Keeping the X on a round column bench mill

Canuck75

Active User
H-M Supporter Gold Member
Joined
Jan 1, 2014
Messages
233
I have posted some information on this forum before about my machinery and have made many modifications to the mill, but this article is specifically about addressing the problem of keeping the X on a round column bench mill when changing the head position.

My particular "King" mill has a 19.2" X, 8" Y and 5 3/4" quill travel plus the head travel on the column. This capacity meets all my hobby machining needs. I dreamt of having a universal turret/ram/ knee mill but space and money dictated my choice. I saw this used bench mill on Kijiji and within driving distance. Mill looked a little scuzzy so did a full teardown, cleanup, repaint. Also started plans to make improvements.

During operations I find I often raise and lower the head to keep everything as short as possible for rigidity sake and to get room for the boring head, chuck and drill etc. Redialing or edge finding after each head change was required. The question of keeping an accurate X during these changes kept bugging me. Other owners of this style mill have posted opinions on this subject indicating we all want to make one of these basic mills do more than they were originally designed to do - in other words make a silk purse out of a sows ear - well why not! I went through a lot of "what ifs". I established several parameters relative to what I needed my mill to do. First: I don't do work big enough that I need to rotate the head to reach far corners of a work piece, or, even to get it out of the way; Second, if you do rotate the head you relatively loose some of your Y travel range not having a "ram" to compensate (even in a commercial machine shop I seldom had to rotate the head and when I did I always needed to use the "ram" to offset the change in Y reach): and, Third, I do have to raise and lower the head often during machining operations thus a desire to control the X accuracy setting if possible.

Having accepted this, it was just a matter of what way to do it. Others have made made very good stand off frames to control the X which also retain a head swing capability but I'm not sure what guarantee of accuracy you will have when you swing the head back and clamp everything up again. Based on my personal "parameters" I chose the non rotating head solution. I know others have thought of fixing the gear rack to the column and using guide blocks (I presume) on the head, so thought I would give it a try. It could always be undone with no harm done. Also the whole mod is basically invisible so the aesthetics of the machine are not altered which is a personal preference.

The attached series of photos shows the results of my effort. Subsequent up and down tests show an accuracy of .001". If on occasion I am really paranoid, I can always edge find or dial indicate for that particular operation.

First photo shows the purchased condition just for interest. Photo 2 - pinning the rack moving up the column with the assist of a dial indicator as in photos 3 and 4. Photos 5 and 6 show the top and bottom guide blocks bolted to the head - this was done with the head on the column, in position, and using the pre drilled holes in the guide blocks to drill the head and then tap. The last 3 photos show the test results. First with the head at the bottom and after a X and Y edge find setting the DRO to zero. The second to last photo is with the head raised 5.5",quill extended, and table brought back for X and Y edge find to see how it would compare to the DRO - .001", and last photo is with quill retracted and head back down and another edge find comparison to the DRO - .001".

I,ve added one photo of the top guide block showing the slightly modified middle pulley bracket to accommodate the movement of the bracket when loosening and tightening the belts.

IMG_6298.jpg

I welcome any comments.

Shop 228.jpgIMG_5659.jpgIMG_5661.jpgIMG_5665.jpgIMG_5684.jpgIMG_5716.jpgIMG_5689.jpgIMG_5690.jpgIMG_5691.jpg




IMG_5659.jpg Shop 228.jpg IMG_5661.jpg IMG_5665.jpg IMG_5684.jpg IMG_5716.jpg IMG_5689.jpg IMG_5690.jpg IMG_5691.jpg IMG_6298.jpg
 
Last edited:
VERY nice! Question...you pinned the rack in the middle? This doesn't interrupt the operation up and own, presuming the bolt head is sufficiently countersunk....more pics, if you have them, would be great!
 
That looks very interesting. I've been giving some thought to trying something like that with my RF31 - now you have revived my interest in doing it. Thanks! :))
 
VERY nice! Question...you pinned the rack in the middle? This doesn't interrupt the operation up and own, presuming the bolt head is sufficiently countersunk....more pics, if you have them, would be great!

