David R.Birch wrote:
Steve W. wrote:

Most shops are not set up to hob gears, for one thing. Even if they were
to make one, including tooling, setup and cutting, you could easily have
20 hours @ $50+ an hour, and while that work is being done, those
workers aren't doing what they were hired to do.

Not applicable in a home shop in many cases we are doing it for fun.

True, when I contacted the surly dude who owns the rights for
Benchmaster parts, his price quotes suggested he had no stock and made
parts one at a time. I built my own X-axis lead screw instead. My
Benchmaster came with Y and Z lead screws, but a rack and pinion setup
for X.

Personally I think this is one of the reasons why the US is losing the
global economy battle. Don't repair that machine that you have, scrap it
and buy a newer one.

The window where a machine has paid for itself and is still cost
effective to run is getting narrower. We have 2 Mazak LASERs at work
with the same 4kW cutting ability. The older one, about 5 years old, has
a CNC probably designed in the early '80's, only data input is a RS232
serial port. The newer one runs Linux and has network inputs, USB 2.0
ports and....a RS232 serial port. The old one we use for long running
jobs, the new one serves the customers who want one part yesterday.


If all they have is the worn out part to work from, they have to reverse
engineer it and guesstimate the pre-wornout sizes & tolerances plus
guess what material will work.

Not really, parts duplication would be easy. You already know what
material to make it out of, you have a sample in hand.

That part in hand is made of steel. 1020 or 8620? What kind of heat treat?

Easy enough to do a test on a piece of it and tell the composition. Or
use a steel currently used for a similar use.
Heat treat to current spec for a gear in similar use. No real need to
use the original factory spec if you know a better one.


Look at the books and you would know what the original specs should be.

Where are these books with specs for parts made 60+ years ago by
companies long out of business?

Depends on the item. However if it is a gear the simple solution is to
use published specifications of common gears of the time. The harder
spec would be to look at the back of the teeth and get the information
from there. In the event that the wear is bad enough to nullify that you
will likely be making a few others to match existing parts and bring the
machine into spec. in that case you simply need to use the original for
it's diameter and tooth count. Then cut a new gear for each worn position.




I've done work like this and you really have to need THAT machine up and
running to make it cost effective.

The only parts in a machine that I can think of which are likely out of
the abilities of most shops would be semiconductors and other
electronic
components.
Parts are made in production lots of 50, 100, 1,000,000, whatever. The
most expensive is the first.

David

True to some extent.

True for every job our company quotes. True for every part I've made in
my basement.

David

However your not making readily found parts in your basement are you? If
you are then why?

I look at it this way, just because I can go to the shelf and pick up a
1/4-20 X 1" cap screw, doesn't mean I shouldn't know how to make them.
I've visited countries where "automation" advances are made by
installing more people with files and grindstones! Ask them about CNC
machines and you get blank stares.

--
Steve W.