Sorry for this essay but I am having a bit of a nightmare diagnosing this issue as I have had a couple of interconnecting items fail which are confusing the matter but here is what I do know so far

Car has been loosing coolant – temperature gauge remains steady at 90 all the time once the car is up to temperature.

I have (right or wrong) assumed coolant flow as there is a steady flow from the small return line to the expansion tank :wonder:

Initial checks had the fan working once to temperature and no signs of leaks suggesting the head gasket as the most likely culprit however the oddity I found was that when I did a long run I couldn’t avoid I didn’t lose any more coolant that a short run. It was only when I stopped in traffic that the coolant alarm came on due to lack of coolant (still no overheat according to the temperature gauge)

The other day I noticed that the fan is no longer running at the end of a run and on inspection found that the thermostatic switch actually contains 2 switches and one of the wires had been rubbing and severed. now I realise that I did not fully test this and it was probably only running as the air con was on

Having repaired the connections and tested that both circuits operated (one switches the fan in at low speed only with ignition on and the other at full speed irrespective of ignition) I found that the temperature sensor was also faulty so ordered a new one

I have just fitted it and when testing found that the radiator was not heating up – can’t believe it but the thermostat has failed in the last 48 hrs!

since I can’t get replacements until the morning I have temporarily removed the thermostat to continue testing and now the cooling fan comes on when I turn on the air con but not when the engine is up to temperature…..looks like the new switch is also faulty

Took the car out for a quick run to make sure I got it hot as without the thermostat it will take longer to build up heat and now had the temperature warning flash up as the temperature gauge climbed over 90 for the first time….

As it stands I am fixing the items I know to be faulty as I can’t tie this down with interconnected faults muddying the waters but I have a few current thoughts

Head gasket – if it wasn’t at fault then an overheat is a great way to make it fail so not ruling it out!

Thermostat – although it did completely fail it could have been constricting the flow causing the other issues however with it removed there should have been no reason for the car to overheat other than….

Water pump – a lack of circulation could certainly cause these problems but this was recently replaced when I have the cambelt done. Although this work is warrantied by the garage that did the work I would want to be certain that it was at fault before taking the car back as It would be a big bill if I sent them back in and its not at fault.

Is there any way to check coolant flow?

Any ideas spring to mind? Have I overlooked something?

I don’t mind admitting that this one has me feeling like I am going in circles

Have you bled the system ?

Sent from my GT-I9000 using Tapatalk

Hi, yes the system was bled again tonight when I replaced the fan switch.

Filled until it ran from the sensor housing then filled and top hose squeezed until all air came out.

Ran until no more air bubbles from the return to the expansion chamber and topped off.

If there is a better way then I have a feeling I will be doing this a few more times before this is sorted :-)

Well I re-bled the system yesterday to be sure and from todays runs can add :

If I thrash the engine I can induce an overheat but the temperature drops back to normal quickly.

Without any indicated overheat there is still coolant loss but as before it only seems to happen in urban driving.

a sustained motorway cruise does not seem to be losing coolant but within a few miles at town speeds the low coolant alarm is back.

the system is pressurised however, and on releasing the pressure the coolant returns to the max level which does rather suggest there may still be an airlock somewhere as the coolant should not be compressible.

Any suggestions on a better method of bleeding the system?

Well it gets stranger by the day…

Noticed today that the cabin heater blew cold for a while even though engine was to temperature.

Still not convinced the system was bled I looked for a convenient place to get access and split a join by the heater matrix connections. I have then used a hose to push water through the system from both directions and lots of air was pushed out.

Having just been for a run the cabin heater blows cold at what seem like completely random intervals!

Sometimes at low revs, sometimes high. Sometimes preceded by a drop in temperature on the gauge which then rises back to normal though the heater is blowing cold sometimes with the gauge at operating temperature all the way through.

Of course it never blows cold at standstill when I am able to test :wonder:

Next plan of attack will be to get a temperature probe and test at various points through the coolant system. I still have a nagging doubt about the water pump so want to look for a good temperature gradient across the radiator as this is the only way I can think of checking for a good flow.

