Hi All:

I just finished rebuilding my Atlas HL-54. I documented the
whole process, and I'll post a web page in a few weeks. Now it's
time to move on to my KTM-380 mill.

I removed the table, and it looks like the ways on the knee
are worn where most of the work was done. There is also a deep
gouge where it looks like a chip worked it way in and got dragged
back and forth. What kind of shop should I look for to remachine
my ways, and any idea how much it will cost me?

--
Andrew

Andrew Tubbiolo writes:

I removed the table, and it looks like the ways on the knee
are worn where most of the work was done. There is also a deep
gouge where it looks like a chip worked it way in and got dragged
back and forth. What kind of shop should I look for to remachine
my ways, and any idea how much it will cost me?

Sounds exactly like my Bridgeport. This is the typical wear pattern,
progressively more toward the middle, with a slant down to the outside.
I had some scoring ("gouge") like you mention, and in one case I
actually found the cause to be a speck of carbide or diamond embedded in
the mating cast iron surface, ouch!

I improvised some effective methods of doing the machining of the
horizontal knee ways, get this, on the machine itself!

I "self-machined" the horizontal knee ways with a dovetail bit, rotating
the turret and sliding the ram in small increments along the surface.
Each increment required lowering the knee,
unlocking/rotating/extending/locking the ram, and applying power while
raising the knee. The dovetail bit I made by brazing carbide inserts
onto a frustrum of mild steel, such that they would scrape flat 2.25" OD
/ 1.25" ID donuts. (A regular dovetail bit doesn't do this, since it
has a slight taper up towards its center, like an endmill.)

Properly overlapped, this yields a surface flat and automatically true
to the ram's datum alignment, with a "noise" variance of only about
0.001" to be tidied up by scraping.

The turret swing and ram extension doesn't quite reach the ends of the
knee ways, so there are a few inches at each end to be scraped down more
or less by hand. After scraping off the entire 0.015" by hand on one of
these 4 areas, which was exhausting, I made another tool, a "precision
plow", which consisted of a 1x6 in cast-iron shoe with a scraped-flat
bottom, with a length of 1/2" round bar attached, and a carbide
threading insert on the end of the bar. This I could adjust on a
surface plate to have the tip of the threading insert precisely 0.003"
or so above the horizontal plane of the "sole" of the shoe. Then by
skating the shoe on the already self-milled areas by hand, I could plow
a network of grooves into the unreachable unmilled areas. Once so
grooved, I used a die grinder to quickly remove the bulk of the material
(about 0.012 of the 0.015 thickness), using the grooves to monitor and
control the depth of hand grinding. The final 0.003 or so I hand
scraped.

The grooving is critical. A die grinder removes a lot of material fast,
but it is uncontrolled and inherently imprecise, and therefore useless
on its own. The grooves are a kind of "topo map" to measure how far
down you have ground. I chose the 0.003" as a remnant because it seemed
like the worst-case error I might make between the plow and the grinder.

Making the dovetail bit and the plow took longer than applying them.

I'm not sure exactly how a knee is machined in the more conventional
manner. It would take a very big dovetail bit on a very big machine to
handle the size of the knee.

In theory one can scrape it all off by hand to return to alignment, but
I learned that hogging off material with a scraper is days of hard work.
Much better to machine off the bulk and finish with hand scraping.

I am in miami fl and i grind knee ways mostly bridgeport but i will do any
knee. what will get you is shipping i usually charge $450 but i could work
something out if you let me now that you are from the group. email me because i
do not read all the post

Andrew Tubbiolo wrote:
Hi All:

I just finished rebuilding my Atlas HL-54. I documented the
whole process, and I'll post a web page in a few weeks. Now it's
time to move on to my KTM-380 mill.

I removed the table, and it looks like the ways on the knee
are worn where most of the work was done. There is also a deep
gouge where it looks like a chip worked it way in and got dragged
back and forth. What kind of shop should I look for to remachine
my ways, and any idea how much it will cost me?


