I have taken a different approach to a repair, which has shown to be somewhat effective under controlled conditions - cheap, and not a drop of Captain Tolleys in sight.
After some head scratching, I managed to remove the headlining and clean up the roof channels on both sides.
Although I saw a drop of water on the underside of the cassette a few days ago, it was only one drop and I wanted further convincing where the leak location was. How a single drop translated to a flooded floor pan did not add up.
I dammed both the OSF drain hole on the cassette and the front lateral section and filled the front corner with water so the plastic-aluminium joint was fully submerged. There was no indication of a leak from the jointed section. As others have suggested the frame is weak and can flex when pushed from beneath, I applied a little pressure to the overlap from below but still nothing.
I’ve been puzzled by the irregularity others have described trying to correlate rainfall (or lack thereof) with leaking. It would seem if that jointed section is fully under water, it will never see more water than that, has minor hydrostatic head pressure and should leak. Given it doesn’t, Are we really looking at the right point of ingress?
Anyway, I continued to slowly pour water into the dammed corner until it finally poured into the cabin. Thankfully I had prepared the seat for this as it was free flowing.
Water exited from the usual place others have highlighted - the rectangular slot and opening above the worm drive (image below).
Considering I thought the corner I had dammed was enclosed plastic, I was curious as to the communication to the holes below.
There is a huge gap between the outer sunroof seal and the plastic front tray where the cotton bud is inserted (image). Water can obviously pass through here very easily and is more in line with what I expected to see given the depth of water in my footwell. The plastic tray should never be that full though, and since the drains flow well enough. I remained puzzled as to how it entered the cabin.
Finally with the roof closed, and the headlining out, whilst wetting the front of the sunroof, I managed to see water pouring in. It came down the gap with the cotton bud inserted, not the joint everyone has tried to seal with Captain Tolleys.
When wetting the roof, there was ingress at the front (and possibly rear) aspects of the outer sunroof seal. When water passes the seal, it should run into the plastic tray and down the tubes. No prizes for guessing where it actually goes…
There is a skirt on the same seal similar to a shower screen seal, which is supposed to guide the water into the tray. However it sits forward of the tray with no overlap. Water passing the seal of the closed roof is capable and does flow directly through the rectangular slots in the first image (there are two slots per side but only one is shown here).
Since I found a leak which flows into the car with the roof closed, when a relatively small volume was sprayed over it, I think it was reasonable cause to fix that and ignore the plastic-aluminium joint.
Considering the size of the gap, there’s no way you’d seal it up. This needed a different approach. I was thinking roof tiles. Not literally, just the principal. To give the water a helping hand into the tray. Since I wanted my roof tile to have no - to minimal interference with the operation of the roof, it needed to be ultra thin yet water proof.
Off I went to buy a bicycle tyre. £4. I was thinking narrowest but largest diameter, so a road bike/racing tyre.
I sectioned the valve out, then gutted the remain tube so it opened flat. I inserted this under the sunroof seal, lapping it over the lip of the metalwork and into the plastic tray.
Although the gaps I was trying to bridge were only at lateral aspects and whilst the factory seal overlapped much better in the central position (above the rear view mirror), I decided to fit a full length of rubber as it would be less likely to move around than two independent sections.
Given the amount of excess sitting in the gully, I cut it in two so I had two long strips. That would allow the wind buffer to fit back and not trap water on the underside of the tyre. I neatened the ends so the didn’t interfere with the guide rails and opened and closed the roof a number of times. There was no movement which might lead to it moving out of position.
Time to test the water - literally
Sunroof closed, I gave the area a proper soaking with a specific focus on the entry points of the roof seal I’d found earlier. Nothing came into the car but a trickle was exiting at the designed exit points behind the doors. This was encouraging. Opening the roof again revealed additional trickles on the rear facing aspect of the tyre. This is the water that would ultimately find its way into the car during any rainfall.
I’ll keep the headlining out for a while to check on progress but there was nothing following the evening rain that followed.
So, if you have tried sealing the OP’s suggested joint area and had no success, give this technique a whirl and let me know how you get on. It’s about as a cheap a fix as you could ask for (about 1,000 times cheaper than main dealer!) and doesn’t actually need the headlining removing.
I have cleaned and dried the OP’s joint too and will add some Tolley’s as I have seen moisture in the joint.
A note on drying the area, I found using a pick and paper tissue to soak the water out of the joints works really well. The smaller the piece of tissue the better. If it’s double ply, separate it. Aim for an area of tissue about the size of your little finger nail then poke it in. Remove it and repeat. It’s surprising how much continues to come out which corroborates what everyone has remarked upon whilst adding Tolleys to the joint.
Weti