Its Bruce Johnson, and I too considered the anti-rotation problem for a
round column mill about five months ago soon after acquiring an RF-31, my
first mill. I was advised by a local career machinist of 35 plus years of
experience, to stay away from the round column due to the lack of
repeatablity to index a center as you move the head up and down the column
and he suggested I look into a small knee mill, maybe a Millrite or
Clausing. I talked to a few more people, and a Rockwell was added to my
prospective list. Then I started looking at the spindles and decided the
preponderance of tooling available, (I had zip) was much greater for the R8
spindle and the power of the Taiwan made machines was about double, and yet
the prices were about one half on the used market. I then gave some thought
as to what I might be making on the machine of my choice and what accuracy I
would need. The answer was that as a newbie, accuracy was down on the list,
and most things that I machine on my lathe that I have owned for 30 years
are just repairs for other tools, simple turnings and bushings, etc. On the
other hand, a resonable counter bore on a mill with decent concenticity was
desirable. The amount of time spent on a machine was also a factor in the
trade-offs. I kept asking myself, will I really need this parameter or
feature in the near future.
I decided to look into pinning the gear rack onto the column as a start,
because I had heard just about what you heard, not to many details, and no
data as to the end result from people who had tried this, and I figured that
I could always move the work relative to the head to get bye for my needs.
But in the back of my mind, I had the capability of the traditional flexible
head movement of the smaller American made machines still imprinted. The
result was I decided to try and come up with a modification that would let
me keep my center for the majority of the time, but for that rare occation
when I needed to rotate the head around the column axis, I wanted something
that was semi-permanent and the roll pins to attach the rack to the column
were not the solution for a couple of reasons: The first was difficulty
getting them in and out, and secondly, I was not getting the rigidity of the
rack against the column to maintain linearity and parallelism with the
column axis. I did try several different combinations of pins with a
countersunk screws on the top and bottom of the rack close to the teeth.
This was an improvement but I still had to remove three pins equally spaced
between the top and bottom screws.
I then had a breakthrough and tried allen drive set screws where the three
pins had been and clamped the rack tight while tapping through BOTH rack and
column. I left a very small amount of material in the column thickness so
the screws would bottom out and not drop into the column ( my column is
still empty but I plan to fill to increase rigidity). The set screws have
to be the right length, so that the top of the screws are just below the
bottom of the gear teeth, and centered on the width of the rack. I had to
do a little careful fillng on the head of one screw to match the gear tooth
profile. I had previously measured the width of the rack at several points
and as I recall, it was less than .001 inch over the length of the rack and
straight using my best straightedge. Now I have 5 screws to pull using the
same allen-T wrench and the head now swivels as it did before. You do have
to crank the head up and down during the process to get at the center three
The next task was to design a better method for the head to " track" the
gear rack. I noticed two important things as I looked at the dimensions of
the rack and the way the head moves against the rack. The sides of the
rack are not flat and the head travels against one side of the rack due to
gravity and the head weight against the sloped gear teeth. The notch at the
bottom of the head which is considerably wider than the rack is what makes
contact with the rack side on my machine. I essentially decided to close
this gap using a copper bar that is slightly crowned at the rack contact
point (side) which decreases friction and it is mounted on the adjustable
ring on the top of the head. The bar is slotted down the center allowing
for contact adjusting pressure using two allen head cap screws and washers.
I chose the copper since I did not want to gall the gear rack sides and I
had it "on hand"-Any semi-soft metal would work most likely. Rack and bar
were lubed on contact sides and on rack teeth.
Your roller idea on a flat ground surface may be a better method over the
long haul and should take out some column to rack error for the Z-axis but
my method is a cheaper fix, another example of cost vs. accuracy. Speaking
of accuracy, the bad news is I have yet to take the final data as personal
problems and bad weather have stalled my hobbie time. When I finish, I will
release some pictures to the group and may in the future make some direct
comparisons between the RF and a Clausing 8520 which I have access to. Oh,
definitely torque the head bolts using a torque wrench and use the same
sequence each time. I would be interested to see the variation between
machines in regard to the bolts and any differences due to the roller
contact. I would appreciate an update on your progress if you decide to go
ahead, and remember, where there is a will-there is a way! Best regards,
.original Message -----
From: "Martin Kennedy" <martin@...