If anyone has any ideas then please feel free to jump in as I feel like I am going slowly mad and could even be talking to myself

Still got an airlock somewhere, phone a plumber.

Still got an airlock somewhere, phone a plumber.

think I need a bigger hammer

never found a problem there wasnt a big enough hammer to fix, this one is doing my nut in though.

oh well, next round of bleeding tomorow night then.

seen so many posts saying that "these engines dont airlock" on several forum - have to say, my experience is a little different

Something to cheer you up.

Something to cheer you up.

:rofl: thanks

I think I have a plan for tomorrow....

will connect a hose to the drain valve and work my way around blocking and opening pipes to direct the flow through the cooling system. if that dosnt shift the airlock then I may have to resort to some TNT

I am wondering about the water pump, could be that the impellors are damaged so it can't push enough water round the system when you thrash it. Also bits of impellor could be causing blockages around the system at random intervals.

I am wondering about the water pump, could be that the impellors are damaged so it can't push enough water round the system when you thrash it. Also bits of impellor could be causing blockages around the system at random intervals.

this is the item that would explain all of the faults. I cant think of any way to test other than removal :S

I had the car mapped a while back and as this was about the right time anyway I sent the car to an independent VAG garage to have the timing belt done early. At the same time I had the water pump replaced with a metal impeller version as the plastic ones have been known to fail and at the time it seemed sensible. :wonder:

That work and the pump is covered by the garages warranty but without knowing for sure that the pump is the problem I don’t want to be billed for them to go back in and find it is fine.

Either way I will head over and have a chat with them at the weekend if I still cant get it sorted but I would like to figure out the problem if possible.

If it is a metal impeller pump then I wouldn't think it would break up. Perhaps the old one had already started to break up an there are bits floating around the system. Though that would perhaps risk a fragment jamming the pump which could lead to it throwing the cam belt off.

Alternatively could the water pump be defective so that the impeller is not being driven constantly? I don't know the exact construction of them though and whether this is possible.

If it is a metal impeller pump then I wouldn't think it would break up. Perhaps the old one had already started to break up an there are bits floating around the system. Though that would perhaps risk a fragment jamming the pump which could lead to it throwing the cam belt off.

Alternatively could the water pump be defective so that the impeller is not being driven constantly? I don't know the exact construction of them though and whether this is possible.

yeah its a double edged sword...

I would hope that if when they removed the old water pump had they found parts missing they would have raised it as a concern but you never know

from what I had read at the time the plastic impeller would fail on its connection to the shaft. I would hope that was not possible with the metal one or at least far less likely. but with my luck at the moment nothing would surprise me

Well last night I replaced the thermostat and spent a long time forcing water through the cooling system from all directions until I was fairly certain that no air could remain….

Testing showed that the thermostat is opening and the fan switch operating although both seem to operate far slower than I would expect. The engine has been up to temperature for 5-10 min at fast idle before the thermostat opens and its almost impossible to get the fan to kick in but if I hold the revs high enough for long enough then it will (briefly)

On a 50 mile motorway run again with the cabin heater on max the temperature of the heated air would fluctuate dramatically from very hot to tepid to cold. It is still fairly random but it can be triggered more often than not when the revs go over 3K.

I think I am going to have to go back to the garage that fitted the water pump as I cant think of any other fault that could cause these symptoms although the lingering doubt is that I would have expected if I was getting reduced or no flow that the temperature gauge would show an overheat as once flow returned the sensor should be seeing a hot blast of water but the gauge is now holding steady at 90 :wonder:

Any suggestions always welcome

On a 50 mile motorway run again with the cabin heater on max the temperature of the heated air would fluctuate dramatically from very hot to tepid to cold. It is still fairly random but it can be triggered more often than not when the revs go over 3K.

Have you got two problems ?

1 - Coolant loss

2 - You may be confused by the cabin heater fluctuating, it's very common for the 'flap motor' in the cabin heater system to fail giving the wrong hot - cold temperatures expected.

