Also, does anyone know, or can tell me where to find out, what the two small diameter pipes going to the expansion tank do? I can't find a detailed description anywhere, even the Haynes only gives a general overview.

If the fan is cutting in and out,then the temp sensor must be working.?However you don,t know if this is first or second speed.It is possible to do a simple test,when fan is running,disconnect the speed resistor,if the fan stops,then this is the slow speed and the resistor is ok.As for the coolant pipes,these will be return flow pipes back to the tank.With engine running,there should be a flow of coolant back to this .

High speed works - I manually linked out the relay by removing it from the holder and putting a jumper wire across the switched female terminals of the connector.
When i did this to the other relay, nothing happened.

Both relays were removed and bench tested for operation AND continuity across the load terminals! All connections given a good clean and squirt of contact cleaner.

In your opinion is this a failed resistor, or a faulty fan motor?

The only other odd thing is there is absoutely no where a radiator switch could be connected.

(This is what the part looks like i.ebayimg.com/00/s/MjQxWDI0MQ==/z/-WwAAOSwI~VTzKza...G)

There IS what looks like a blank plate on the radiator near the top hose, but the only blue connector in sight is the one that has the resistor mentioned above plugged into it. This is not on the radiator but on the surround/mounting plate.

No loose or disconnected cables were obvious even after removing radiator to access the relay comparment.

Looking in the Haynes they use the phrases "Where fitted" and "If applicable" liberally so I'm hoping there is not a radiator switch on this model

Again thank you for your help elekie

The way I see this so far.the fan motor is fine ,the temp sensor is fine.this car does not have a temp switch in the rad.There is a fault with the low speed circuit,which means a fault with the speed resistor or related wiring.The resistor is easy to test with a digi meter or test bulb and wire across the battery.

Thanks elekie - I will get the tools out again and update this thread.

Can I just confirm I should have only one connector plug at the fan, and that high/low speeds are switched in elsewhere?

I will also add here for reference that cutting some slots in the thermostat surround before installing it again is helping keep the temperature nearer 90 celcius, although of course there is a longer warm up time. Not a solution I would want to use long-term, but it should give the engine an easier life for now while I do more investigation and order any parts.

Wish I'd done it before towing 2 motorbikes 1700 miles to and around the Pyrenees last week. Poor engine was at 100 most of the time, and we had to have the heater on full blast on long motorway climbs and on the steep Pyrenees hills. Fun.

Just one more question - would this fault have caused a car to run hot on the motorway at speeds over 60mph and when climbing hills? I did not think fans operated at motorway speeds

The symptoms were identical to either failed HG or dodgy water pump impellor.

Good to see a result.Once on the move over approx 30mph,the fans are not very effective,unless the coolant temp sensor requests operation .

Thanks - it could be a couple of things then as I originally had a problem with it getting hot over 60mph on the motorway or when climbing hills. Around town and bimbling along on the flat it will stay at just under 90 unless there's a traffic jam, and previously I've driven miles and miles with my foot to the floor and it's stayed at just under 90C.

I might book it in for a timing belt and water pump change as it may be the impellor. Annoying as I had it done only 2 years ago.