I fully expect that at least one screen (and, by extension, laptop) will not survive this. Anyway, onwards with more delayed photos!
First off is removing the screen surrounds again. Presumably there's a way you're supposed to do this, but I ended up taking my usual approach to plastic clips which is to jam a screwdriver in and use it to lever the clips apart. This is a bit easier on the second screen as then I know where the clips are hiding.
Above is the R50e's display, below the R50p's (with a set of 802.11 and Bluetooth antennas). Ignore the different perspectives - both screens are in reality the same size.
The inverter board just unclips, and then the screen itself comes out along with the hinges and a cable assembly (including that little white bit top centre, which is a keyboard light!) once the rest of the screws are removed.
Annnd now all my photos are blurry as my camera's auto-focus is made of fail (not to mention that there's a sudden jump to a long minimum focus distance once you zoom in more than a little - I was using around 1.5x zoom to reduce barrel distortion). Anyway, next is to take the hinges off and unplug the cable, leaving a bare LCD panel.
Now I've reached the end of the HMM instructions and am heading into uncharted waters. I ended up using this site as a rough guide to the job.
The CCFL tube is most likely along the bottom of the screen - the lead in the bottom corner is the power cable for it - and so to get at it I probably have to remove that bottom panel. So I peeled back the foil tape, and then undid the many teeny tiny screws around the metal frame. This I then lifted away slightly using the top of the screen (with all the column driver cables) as a hinge, at which point the multiple layers decided to separate and threaten to escape from the panel. This would be Bad, not only because I'd have to make sure they went back in the right order but because I'd need to be very careful to avoid getting dust inside the screen. So I used a tip from another site (I forget where) and taped the screen together.
Next I unscrewed the metal strip along the bottom and then set about trying to remove the tube which I could see was tucked in along the bottom edge. What you can't see from the photo is the strip is actually the reflector and has a sort of U-shape, with the tube in the base of the U and one of the wires running along the front side of the screen (inbetween the frame and the outside of the U). And to make things even more awkward, the reflector is taped or glued in some way to the front of the panel.
This is something that's very hard to work out when you're trying to see what's going on while being careful not to unhinge the frame by more than a small amount lest the many layers separate and escape, and being very very careful not to break the CCFL tube attached to it (did I mention that those contain mercury?). Eventually, however, I did extract an intact CCFL tube!
Unfortunately while the glass envelope of the tube survived, the cable connected to it didn't. That tube actually came out by sliding it along the reflector, which is not something I'll be able to do with the other one if I want to leave the leads attached.
That is entirely beyond my skill to repair. There's nowhere near enough left of the electrode to solder to, and as a result one of these panels will now be totally dead.
That said, this is the failing tube that I've killed - the photos above are of the R50p, which is the panel that I'm trying to repair. The working tube is still intact and waiting for me to take a screwdriver to the other panel. I now also have a better idea of how the thing is actually attached, and with any luck will be able to remove and replace it by essentially dropping it into the top of the U-shaped reflector.
Project frankenlaptop is not over yet!