Regarding the problems with the SD cards, it would be quite helpful if someone were to post me one of the SD cards that is not working, or if you prefer, on a UNIXy kind of OS, use dd to copy out the first 128MB of the SD card, zip it up, and email it to me.
I can then look at making the partition table code more robust. There is probably just a stupid bug in it somewhere.
As for whenever a ROM boots to the 80-column MLM monitor instead of C65 BASIC, this is a sign of a definite bug in the FAT file system code, where it incorrectly calculated the location of a cluster on the disk. We need to track down and fix this bug. If anyone would like to go through kickstart_dos.a65 and hunt for it, they will receive a smily face from me if they are successful.