raised seat inside the radiator filler fitting. 
  
• The larger of the two fits right up inside the cap and is a loose fit around the central boss and provides a seal between the cap and the screw fitting on the radiator filler fitting. This is the one SCP and Inge are talking about.
  

If you look very carefully (easier if your remove the large seal from the cap), you will see 3 equi-spaced semicircular(ish) ports cut into the central boss right at the very top where is is joined to the radiator cap. These are the ports that allow coolant to flow to or from the expansion tank - providing the large seal has not distorted/swollen and blocked them all. Ordinarily, there looks to be a good 2mm gap between the seal and the central boss all the way round, so the large seal should not normally be able to block any/all of the ports.





If you look inside the central boss (my spare has its gauze strainer missing, so it is easy to see) - you will see two small valves. 

• The smallest one looks like a spoked wheel and is held shut by a spring mechanism. This valve allows coolant to be drawn in from the expansion tank (the suction effect pulling the valve off its seat against the spring tension.
  
• The larger one has a black tit in it's centre. If you push this tit, it opens the valve against spring pressure which would allow excess radiator water to be discharged to the expansion tank (as the temperature rises in the coolant system, the volume of water tries to expand - which it can't do, so the system pressure increases which then acts on the black tit to open the valve and discharge water).
  

Clear as mud?

Much clearer, thanks! This forum is really good  

I poked/prodded the two valves on the inside but never noticed the ports around the top, if mine sucks the pipe in again I will interrogate the cap more thoroughly.

Snod Blatter wrote:Much clearer, thanks! This forum is really good  

I poked/prodded the two valves on the inside but never noticed the ports around the top, if mine sucks the pipe in again I will interrogate the cap more thoroughly.

  To be honest, I didn't have a clue how it worked (or even realise how complicated it was) until I started looking as a result of this thread. That's what I love about the forum - it gets you thinking! 

I'll probably never have a problem with sucky in hoses again, but I've learnt a whole lot more about how my bike works!

While out on my jollies this last weekend I found another K100 doing this thing they don't do..



This bike apparently also has a problem with overheating, which may or may not be related. My bike has not sucked the pipe in since I poked it about, how's yours doing Paul?

That looks like another K in the background..........wonder did it catch the suckage from that one.

Same story, check the radiator cap........before worrying about the overheating.

Check thermostat is working ok, by feeling the coolant pipes heating up.

If it doesn't solve it check the rad for being a bit blocked up.....as in flush it.

If its not losing any coolant anywhere the water pump impeller is a thought.

Overheating-does that mean a problem with the fan not kicking in? Sensor or seized up fan?

Actually the new radiator caps are quite different (and only about 12 quid from 'bins)



SCP

sidecar - I had seen that and it is very interesting. Maybe it's an 1100 cap? Or an even later revamp?

The sucking K is fine if the fan is kept on with a switch but will overheat in traffic without it. Being okay with the fan on suggests the water is circulating at least somewhat properly but it could just be gummed up or the cap could not be holding pressure letting it boil over at a lower temperature. It is quite high mileage so anything is possible.

My LT was inclined to run the fan quite a lot in traffic and it had a manual override, something that's great for winter riding.

I figured high mileage when I suggested the impeller. But I think initially I would swap the cap, its as easy to do, but also check the pipe back to the reservoir. If it collapses for any reason it has the effect of preventing coolant being drawn back in as it cools. The cap and this pipe are very small price items and saves the hassle of a tank off job. Also, coolant mix is more effective than pure water so check this is correct, worth doing a coolant change if its not been done. If a service was recently done check the plugs. There was a reference here to using D7EA but over here it seems those in the know are using D8EA which are a colder plug. I am not expert on these but have been using the D8EA for over years.....

You can kill the engine in traffic when you stop and this is very effective in keeping he temperature down. I had to do it when the LT fan packed up.

Snoddy, I've no idea when the design was altered, but I would guess it was to overcome the problems that are now becoming more common with our K's......only a 30 year life span for a rad cap, tut tut!

Paul.

A stationary K has very little cooling. The fins on the sump require air movement to be effective, same with the radiator. Stationary they have very little effect. The fan will therefore run almost continuously in very slow traffic, only cutting out for the time between the low temperature cut out and the time the high temperature cuts in so it does not necessarily mean there is a problem to be worried about. I have had a complete new water pump in the RT, flushed and new coolant. In the warm weather in very slow traffic the fan is constantly coming on but the entire system is running ok.

Finally sorted my overheat issue. 

Having fitted the fan manual over-ride switch a few weeks back, I must admit that I wasn't in a hurry to tear the bike to bits to get at the water temperature sensor (behind radiator). The over-ride switch came in handy several times when commuting though heavy Bristol traffic on some pleasantly hot days.

Then, this weekend, while riding out to look at a "new" car (a BMW - coincidentally exactly the same colour as the bike, I just need to get a bike trailer now for the perfect combo), the temperature gauge started reading erratically and then went full scale. Since I was relying on the gauge to operate the over-ride switch, loss of any meaningful reading was a bit of problem. 

My first thought was that the element in the water temperature sensor had now completely failed and was giving a full scale reading. I guess this was my prompt to tear the bike down. So off with the lower fairings, off with the radiator fairing, off with the fuel tank and off with the radiator. Got the water temperature sensor out without any bother and set about testing it. Totally confounded - both elements appeared to be working fine, reading 2.7K Ohms cold and 400 Ohms when immersed in a cup of boiling water. Tested my donor bike water temp sensor and that behaved in exactly the same way. I guess the sensor was no the problem after all! Throughly cleaned up the electrical connector - it is in a pretty grimy and inaccessible place - and checked wiring continuity was all OK. Just to be on the safe side though, I put the donor bike sensor into the coolant stub pipe, refilled the coolant system and refitted the fuel tank.

Fingers crossed, I turned on ignition....bugger, temperature gauge went full scale immediately. So, off with the screen for a closer look at the gauge itself. The problem was blindingly obvious, the brown (earth wire) to the gauge had broken. Easily fixed by crimping a new connector on and hey presto, gauge reading correctly again. Screen back on, ignition on...so far so good, engine on....set revs to 1300 - 1500 and waited as temperature slowly came up and, hallelujah, at the appropriate point...fan cut in automatically. Yeehaw.

I guess the problems were all down to wiring - a poor connection at the water temperature sensor plug and a broken wire at the temperature gauge, oh, and the original problem with pipe suckage... I put that down to a gunged up valves in the radiator cap (I haven't noticed the suckage thing since) - a latent defect that showed symptoms of things not being right but was not understood until something else failed.

The real bonus, apart from having the bike working again as it was designed too, is I now have a fan over-ride switch fitted as a permanent back up for when I feel like it!

Temperature gauge doesn't use the double sensor in the stub. It uses a separate sensor on the water pump.

+ 1   

Avenger GT wrote:Temperature gauge doesn't use the double sensor in the stub. It uses a separate sensor on the water pump.

OK - thats interesting, didn't know that ....so that was three separate faults that all conspired against me to raise my Knowledge level! 

The temp gauge problem was a broken wire at the gauge, so easy to spot/fix...assumption did the rest (dangerous thing!).