Well, I blew up the 3.3v buck converter somehow. The voltages all started dropping and the converter got incredibly hot, and now it doesn't work. Fortunately, it doesn't seem to have damaged anything else. Also fortunately, I had a USB power board with a 3.3v linear regulator on it lying around, which I am now using instead. I've ordered an LD1117 to put directly on the breadboard to save some space.

I spent far too long trying to get my clock generator working this evening. I was hampered slighly by the fact that I don't have anything which can measure a 4MHz signal, and while I had my CPLD set up to divide its clock signal by 2^20, I forgot that I'd also added a button input just to try things out, and so the signal wouldn't be divided unless I pressed the non-existent button. When I reloaded the CPLD with a new programme which didn't require the button, I discovered that my clock is working just fine, and probably all my other variations of clock circuits would have also been fine if I'd been able to test them properly.

I'm going to remove the flip-flop divider from the breadboard and divide the clock on the CPLD instead, because it might be useful having a 16MHz clock in there when I come to adding an SD card, so that it can read and write a little faster. Depends if I have enough macrocells left when I get there of course, and whether the CPLD can cope with having two clocks.

Another thought: TTL devices should be able to cope with the 3.3v signals from the CPLD, but will the Z80 be okay with having a 3.3v clock? We will see…