Well, you can do it yourself with a little bit of gear, unless the ways
are hardened or Chromed. Then you really will need to have it done
for you. The problem is the base of the machine is too big to mount
on just about any grinder. There are things called portable bedway
grinders, that are essentially a pair of lightweight rails that you
mount to the machine part, and it allows a grinding head to be
run up and down the ways. Some people will come to your location
and do this, but it is going to be WAY expensive to have this done.

First, how bad IS the wear? Attach a dial indicator or dial test
indicator to the spindle, and put a flat, vertical surface on the
table. A precision square can be clamped to the table, or the
fixed vise jaw can be used. Put the table at the elevation of
the most wear, and get the indicator reading on the vertical
surface. Raise and lower the table while watching the indicator,
and see if you get change on the indicator. If the indicator
only moves .001" or so, I wouldn't worry about it at this point.
Get a set of way covers to keep the chips out of the way wipers,
and keep up the lubing of the ways. I have about this much wear
on my ancient machine. I always move the table UP last, before
machining, so that the knee screw is pushing the knee up against
the base ways.

If the top of the knee is the big problem, you can make sure the
gibs allow the saddle to move all the way forward and back.
Then, rig a dial indicator to measure rocking of the saddle,
and use the table as a lever to try rocking the saddle. See how
much looser it is in the worn area. This is differential wear,
and you really do need to get it reground. Grinding the saddle is
not a big deal, but again, the top of the knee is pretty tough,
due to the size. You also want to maintain the right angle
between the back and top of the knee.

I have no idea what it will cost to have this done, I think it is
extremely area-dependent.

Jon

Jon Elson writes:

If the top of the knee is the big problem, you can make sure the
gibs allow the saddle to move all the way forward and back.
Then, rig a dial indicator to measure rocking of the saddle,
and use the table as a lever to try rocking the saddle. See how
much looser it is in the worn area.

The "shake" test is OK as far as it goes, but the knee-saddle-table system
involves no less than 24 different precision bearing surfaces. A simple
straight edge and feeler gages will map some wear, although jigs and other
standards are needed for complete characterization. For any detailed
measurements you must remove the table and saddle. Mating surfaces with
complementary concave/convex wear maintain rigidity and pass the "shake"
test while degrading positional precision.

Richard J Kinch wrote:
Jon Elson writes:


If the top of the knee is the big problem, you can make sure the
gibs allow the saddle to move all the way forward and back.
Then, rig a dial indicator to measure rocking of the saddle,
and use the table as a lever to try rocking the saddle. See how
much looser it is in the worn area.


The "shake" test is OK as far as it goes, but the knee-saddle-table system
involves no less than 24 different precision bearing surfaces. A simple
straight edge and feeler gages will map some wear, although jigs and other
standards are needed for complete characterization. For any detailed
measurements you must remove the table and saddle. Mating surfaces with
complementary concave/convex wear maintain rigidity and pass the "shake"
test while degrading positional precision.

While it is possible for the surfaces to wear into mating curves, it
seems that in most cases of one long and one short piece (eg. a long
table on a short saddle) the wear is distributed such that the curves
are very unlikely to actually match. That is certainly the case on
my 1938 bridgeport mill. The underside of the table is concave,
the top of the saddle appears to be a convex curve of shorter radius,
allowing the table to rock slightly at any position.

The Y ways look like they were used as an anvil, apparently by the
daily dropping of wrenches, etc. on them. One of these days I
really have to tear the whole machine down and rescrape it, but I
can't afford to have the machine down for the 6 months it would
take me to do it.

Jon

Hey Folks:

Thanks for the feedback so far. I'm still considering what to do about the situation.
Starting with how to measure the issue.

--
Andrew

I just listed both of my metal planers on ebay. Either one will do an
excellent job of recutting bedways. One even has a fixture to cut
jibs from scratch. It is probably cheaper to get somebody to do it
for you, however.
Paul

6e70 writes:

I just listed both of my metal planers on ebay.

URL?

Richard, I'm too ignorant to URL. You could search by seller (papapault). Thanks,
Paul