So I recall having a similar problem about 4 years ago when I first put together my zed pro. At that time I had no tools other then a cheep DMM and was just starting my computer hobby journey so I knew next to nothing. Didnt have a scope yet so couldn't check timings of reset line rise/fall.
Anyway I was seeing very similar behavior as you describe, where the backplane reset worked fine, andthe dual clock reset would lock up the machine. I did what I could to try and debug this issue with just my DMM, which meant poking around to see if anything was dead. Not the ideal solution but it did reveal something anomalous. I found that if I probed across U1pin1 and U1pin4 of the clock, the computer would just boot. At the time I just assumed that the U1B inverter was bad so I just whacked a 10k resistor across
U1pin1 and U1pin4 and boom, card has worked since then.
Couple of caveats. I now know that this is kind of a nonsense fix, that at best acts to pull the reset line low faster or in a less bouncy manner, but really a scope would have allowed a proper diagnosis. I also seem to have swapped C1 to a 1uF cap at some point, so YMMV. Finally I have no clue why changing the rom would cause this effect when it wasnt happening before, so that's weird.
Dont know Id advise this fix explicitly, but at the same time it really shouldn't hurt anything. If you do try it, let me know how it goes.
- Tom P