On 2 Sep 2010 04:04:34 GMT, "DoN. Nichols"
wrote:

On 2010-09-01, Phred wrote:
On Tue, 31 Aug 2010 21:22:36 -0500, Sunworshipper
scribed:

On Tue, 31 Aug 2010 20:03:32 -0400, Phred
wrote:

On Tue, 31 Aug 2010 19:02:38 -0400, Bob Engelhardt
scribed:

Phred wrote:

http://www.metalworking.com/dropbox/B-17_Part.pdf

[ ... ]

I certainly don't have the time and pulling out of that groove is over
my capabilities.

The choices there would be either to have a CNC lathe to do that
work with, or to hand turn the spindle with a crank on the end (as some
people do when cutting non-native threads on a lathe (metric threads on
an inch lathe, or inch threads on a metric lathe). The reason for this
is that the threading dial does not work in a converted mode, so you
can't disengage the half-nuts while threading in the "wrong" mode.
Couple this with the radius needed in the bottom of the groove and that
is going to be a muscle workout.

Having the groove end in holes reminds me of how old threading
was sometimes done -- with the thread ending in a hole instead of a
runout groove. I suspect that was a lot easier to do with overhead
lineshafts and flat belt drive which you could make slip at your desire.

The end grooves, with their radiused bottoms would call for a
fairly heavy lathe too -- or ball end mills and a circular table.

But, even looking at it sideways, I wouldn't put in
the two circular grooves on the sides to reduce weight. The most of
the stress if it binds will be on the outsides of the cable rapped
around. What's a cubic inch of aluminum for a proven war bird that
will never carry again what it was designed for.

I think that the grooves are to provide access to the ends of
the cable for tying off. As I see it, the cable starts through the drum
with a knot in the end groove, takes several turns (probably four), then
goes off to the drum on the control at the other end of the aircraft
around its drum, and returns to go a final four or four and a half turns
around the original drum before being tied off in the end groove again.

And actually -- it looks more like a hole drilled down to gain
access to the end of the cable, not a turned groove. But why the
radiused bottom?


I didn't see those holes, it is uncomfortable to view. Can't seem to
wrap my head around it, and no A&P, but I have worked on a number of
WWII planes. Typically the cable just wraps around and goes right back
in a long loop, that's why it turns so many times, to multiply the
friction. The radiused bottom could be for the "friction brake" (see
my other post), but there is another hole from left to right radius,
plus if the ends of the cables are there also???

Had to do an experiment in the shop to confirm it, what it could be is
each end of the cable are terminated in each of those holes and the
"two" cables wrapped around to meet towards the middle. Then the loop
end is wrapped around the slave drum in the tail. So essentially one
is being reeled in while the other is let out at the control. If that
is right the one in the tail won't be the same.

SW



Like the other post, probably trim, the main yoke would be 5-6" dia.
not 2.5"

SW

Trim control sounds about right... Much less turn than rudder, and the
dimensions sound about right for tight control over a smaller drum
size and force, more precise control anyway... Or is the bigger the
drum, the more control you have?

The bigger the drum under your hands, the more motion at the far
end for a given motion at the near end -- but the more force is needed
to accomplish that motion.

For trim adjustments, I suspect that there is a worm gear
connected to a hand crank to rotate the drum by the keyway inside it
instead of a lever to move the surface quickly.

I was never a physics or calc guy...
I know the bigger a micrometer head, the more precise it is, probably
just cause there are just bigger graduations, more widely spaced,
thus, easier to read and interpret and "split the lines"...

Right. (And, of course, for that to be meaningful, the more
precise the spindle thread pitch has to be. :-)

Anyway - it looks like a job beyond my current capabilities,
especially making the part number marking according to spec. I suspect
that they were pressed in by a special punch, not hand punched one digit
at a time. Presuming that it needs to be as close to the original as
possible, even the size of the punches matters as well as the angular
spacing.

Good Luck,
DoN.