DB.

OK suggestion...

It sounds to me like it is nothing to do with the thermostat etc. any more but everything to do with the heater control. If I recall correctly the heat into the cabin is set by the dial on the centre console but this isn't a simple mechanical connection like cars used to have. Instead there is a flap, operated by a motor that controls the mixing of hot and cold air entering the cabin. It is not uncommon for these to fail and oscillate giving a fluctuating temperature so my guess is that is your problem - not the water pump, thermostat or temperature sensors...

Have you got two problems ?

1 - Coolant loss

2 - You may be confused by the cabin heater fluctuating, it's very common for the 'flap motor' in the cabin heater system to fail giving the wrong hot - cold temperatures expected.

DB.

ahh, damn that makes sense

the more I look into this the more faults I seem to be finding

I can certainly hear the flap close when I shut off the engine but if its an internittent fault then thats not telling me anything either

I was just using the cabin heater to try and give another way of checking what was going on but from what you have said this sounds like its not a reliable way to get information...

I still have the coolant loss so I need to find a better way to determine the problem.

the coolant system is pressurizing blowing the coolant out through the pressure release cap which would normally suggest the head gasket but the issue with that is I would expect on a motorway run to be losing coolant at a constant rate if it was the head gasket whereas I have found that it seems to hold up fine on a motorway run and only blow out the coolant when I slow down. This to me suggests the flow rate but I just don’t know.

I have the parts to replace the head gasket so I may just have to bite the bullet and change it but it’s a lot of work if I then find its not the problem.

I think I will try to get a sniffer test done on it as at least then a positive would tell me for sure, only problem is a negative sniff test doesn’t mean it’s not the head gasket.

Scanning the car for faluts will tell you if there is something wrong with the flap in the heater.

Scanning the car for faluts will tell you if there is something wrong with the flap in the heater.

no faults on a scan :'(

just had a snif test done - negative of course so that wanst much help but I did have a long chat with the mechanic who agreed that as far as the cooling system goes I had checked everything and the upraded water pump was not likely to have failed which only really leaves the head gasket.

guess I know what I will be doing this weekend

fingers crossed that cures the problem :wonder:

Was it checked with VCDS or just a generic code reader. A simple OBD reader might not pick up the faults with the heater.

Was it checked with VCDS or just a generic code reader. A simple OBD reader might not pick up the faults with the heater.

simple OBD reader. I will come back to this once I get the overheat sorted out.

think I may end up investing in VCDS one of these days

well after a long weekend of spannering, the head gasket is replaced. It showed clear signs of leaking on number 2 cylender and so far all of the gremlins seem to have disapeared.

touch wood, this is the end of this saga and I can get back to enjoying the motor

Glad to hear that things have improved

I think head gasket issues are a lot more common than people think - after all it's not exactly much between two solid blocks of alu/iron, under pressure it is the path of least resistance.

Glad to hear that things have improved

I think head gasket issues are a lot more common than people think - after all it's not exactly much between two solid blocks of alu/iron, under pressure it is the path of least resistance.

thanks - its certainly a relief

I have known from the start that the head gasket was the most likely problem but you want to rule out every other possibility before heading in there.

I would be interested to know how long the job should take. For me in a reasonably well equipped home garage it took 17 hours ( 5 to strip it down and then 12 cleaning and putting back together) but that did involve a lot of head scratching trying to figure out the best way in and double checking everything. Also the inevitable mini strip downs when I realized that the next part to put back on meant taking off the last 3 bits I had just put in because they were in the way

I did also spend about 3 hours just carefully working on the mating surfaces to be sure they were as clean as possible but taking great care not to damage them.

I suspect the book time will be less than half the time it took me

I had my MPI head off, checked, cleaned, and gasket changed in 6 hours for a first time... Now it would probably be a lot faster if I had to do it again knowing every step in the procedure.

Same with everything really - generally if a repair is done correctly you'll probably never have to do it again.