Sent: Friday, December 31, 2010 03:33 PM
Subject: Re: [mill_drill] Re: Round Column Mill Anti-rotation Modification
> Thanks to all for the suggestions!
> I have not heard much to give me hope that I can successfully modify the
> mill as I described. I was never positive that it'd work, and everything
> I've read makes me even less certain that it could work.
> Of course, as much as I'd like it, buying a new $10k CNC mill to replace
> one I already have is not much of an option! It would certainly fix one
> problem (the mill) and lead to another (the wife).
> I've tried the laser trick and a string on the far (20' away) wall in
> the past without much success. I don't think it's very accurate, and
> it's definitely not easy to do - I can't see that far to see if the
> laser is on the string, and end up running back and forth as I torque
> the head bolts. The diameter of the laser is relatively large at that
> distance, and that's where the inaccuracy comes in.
> I'm aware of the limitation of my round column mill. Thoroughly. But you
> have to work with what you have. I plan my operations, but setting up
> the mill for different operations like a chucked reamer and then milling
> at the maximum extent of the quill is not a good idea. Sometimes you
> just have to move the head between operations.
> I do like the suggestion of the camera. I had not heard of this before.
> I found the web page that describes it and allows download of the
> software "CentreCam" http://www.miketreth.mistral.co.uk/centrecam.htm
> I'm going to build one and see how it works. I just need to find an
> appropriate webcam to buy. I think that building the holder would be
> simpler than modifying heavy mill iron! Accuracy for the camera is
> claimed at 1 thousandth or so, which is plenty close. There's already a
> computer and monitor sitting right next to the mill to drive the CNC, so
> I would not need to buy much stuff.
> On 12/31/2010 9:15 AM, draggingsteel wrote:
>> I cant imagine the column ever being repeatable. Loosening and
>> tightening the bolts will make a considerable change every time. The
>> idea on pinning the rack to the column might help but even if the
>> sides were ground and you ran guides along them, the loosening the
>> head would negate all efforts. Do you plan on cncing the head or the
>> quill? The reason you dont see it done much is probably due to the
>> extra complications. If you used Arnie's/Zarzul/Wildhorse Innovations
>> camera that mounts in the spindle, you could form a reference point on
>> the table. Find and center your reference point, lock the head and
>> switch to the tool of choice. I have seen lasers used aiming at a
>> place on the other side of a room. That might get you close but that
>> would be about it.
>> --- In firstname.lastname@example.org
>> <mailto:mill_drill%40yahoogroups.com>, Martin Kennedy <martin@...> wrote:
>> > I'd like to modify my round column CNC mill so that the head will not
>> > rotate and lose my center when raising or lowering the head. I looked
>> > on-line for modifications, and although I found people that mentioned
>> > them, no one had any details on how they were done or how well they
>> > I'm thinking about pinning the rack on the round column, and then using
>> > a surface grinder to make the rack very flat on the sides and parallel
>> > to the column. I'd then modify the head to include some ball bearing
>> > rollers that press up against the sides of the rack, similar to the
>> > arrangement on a band saw. (These are on an eccentric bolt so that the
>> > rollers can be snugged up against the saw.)
>> > I think that one of the potential problems is around the two bolts on
>> > the back of the head that clamp onto the round column. If I leave them
>> > slightly loose so that I can move the head up and down, then there will
>> > be play in the head. If I loosen them, move the head, and then tighten
>> > them, they may cause some small movement of the head even with the
>> > modifications above.
>> > Has anyone done a modification on a round column mill to reduce or
>> > eliminate head rotation, and how has it worked?
> Yahoo! Groups Links