Thanks Tom, Pekka!
Pekka the headstock will only move for a short time while being bored, after that it will be slid to its final position and bolted in place.
The shims only adjust the tightness on the ways while sliding. Not the headstock height. They are located under the slide retaining plate where it is screwed on.
The shims in the headstock are probably unnecessary, assuming you are exact in getting a perfect sliding fit when surfacing the "legs" (bosses?) at the bottom of the headstock. Not too sticky a sliding fit, and not too loose. But I'm used to doing it this way, and it is a simpler matter to make the legs a few thou short, and then pad back to a perfect fit on the ways thickness with shims. (remember also that the ways thickness was actually determined by scraping -- not a generic .375" any more)
For regularly sliding lathe parts (like the carriage) the shims also serve the purpose of allowing you to adjust for wear over the lifetime of the lathe. With one thousandth of an inch brass shims, you can remove one at a time to adjust out wear the same amount. With four separate shims at corners, like I have on this lathe, you can also adjust one corner at a time to remove a small amount of rocking if you have any..
But again, this is all unnecessary for the headstock, as it won't wear over time. Just my habit, and very simple to cut the legs a little short, and pad back to a good fit with shims.
Just a word on why the headstock is made to slide. When boring, I will use the tailstock as a boring bar holder. It will be fixed in place. I will temporarily mount a pulley back there and use the tailstock as a headstock.
The real headstock will be attached with a nut to the leadscrew and move while boring.
By doing it this way, the tailstock and headstock will end up with concentric bores. And also the bore will be exactly parallel with the ways, since the direction of movement is constrained by sliding along those ways. Since I started out with a pre-existing finished tailstock casting, I am operating in the reverse of the normal procedure. Which would be boring a tailstock using a headstock to hold the boring bar.
When boring the spindle location by sliding the headstock over the ways, even if the boring bar is slightly canted, the bore will be parallel to the ways. Canting would just introduce a very slight elliptical shaped hole, but still exactly parallel to the ways. Since I will be very careful to align the boring bar first, any elliptical error will be negligible. It would take a fairly big and obvious angle to make a problematic degree of ellipse. If you do the math for a couple degrees off, it still produces a very small circular error.