base timing keeps advancing by itself...
2) timing curve is custom, done by Ray, motor ran well for over 5k miles over a summer with no major changes to timing. 14 degrees initial, 22 mechanical with a fast curve. The problem began after storing over the winter and spring. Detonation from 2k-2500 under load, did not pop up right away but after a few tanks of gas.
Define that 22 degrees better. Is that measured at the balancer, or added by the distributor? 14+22 = 36, which is reasonable. But 22 total is not. It's not clear from your statements which one it actually is, or I've skim read too quickly.
No carb is ever "perfectly tuned", especially for e10. Between the water solubility and the fact that e10 is not stable (it's UP TO 10%, not just a straight 10%) there's no such thing as tuning perfection. Modern efi engines can deal with it fine, but a carb has more issues with it. The only way to know if your car is working consistent is to eliminate other inconsistencies and see if it has an effect or not.
I think you have much too much faith in a lot of the tuning parameters of this engine. Doesn't matter how low compression, low power it is. It shouldn't rattle under load. It could be lean, it could be a cylinder to cylinder imbalance. It could be too little or too much timing... The problem is that none of this can be ruled out until the accuracy of the measured information can be confirmed. An engine can "run great" while slowly, or quickly, destroying itself. Simply jumping at the timing may not be the solution to the issue. Ignoring other possibilities is a great way to wind up looking at a rebuild instead of a rejet or new distributor springs.
A piston stop is nice, but not needed to rule out a 10 degree slip of the balancer. A long screwdriver will do. A dial indicator could work too. More than one way to skin a cat.
A piston stop isn't needed to check the spark plugs and look for signs of detonation or cylinder imbalance. Plugs can also help shed some light on what's going on with the timing too, which could corroborate whether the balancer is actually suspect or not. They can also shed some light on actual mixture and mechanical condition. A compression test is a good way to ballpark timing and engine health as well. Your engine is doing something it shouldn't, and which is uncommon. It's best to get to the root cause instead of bandaid or assume one unproven approach will fix it.
There's a lot of gruff personalities on this site, but they're still here to help. The fact that the timing wasn't actually moving itself, despite your conviction to the contrary, tells us how experienced you are with this stuff - you're not. Which is fine, theres all kinds of things all sorts of people don't know, but it means that no one can take any info as gospel until it's confirmed or proven. The requests and suggestions being made all have a purpose, and you're just not going to know exactly why until you get a better grasp on the systems in these cars. If you do follow some of them, you're likely to learn a whole bunch in a hurry.