Hi Robert. Intriguing problem, and I think I may be able to shed a little
light. My supposition could be wrong, but if you look at the physics of the
thing, it could point to not one issue (“…obviously the contact point between
the leather nib…”), but more likely two issues. You have to raise your
perception above where the symptom’s cause would appear to originate, and look
at the whole picture.
The key to solving your mystery may be directly related to something you
conveyed very precisely, that the issue occurs over an extremely narrow band of
speeds. So, it appears that we have a primary cause (vibration source at the
leather and flywheel), exacerbated or amplified by a secondary condition, which
I will assume is marginal wear in the holes that the leather pad arm’s arbor
passes through (the first of the two axles), in possible combination with the
holes that the speed indicator needle’s arbor passes through, in the indicator
frame. Given that the steel arm that the leather pad mounts to is supported on
a smaller arbor, and given that this arbor is first in line and thus closer to
the governor flywheel where the vibration originates, it’s likely that wear
would be more present at those holes and possibly the associated arbor as well,
than at the larger holes and arbor of the indicator needle. The same wear on
smaller parts will occupy a larger percentage of area, thus having a more
noticeable effect.
The wear I describe would be very subtle. If for example, the holes that the
indicator arbor passes through are normally three-thousandths of an inch larger
in diameter than the arbor (just guessing on this — I have not measured it),
and it’s known that the arbor touches only a specific portion of the wall of
those holes at a specific speed, then you may find that the holes and/or the
arbor itself have worn -slightly- out of round. It would be hard to see, but if
the holes are normally for example .003” larger than the arbor, and now they’re
subtly egg-shaped and .005” in only one direction (out of round), then any
vibration whatsoever at the contact point (leather pad & flywheel) would cause
the arbor to vibrate in a unidirectional orientation, within its holes.
Wear would naturally occur on the associated parts only in the specific areas
of the arbors and holes that would be in constant contact when the machine is
set to run the correct speed, assuming that for most of the machine’s life it
was set to run at the correct speed, thus causing the wear at those limited
locations.
Keep in mind that even though the holes may appear to be a very close fit to
the arbors and thus touching them all around, the reality is that the contact
point between arbors and holes can only be over a specific small part of the ID
of the holes. For them to touch all around, the holes would have to be the same
size as the arbors, which would be a zero clearance, zero tolerance hole —
almost a press fit. The holes are larger than the arbors to provide freedom of
movement. Freedom of movement when constrained within certain engineered limits
is a good thing. Freedom of movement outside of those limits can either cause
problems or allow problems originating elsewhere to come to light.
Chances are that this machine has lived with a slight vibration at the governor
most of its working life, and this has in effect caused more wear to occur in a
very narrow and specific way. ANY vibration will lead to greater wear on
associated parts than the same arrangement with no discernible vibration. There
are multiple causes of vibration at the governor and pad, which could be
explored as part of another investigation.
To correct your immediate problem, and assuming nothing is done to correct the
vibration, then you might approach where the visible effects of this vibration
manifest, by sourcing a different indicator frame and arbor. While the
indicator needle serves to show the turntable speed, it also makes a marvelous
vibration indicator!
All of the above supposition is based on imagining all of the related parts
working in the mind, in the absence of physical inspection, and therefore
remains in the realm of theory. To check to see if the arbor holes are worn
slightly out of round, a clockmaker would insert a "smoothing broach" into the
holes (tapered steel tool of round cross section), until the tool gently seats
in the hole. Then they would hold the indicator frame and broach up to a light
source and hunt for a slim, crescent moon shaped irregularity showing at the
hole, around the broach. If present, this would prove a worn hole. The test
would be repeated for each hole. Wear would likely be greater at the hole
closest to the steel arm and hairspring that supports the pad, than at the
opposite end of the same arbor. You should also inspect the arbors themselves,
where they touch the holes. I would expect to see that a narrow portion of
their OD would at minimum appear