Well I spent about 4-5 hours on the job today but its still not finished, so I've had to take tomorrow off work to finish it as that is the last non-raining day for quite a while.
Changing the thermostat on this engine is the most unbelievably, unnecessarily difficult job imaginable.
The amount of stuff you have to remove to get the damn thing out defies belief. Aside from the obvious suspects like the air filter box I also had to remove:
1) Air filter sock support.
2) Main steel supply pipe from the pump into the pressure regulator (I only removed it at the pressure regulator end and removed most of the mounting supports for it so it could move it out of the way enough)
3) Mounting bracket for the gear selector lever so that one of the thermostat bolts could be removed.
4) Loosened the two bolts that hold the plate beside that to tilt the plate out of the way a bit
5) All the wiring loom up around the throttle plate area had to be completely extracted to make room to get a screwdriver in to undo the hose clamp on the smaller of the two coolant hoses at the rear of the housing, which was a complete pig to get the hose off.
6) A curved black bracket down on the right hand side of the thermostat housing has to be bent out of the way to get at another bolt it hides, (wtf ?) and that bolt is at a depth that is too deep to reach with an offset crank spanner, and too shallow with the clearance that is available for a normal ratchet an socket extension. I had to come up with a very improvised solution to get that bolt in and out, consisting of a deep reach socket, a universal joint drive and an extension bar used in place of a ratchet!
7) The thermostat was stuck fast in the side of the block so I had to try to remove the housing past the thermostat, which gives far less clearance than refitting it with the thermostat sitting in the housing - but it JUST possible to do this, and I mean JUST.
I finally got it all out and although I haven't got it back together and tested yet (reassembly is about half done before I gave up for the day) I believe I've found the true cause of the problem. As soon as I got the housing off I noticed this:
The picture doesn't do it justice - it's a hard white, rough lump on the side of the thermostat at the end where it would be trying to slide through the gasket. There are actually two of them about 120 degrees from each other, only one is visible in the picture.
Under the gasket in the housing there is a ledge divided into three parts that locates and supports the gasket. There is a minor design flaw IMHO in that the diameter of the ledge is only JUST slightly bigger than the inside diameter of the gasket, so the clearance between the stainless steel cylinder of the thermostat supported by the gasket and this alloy ledge is ridiculously small - I'd say only about 0.1 - 0.2mm clearance at most.
So what has happened is two of these ledges have corroded and pitted leaving both pits and raised bumps of hard aluminium oxide, with the bumps actually scraping on the stainless steel cylinder as it moves. On one of them with the gasket in place the alloy ledge sticks out visibly proud of the gasket. It's difficult to see the corrosion in this photo (gasket removed) but I've highlighted the ledges that are corroded and pitted:
And here you can see the very obvious smooth mark where the oxide has been rubbing against the thermostat cylinder, along with the oxide bump at the end:
That's not something that has only started happening in the two weeks since Forte was added, however, it does look like the Forte built itself up on the lump of aluminium oxide that was embedded on the stainless steel cylinder making the bump even bigger - it may have tried to "seal" the gap between the two contact surfaces and this was the final straw that took it from rubbing and sticking slightly to jamming shut altogether. Some of the bump rubbed off quite easily - possibly Forte, but what is left is hard and embedded enough that it won't come off with a fingernail.
If there had been no corrosion Forte would probably have been OK, (although I'm not sure I would risk it knowing what I know now, if the state of the thermostat was unknown) and I also believe that had Citroen given just a bit more clearance between the alloy ledge and the thermostat then a little bit of corrosion would not have been able to expand the ledge until it touched the cylinder. They could have easily given it 0.5mm to 1mm of clearance as the gasket is very hard and strong and the ledge is only there to locate the axial position of the gasket - it doesn't need supporting right out to within 0.1mm of the edge in my opinion, so I consider it a minor design flaw.
My solution was to carefully file all three ledges back with a rotary drill file attachment so they're recessed about 0.5mm behind the gasket, so that there is no possibility they can come into contact with the cylinder, even if they were to corrode further. So hopefully this won't happen again!
I have the thermostat housing back bolted in place but the rest of the grunt work of reassembly and of course filling, bleeding and testing is still yet to be done. Fingers crossed that goes to plan tomorrow.
As for the saucepan test - the new thermostat opens at 82 degrees as specified and opens about 10mm by about 86 degrees. The old one starts to open at about the same temperature however only opens about 7mm and not until nearly 90 degrees - so it works, but it is "lazy" and doesn't open as much.
To be fair, to get it out of the side of the block I had to twist it and lever it out by the gasket so the innards are now slightly bent so it may not be working as well as it once did! I didn't have much choice though as it was really stuck hard in the engine block, even though no gasket sealant appears to have been used. (I didn't use any either - the rubber flange on the edge of the thermostat looks perfectly adequate by itself!