Thanks Joe & Greg for your responses,
Joe, your comments confirm a lot that I had suspected about exposures and the eclipse calculator. My hunch is the previous calculator I was using was based on the long published tables.
Looking at your examples,
The Libyan eclipse was high in the sky so extinction can be ignored. 1/60 at F7.5 (and assuming ISO 100), is almost equivalent to the 1/80 at F6.3 predicted by the first calculator I was using. I agree that looks overexposed, but a burst of exposures might capture the moment when the proverbial "last little ray shining through the last valley on the limb", and yield a nice coronal display as well
You said the diamond ring lingered during the 2010 eclipse and one of your images was shot 20seconds before C2. I can only assume that is because the duration of totality was significantly longer during that eclipse. The Bailey's Beads simulator on Solar Eclipse Maestro suggests 7-10 seconds before and after C2/3 will be the timeframe for this shorter eclipse - any comments?
The 1/500 at F6.3 (again assuming ISO100) shot, if taken a few seconds later might be "exactly what I'm after"! It was also interesting to look at the effect of cloud on exposure. Hopefully that doesn't affect us in Port Douglas, but if it does, my plan is to try the exposure compensation in Solar Eclipse Maestro - looks like light cloud might need ~2 stops of compensation.
After reading your pages, I think I might proceed with my planned burst runs, fast on the way in and slow on the way out. I might keep the run shorter prior to C2, so as Greg suggest it doesn't tie up the camera while it clears the buffer. It doesn't really matter how long the camera is tied up for after totality ends, so I might shoot that sequence from well before the predicted C3, in case timings are a little out! I have ordered a new CF card that should be faster than my existing SD cards. I'll have to do some testing to see how fast the buffer clears when writing my scripts. With the automatic scripting for camera control, you must wait for the buffer to clear before it can accept a new command from the computer. Xavier tells me this is a consequence of the way the internal camera software is written, no way around it. I do intend to switch the camera to DX mode for these burst runs, so the file size is substantially smaller than full-frame, meaning the buffer will clear sooner. I won't need the field of view provided by full frame to capture the diamond ring - assuming I have the sun centred in my frame!
Greg - I agree with your comment about "underexposing slightly" with the D800. You can definitely pull detail out of the shadows, but with every camera, highlights are gone.
Overall, preparation for the eclipse has been a fascinating process. In a lot of ways I'll be glad when it's over - I can get back to a lot of other things I've been neglecting.
See you in Cairns!
DT