Yes, x4. I figured where I wanted the roll pins to go and pre drilled the holes in the bottom of the gear teeth on the drill press first, thus keeping their size and being square. The bottom 1/4"-20 bolt hole had already been drilled and tapped by the previous owner which put the the head 90 degrees to the table which is what I wanted anyways. Yes the heads are flush. The top 1/4"-20 hole was pre drilled at tap drill size and countersunk part way. Put the rack in place and tightened the bottom FSHCS in with locktite. I advanced up the column using the dial indicator to keep a zero on the rack at each pin position, temporarily clamping it, drilling thru the column, installing a roll pin, and advancing again. When I got to the top, a roll pin next to the c-sink hole was driven home first, then the FSHCS hole was drilled, tapped, and c-sunk to be flush. I did this to ensure that the FSHCS would not draw the rack off even a thou. The FSHCS was the put in with locktite.

Dialling the rack as I went gives a .000" error the full length of the column.
 
VERY nice! Question...you pinned the rack in the middle? This doesn't interrupt the operation up and own, presuming the bolt head is sufficiently countersunk....more pics, if you have them, would be great!


Thanks for your comments. Rather than trying to lump all my efforts into one article with the limit of ten pics per, I will break it up to cover each mod. It's the photos which really help one understand. The next one will cover gib locks.
 
Makes sense to do it this way , arks of a circle cut in different degrees. Should even make it shift able to do circle cutting on larger work pieces. Could add a sprocket and chain drive to do it . Like the up and down only around . Think this one in mine when I do it.
 
Is the sides of the rack straight and smooth enough for the accuracy?

How tight are the guide blocks fit on the rack to eliminate any error?

I thought about doing something like that to that style mill in my friend's shop, but these questions have stopped me.
 
I have had a round column mill for forty years and had to deal with the problem of losing registration whenever the head was moved.

The issue with registering at the column is the registration point is only about 2-2.5" from the column center whereas the spindle is about 4-5 times further. This means that any error in registration will be magnified 4 - 5 x. There will necessarily need to be some clearance in order for free movement to occur. Clearance would need to be less than .0002" to insure .oo1" accuracy at the spindle.

Another issue with using the rack as a registration device is that it was not meant for that use. There is no need for or guarantee of consistent width or straightness. The rack could be ground but even so, it doesn't compare to the dovetailed ways and gib system of a square column mill. In the square column mill, any movement is linear so a .001' at the column is .001" at the spindle.

Finally, at least on my round column mill, the head is split and is secured by tightening three 5/8" bolts. As you start to draw the bolts tight, friction will grab one side or the other first and subsequent movement will be from the other side which will slightly distort the alignment.
 
I have had a round column mill for forty years and had to deal with the problem of losing registration whenever the head was moved.

The issue with registering at the column is the registration point is only about 2-2.5" from the column center whereas the spindle is about 4-5 times further. This means that any error in registration will be magnified 4 - 5 x. There will necessarily need to be some clearance in order for free movement to occur. Clearance would need to be less than .0002" to insure .oo1" accuracy at the spindle.

Another issue with using the rack as a registration device is that it was not meant for that use. There is no need for or guarantee of consistent width or straightness. The rack could be ground but even so, it doesn't compare to the dovetailed ways and gib system of a square column mill. In the square column mill, any movement is linear so a .001' at the column is .001" at the spindle.

Finally, at least on my round column mill, the head is split and is secured by tightening three 5/8" bolts. As you start to draw the bolts tight, friction will grab one side or the other first and subsequent movement will be from the other side which will slightly distort the alignment.
Those are exactly the points I was referring to. My friend is going to give me his mill but the round column bothered me as far as losing X When the head is moved. I couldn't see any way to reliably fix that problem.
 
This issue has come up here before. The best solution is to securely mont a small laser, such as a from a laser level, pointer or gun site, onto the head of the machine, aim it at a spot on a wall as far from the machine as possible, draw an accurately plumbed, vertical line on the wall where the laser dot is. Then whenever you move the head, you simply align the dot to the line. Since the error is greatly magnified, it is easy to get alignment within a thousandth of an inch or so. Of course, the mills column should be accurately plumb and the machine securely mounted. I have used this method on a drill press, it works very well.
 
Back
Top