I didn't put 2 and 2 together, so it took me over an hour to figure this out.
First off, I'm tricking the machine, to which Metallus will probably say serves you right. My clockmaking requires some small holes and very precise (in diameter) holes. I use the Carvewright to drill pilot holes and a drill press to drill the actual holes. For the smaller pilot holes, I use a 1/16" bit drilled only .1" deep with the Carvewright. For me and others who don't have a 1/16" drill bit, I substitute the 1/16" carving bit. This worked just fine. It puts no stress on the machine as the hole is so shallow.
Today I'm machining a part and I get a z-axis stall during calibration. I checked and lubed the bearings. No help. Pulled the encoder cap, and there was a little pile of dust in there. Cleaned and re-installed. No go. Replaced the flex cable. No go.
But now I'm noticing that the bit is travelling fast down to the bit plate, and I get the error when it hits it. I removed the bit and tried again. I see that the head moves fast at first, then, when it gets near the anticipated bit length, it slows down. Aha! I substitute a shorter aftermarket carving bit, and it works. Apparently, the calibration procedure has changed - the old firmware does not use this 2-speed motion. That's when I recalled that I just updated firmware on this card yesterday.
BTW, a work-around apparently is to select a different cutting bit. When I told the machine I was using the longer 3/16" bit, instead of the shorter 1/16" cutting bit, the head slowed before the 1/16" carving bit touched, and I got no stall error. I did not follow through and complete the project - I will try that later.
I see this change may save a few seconds, but I don't think I like it. Making assumptions about bit length can limit the use of aftermarket bits - surely that is not the reason behind this. (In my case, as reported elsewhere, I have had burning issues using newer Carvewright 1/8" cutting bits. The only relief I've found to date is to use aftermarket 2-flute bits, which in my ER11 chuck are longer than standard. That works fine now. If a future firmware change prohibits my use of these bits, I will be in a fix.) And, reporting a hardware error due to a bit length difference sent me off on a wild goose chase. Unintended consequences I hope.
Anyway, I thought I would pass this along to others. If this firmware change stays, I will need to amend the instructions for the Mystery Clock. It would be a shame for others to get sent down the path of pursuing a non-existent hardware problem like I did, and the project cannot be completed as written with the new firmware.