Part way through a repair here, findings so far. The car has been good generally, bought at 60k and have put 10k on with this the first problem. There is an overtemp light but the engine was knocking so badly beforehand that I saw the overheating on the gauge before the light came on. At that point I stopped to check:
* the radiator fan was on, so not an electrics problem
* no heat from the cabin heater, suggesting a circulation problem
* radiator seemed too cool for the situation
* cooling system was holding pressure as hoses were tight, so no major leak
* system water level was above max (due to overheating boil/expansion)
The puzzling thing to me was no cabin heat, which suggested no circulation and so a water pump fault. I'm good at mechanics but I don't know this vehicle. Since the engine was running fine obviously the cambelt was OK. It seemed unlikely that the impellor on the water pump had detached. The last cambelt change was unknown and I had been meaning to do this anyway so I ordered a water pump and a belt kit (belt + tensioner bearing), £54 from Autoparts in Glasgow.
After some browsing on this site I concluded that the cooling system was marginal and air locks were possible. I put 2l of water back in whilst bleeding the system. It was clear some had blown out as it was stitting under the radiator, probably past the pressure release but it's not entirely clear. Radiator itself is not new but no obvious leaks. So now I have a new theory - it's the thermostat. I figure I can use the internal heater to keep temperature under control whilst I go and get my parts.
Partial success. The cabin heater works so that confirms the water pump. However in-spite of this, a couple of miles later and the needle's in the red again. No thermostat at the dealer, plus it's £13! On order.
Well the thermostat's easy to remove on the side of a main road, unlike a cambelt swap. It's clear why it costs now - it's integrated into the housing. I put it in a pan of water on the stove and it opens up. Bother. Not 100% as a test because the flow and pressure make a difference, but a good indicator. Block behind looks quite clean too, but I put in radflush anyway.
Now I need a new theory, and it's this: it's still the thermostat, which has lost some wax or something that causes it to open at a bit higher temp. I notice that the heater pipes are above the nominal water level: so the cabin heater keeps the temp down until the revs are too low, when some area boils. This rapidly causes an air lock, circulation stops and the car boils. As I have a new one on order I think I'll remove the valve from the old one. However, it's only held together by spring pressure so comes apart in a non-distructive way. The empty housing is refitted. A drive test at this point shows slow warm up as expected, and even at idle it only reaches normal, with no fan.
More later.
* the radiator fan was on, so not an electrics problem
* no heat from the cabin heater, suggesting a circulation problem
* radiator seemed too cool for the situation
* cooling system was holding pressure as hoses were tight, so no major leak
* system water level was above max (due to overheating boil/expansion)
The puzzling thing to me was no cabin heat, which suggested no circulation and so a water pump fault. I'm good at mechanics but I don't know this vehicle. Since the engine was running fine obviously the cambelt was OK. It seemed unlikely that the impellor on the water pump had detached. The last cambelt change was unknown and I had been meaning to do this anyway so I ordered a water pump and a belt kit (belt + tensioner bearing), £54 from Autoparts in Glasgow.
After some browsing on this site I concluded that the cooling system was marginal and air locks were possible. I put 2l of water back in whilst bleeding the system. It was clear some had blown out as it was stitting under the radiator, probably past the pressure release but it's not entirely clear. Radiator itself is not new but no obvious leaks. So now I have a new theory - it's the thermostat. I figure I can use the internal heater to keep temperature under control whilst I go and get my parts.
Partial success. The cabin heater works so that confirms the water pump. However in-spite of this, a couple of miles later and the needle's in the red again. No thermostat at the dealer, plus it's £13! On order.
Well the thermostat's easy to remove on the side of a main road, unlike a cambelt swap. It's clear why it costs now - it's integrated into the housing. I put it in a pan of water on the stove and it opens up. Bother. Not 100% as a test because the flow and pressure make a difference, but a good indicator. Block behind looks quite clean too, but I put in radflush anyway.
Now I need a new theory, and it's this: it's still the thermostat, which has lost some wax or something that causes it to open at a bit higher temp. I notice that the heater pipes are above the nominal water level: so the cabin heater keeps the temp down until the revs are too low, when some area boils. This rapidly causes an air lock, circulation stops and the car boils. As I have a new one on order I think I'll remove the valve from the old one. However, it's only held together by spring pressure so comes apart in a non-distructive way. The empty housing is refitted. A drive test at this point shows slow warm up as expected, and even at idle it only reaches normal, with no fan.
More later.