replication, or big machines from small ones was Re: Soliciting Feedback:OS

Ron,

Snips, inserts follow...

>In CCED, R Rogers <rogersmach@...> wrote:
>Just some input here and it's probably 180 out :-).

<snip>

>The old adage that "the only machine that can reproduce itself is
>the engine lathe" (or Bridgeport depending on who's saying it) is an
>interesting claim but unfortunately with little merit.

Really?

>One could produce a new milling 9x42 table on a 9x 42 mill?

Definitely. Leadscrews too...

>Machine the dovetail ways on the column?

I might modify the design to have two component parts to make it
easier; but again, it's surely possible.

I take it you've never set up work larger than a machine's basic
envelope? It's a pretty standard thing to do in many shops.
(BTW, CNC is pretty helpful in this because you can set the work up
and THEN rotate the machine axes to match, whereas we used to have to
dial in the work TO the machine instead.)

There are even commercially available "sled" type indexing tables to
do this type of work more easily... And again, to bring this on
topic, such sleds and indexing techniques are VERY much used and
useful for folks with hobby CNC machines; whether router or mill.

>So whatever the envelope of the machine is, so will be the
>limitation of what it produces.

While this sounds logical; and is in some ways true... It is ALSO
true that the person or shop who can only make things the size of
their machine's basic envelope is not really trying very hard.

You can most certainly set up a BP to make the necessary cuts to
replicate its table; then you'll hand scrape the ways to complete the
task(as was done on the original!) Point is, the accuracy required of
the milling operations involved is attainable using multiple
setups.

>No metal working machine can reproduce itself.

IMO, a false statement. (I guess if you're saying that the reproduced
machine has to have the exact same construction and materials as the
original it may then be true.)
But I'm pretty sure most folks saying that if you have a 9x42 mill
you can make a 9x42 mill mean that you can make a machine with all
the capability of the original.

As a final "proof" and "truth", one simply has to go to CNCzone and
look at one of the many 2nd or 3rd generation "builds" by folks
there. Joe2000 is one VERY good example; his 2nd machine is larger
AND more rigid and accurate than the first and it WAS built using his
1st machine very nearly exclusively... (AND he's made a table for his
new router that is bigger than the one on his old router USING the
old router. As I said above, if you can't figure out ways to do that,
you're not trying very hard!)

Successive approximation and iterations solve more than math and
software problems. Bigger machines have been made from smaller
machines since machines have been made.

Ballendo

P.S. A shop I worked with last year had recently completed an
aerospace part--with aerospace tolerances--that was 6 times the VMC's
longest axis capability. They had cut holes in the VMC enclosure
sides to enable this "too big" part to be made. LOTS of setup time,
LOTS of checking and re-checking; but they got the job done and won
over their customer "for life" by doing so...