loading image

Greetings, Lathe Ops, cutting engineers, and aficionados of music recording and or engineering:

R ecently, we were discussing the bedways of the Scully lathes.  I like how the trolley wheels ride on flat surfaces that are claimed from the Aztec / Mayan pyramid-like silhouette of the vee ways.   In order to address a problem with the Finish Cancel Trip signal (discussed below), I had to remove the carriage.   While it was off, I decided to do a photo shoot to add to the

documentation on the ways. 



As for the "ways" they made them, these Scully pyramids "boogie" "mesas."    Notice, also, how the wheel at the tail end of the bed goes. " \_/," so as to hug, with discouraged, but not disallowed, wiggle, providing the rudder of the carriage an even footing

loading image



These Finish Cancel Trip micro-switches were nearly boggling me mind, for a while.    I know that they say to check the connector, but it was the last connector I'd ever had thought of, since it is normally hidden, seldom disturbed, and was still tightly connected.   The micro-switches are bundled in a snake, which mates with another section of snake, under the lathe bed.   One of the pins on the male connector, there, was invisibly oxidized enough to sort of fall asleep, electrically (or that's what I've decided).   Once I unplugged that connector and replugged it, together, it worked.   I put some Caig DeOxit on there and plug-cycled it about 20 times.   Have been doing contact maintenance on all the surfaces.   Alas, Mr. Grundy did not restore this vintage machine before I bought it.   (Note: Caig D5 and Gold (follow-up) done to all contacts, September 2014, with immediate improvement in logic supply efficiency).




Before I found the solution, I naturally checked every transistor and integrated circuit on all seven boards. 
The micro-switch showed continuity when tripped.  The wire ("W") on the pin of the turntable speed selector switch,
which is designated for the 12" Finish Cancel Trip,* was showing continuity with its destinations in the card cage and
with ground, when selected.  

Yes, I did have the ability to do manual mode lock-out cutting at the end of a side, but when the thing is working and is
automated to precise RIAA disc-diameter specification/suggestions, without the operator breaking more of a sweat than,
say, doing the work involved in selecting a side on a juke box, I could not rest until the answer was found and the problem
solved.   (rest, resumed)

 * Finish Cancel Trip happens when the carriage has been fed far enough to the left that the micro-switch, which causes the
Feed to be canceled, without stopping the turntable or lifting the head before a complete revolution and a half has occurred,
thereby creating a locked groove at the end of side, is tripped.



I was concerned about the amount of turntable runout, and, when the tightening and loosening of the platter hold down bolts was insufficient to tame the variance of flatness during the revolution of the platter,  I resolved to check the isolation mount posts for their correct adjustment.   Fortunately, only one of them was not correct.    One can loosen a nut under the steel weldment and, after removing the aluminum shroud, rotate the isolation mount post to raise or lower the given corner of the A-table.  

loading image

Bottle regs:  "Lefty Loose and Righty Tight," for off.....

Normally, two assistants are required so that each end of the A-table can be supported while the spirit level is made to
rest horizontally and you rotate the coffee mug-shaped post to make the necessary adjustment.  Then you tighten the nut,
underneath the weldment, and replace the shroud... 

In this case, the other posts were correct, already, so I was just able to rotate it by myself, with only some difficulty -
"sweat equity?," but with the WPI (work progress indicator) that the air bubble had become, once again, centered throughout
the turntable revolution.  

It is now holding that corner up flat, and the depth feeler is showing the runout to be, once again, reasonably in spec.   The goal is +/- 1 mil during one slow
revolution (by hand).

loading image

While doing the photo shoot, I thought I'd document the captive suction mounting vacuum line.   
The original version of the LS-76 was like the Neumann VMS lathes which have a vacuum tower above the table, with a hose terminating
in a chuck which is put on top of the spindle, once the blank workpiece is in place.   This causes vacuum pressure to be an invisible mandrel,
using the system of air canals that provide suction mounting through holes in the platter.   There was no edition of the LS-76 manual which
demonstrated the captive vacuum line, but it is in this Scully (#656), and also, at least, on #660.   



loading image


...and, topside (shown, below - taken while platter was removed, of course):

loading image




There are three threaded holes on the top of the turntable that the platter bolts onto.  The dipstick hole is just showing over the rear edge of the spindle.




























H
ere's a view (below) of the strobe light mirror window, next to the turntable.   
The clear mylar belt used for turntable rim drive is resting next to it.   The top of this housing is normally covered by
the platter, and the light bulbs are thereby concealed.   But the little window looks in on a tilted mirror, which reflects the light
display from above for non-glaring optical confirmation of steady revolution at the selected speed.  This is rather like a periscope. 
The article, only half-jokingly titled, He Who Lathes Best, mentions that Scully made a lathe for the Navy to use in a submarine.
Perhaps there was some cross-polination of design and engineering, over the years, between Westrex, Scully, The Armed Forces, and
Tom Dowd.  One can dream, can't he?



image loading