"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:
"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:
"Jim Wilkins" wrote in message
...
On Nov 24, 1:04 pm, "Ed Huntress" wrote:
"Jim Wilkins" wrote in message
...
I just checked my two best sources, Holtzapffel #2 and Oscar
Perrigo's
1916 "Lathe Design", ...
...
As I think about this, I'm remembering what I thought about it at
the
time,
30 years ago. I believed then that the issue was the difficulty,
without
planers, mills, or big surface grinders, of getting the four planes
of
a
pair of V-ways coordinated for straight and smooth travel. One way
to
interpret this is that you can adjust the single plane of the flat
way
a
lot
easier than the pair of planes you have with a second V. So to say
that
it
was simpler to correct accuracy with the V-and-flat could just mean
that;
if
the ways are hand-finished, you're correcting accuracy, and
V-and-flat
is
a
lot easier to correct than two V's.
Maybe. g
--
Ed Huntress
If I read Holtzapffel correctly the two sides of inverted vee ways
were at first made separately, joined and aligned afterwards to fit
the fixed and moving poppit heads (headstock and tailstock to us).
Yeah. In the very beginning of modern lathes, the V-and-flat
combinations
were assembled the same way. The first screw-cutting lathes had wood
beds
with bolted-on iron ways, IIRC. (I'm doing this from memory; don't
bite
me.
g)
"This slight width of base does not afford sufficient lateral support
to the heads, which with only moderate force in turning are liable to
vibration; while exact parallelism of the two angular edged bars is
also necessary. Improvement in stability was sought by making one
side
of the bearers flat and broad, fig. 72, with a corresponding flat on
the underside of the lathe heads; retaining one angular side, to give
the direction or common axis. This arrangement also facilitated the
construction, as the parallelism of the two bars was no longer
essential,..."
Right. That sounds familiar.
Fig 72 shows one flat and one inverted vee way on a cast iron bed.
The difficulties of the early machine builders that Holtzapffel
recorded aren't that much different from those of a homebrew machine
tool maker today, except that we can buy ground drill rod and flat
stock and they could hire cheap child labor for tedious hand fitting.
Right. Maybe you've peeked at my ideas for a ferrocement lathe with
steel
ways. d8-)
(Having finished reading Naaman's _Ferrocement & Laminated
Cementitious
Composites_, I'm less enthusiastic about that construction.)
Significant work has been done on concrete-filled fabricated metal
frames for precision machine tools.
The place to start is MIT professor Alexander H. Slocum
(http://meche.mit.edu:16080/people/index.html?id=80). A good
discussion
and many references may be found in his book "Precision Machine
Design".
Joe Gwinn
Thanks, Joe. I talked to Slocum not too long ago -- maybe a year or
two --
and someone here brought up his book before (maybe you?)
Very likely. I recall posting the reference before. Yes. The thread
was "Epoxy grainite build yer own machine frame" in March 2009.
-- it's on my list
of things to get for my library. I had a copy for a couple of weeks on an
interlibrary loan, and I read what he had to say about long-term
stability
and so on, which helped a lot.
I bet you can borrow it again.
I can, but I'd like to have it. It ain't cheap, and my list of desired books
is long.
I should point out that I was studying and writing about concrete- and
polymer/aggregate-base machines around 30 years ago, and I've visited
manufacturers of them in the US and in France and Italy, mostly around
that
time but as recently as seven years ago. I'm familiar in general with the
various approaches, the materials, and the design philosophies. But I've
never claimed expertise; it was more on the level of a good journalistic
exercise.
The commercial applications are one thing. This wild hair I've been
chasing
has more to do with exploring strategies for building at home, at very
low
cost, and to see what can be done with a material that's intrigued me for
40
years -- ferrocement. The obvious approach to building a machine like
this
would be either a fairly massive, fiber-filled casting or a
screw-tensioned,
post-tensioned cast structure. The former is my next area of study and
it's
a big one. The latter is something I've abandoned because of problems
with
long-term stability and engineering that's trickier than it looks. But it
would be a great way to go, if it wasn't for the stability issue.
Slocum runs hot and cold about concrete versus lots of cast iron, but he
isn't trying for infinite life either.
If it's cheap enough, people will live with something that doesn't last
100 years. Most HSMers won't last that long.
Right. g I don't think that the limitation is the life of the machine, but
rather of dealing with growth or shrinkage over a period of years. There are
ways around it. I just haven't thought it through. But, again, we're talking
about the difference between, say, a South Bend and a Hardinge. You can
build something that will fit into the SB category. But concrete is not the
stuff to use if you want Hardinge.
Then again, if you want Hardinge, you'd better mortgage the house and buy
Hardinge. g I own a South Bend, and it's all I could want for my hobby
machining.
I'm convinced that a concrete machine can deliver ordinary lathe accuracy
and could be perfectly suitable for the hobbyist. Problems crop up when
you
go for high-end, toolroom-grade accuracy and long-term stability. The
polymers are better for that but they're *very* expensive, and defeat the
basic goals I'm starting with.
Anyway, thanks for the tip. I would love to go for this in a big way, but
it
isn't in the cards right now. I'm engaged in conversation with Dr. Senft
on
Stirling engine lubrication, and the outcome is going to eat me up for
months to come. d8-)
Is there a good solution? I recall that this was one of the big issues.
You'll have to read my article, should I succeed in gathering enough info to
write one. There are solutions that work; different solutions for different
realms, from low-temperature-differential types to fractional-horsepower
mule motors, up to 100-hp-plus automotive engines. The solutions are all
different. Senft has put me on to a guy who apparently is one of the world's
experts on automotive Stirlings, and who knows the big-time lubrication
solutions, both for kinematic engines and for free-piston types. I haven't
talked to him yet. I'm looking forward to doing so.
BTW, if anyone is interested in the approach you mention, which is
various
forms of making a light welded or bolted steel structure and filling it
with
concrete, it has possibilities for the home builder. But it's a lot
trickier
than you might think. There are bonding issues and problems with the
different coefficients of expansion between steel and various...er,
"fillings," plus retained-stress issues with the steel itself. And
concrete
grows (or shrinks; I forget) for decades after it's cast. It's not enough
to
matter in building structures but it can be an issue when you're dealing
with thousandths of an inch.
Yes. Slocum goes over this.
The issue is to replicate the static and dynamic stiffness of cast iron
more cheaply (or with less weight) than cast iron.
A fabricated box-beam frame has plenty of static stiffness, but has far
too little damping, so the dynamic stiffness is scant. Filling the
frame is an attempt to sharply increase the damping and thus dynamic
stiffness, to make the chatter threshold more remote.
Joe Gwinn
Yes. But a torsion box (box-beam) made of concrete has lots of natural
damping. As you say, the filled-base commercial machines have been attempts
to build in damping with low shipping weight (and lower costs) for the
machines.
It's a very tricky engineering problem, but it doesn't require a lot of
knowledge or heavy math. It just requires a good feel for materials and
structures, combined with a lot of patient thought and analysis.
--
Ed Huntress