South Bend Milling Machine - squaring the head

Update:

I'm a little disappointed, I must admit. I removed the head last evening and used the same stoning method that made some progress earlier in the week. The mating surfaces still show some 'frosting', but there are some suspicious looking sanding marks that I think may be someone else's attempt at squaring the head. The sand marks are not quite in the right spot, but I can't imagine any other reason someone would have done that.

Anyway, I worked pretty vigorously on both mating surfaces at the head mount with the honing stone, but to no avail. The same 2 thousandths of deviation remains. I'm eyeing the ways now, considering that I may have been too dismissive of them before. Not sure how to go about that... need to build up some courage before diving in to that.

Perhaps I'm too leery of shimming the column-to-ram connection. It feels like that will de-stabilize the rigidity of the whole machine? It seems the ability of the machine to absorb noise, vibration and harshness depends on solid connections between pieces? Sometimes I think I worry too much...

:cautious:

2017-10-19 19.42.29.jpg
 
Shimming is not the proper method of gaining the alignment you need here. It needs to be scraped with a scraper. A old file will make an decent scraper to get you by using. There are a few threads dealing with this under "Machine Reconditioning Restoring" I think that right, close enough. The surface shown in your picture needs to be scraped flat to a known flat surface. I'm not going to go into deep detail on doing this. Post a thread over in that section if you like and we'll offer advise on getting it squared up for you.
 
Just curious here, there are 2 surfaces in question here, the head that Jake is showing and the ram mating surface. Wouldn't any misalignment entail both? Again, just trying to get a handle on how to approach this.

Rich
 
I believe you're attacking the wrong area if the head to table distance is changing as you swing the head left to right/right to left.
You need to work on the swivel point on the column to fix that. Right?
Mark S.
 
Mark,

This issue I'm having is that the table is not square to the spindle. I hope I haven't confused anyone on that. :guilty:

The column actually does not swivel on these old SB VMMs. It looks as though it would swivel (just like a BP), but it is in fact fixed and not meant to swivel.

Ken,

I agree that scraping is needed. I've never done that before and would need some serious surface plates and practice. That's a big undertaking for me, but not impossible. More research is needed, but for me that's half the fun.

-Jake
 
Silverbullet said:
From the looks of it , you may have to shim the head where it rotates sideways.
Click to expand...

This is somewhat puzzling: is the mount flange to be scraped (which is appropriate if
it isn't true vertical), or the head flange being scraped (which means the mount flange is
vertical but the head flange isn't parallel to the spindle axis)? Isn't the table (the gibs for
vertical and left-right travel) also a possible site for adjustment?

If you rotate the head so it travels horizontally, is its plunge direction reading parallel
to the bed T-slots? If so, that'd rule out head flange, but not mount flange.
 
Jake,

When you sweep the table with the dial indicator, which side is high and low.

I tend to agree that the swivel portion of the head is not the coupuite of the problem. It's generally the knee. Over time it can wear to the point where the table top will tip outward a few thousandths, showing it to be low on the indicator readings.

This generally turns into a major scraping job to correct. Ready for a challenge?

Ken
 
Hi Jake -- Ok I see. But.. those two pieces of the column separate don't they? Or not? It looks like they are bolted together
Mark
 
Jake, your head can not turn left and right? I know there is no nod. Something that was brought up by Ted earlier was placing a shim at the separation point on the column, not the ram/head connection.
Ken is offering the best solution of course, but maybe a 1-2 thousandth shim at the column point will give you some serviceable time if you need the machine right now.

If you do the scraping, a nice walk thru and pics would be nice
 
Before doing anything that is a mistake, first understand the geometry and usage of milling machines. Why is the head being adjusted? Did it move? Not likely on that machine. The head and ram rarely get moved, and typically do not wear enough to measure. Much more likely is wear in the table and in the knee, which get moved all the time. This leads to table sag from wear, and it may well be different with the table at different distances from the column and at different knee heights. Shimming, scraping, machining, or other mods to the head or ram to make the head perpendicular to the table is a big mistake. The angles may be square, but the axes are not, so you will get in a situation where the cutter appears to be moving toward the column as you raise the knee, and the hole you drill by moving the knee will be oval, not round, with the hole location moving as the cut progresses. The geometry is a parallelogram.

I am not familiar with the details of that specific machine, but you must first start at the machine datum. That is typically flat vertical portions of the column casting, machined at the factory between the Z axis slides. These will be surfaces that do not wear, and should be in pristine factory condition. Those are the datum planes of the milling machine, and all the rest of the geometry of the machine is referenced to the datum. First, make the column knee ways parallel with the datum. Then, fit the knee to the base slides so that the Y axis slides at the top of the knee is perpendicular (square) to the machine datum, with flat and true surfaces in the proper planes. Next, same with the saddle to the knee, then with the table to the saddle, then the top of the base column, then the turret with the base, and the _LAST_ thing to do is to make the front face of the ram parallel with the machine datum and the quill parallel to the datum. Doing it any other way is a short term crutch which will actually make the rest of the machine perform worse.

My strong advice is to use the mill as it is, or to rehab the machine correctly, with no shortcuts. That takes the services of pros or a LOT of study, learning, and practice...
 
Last edited:
You must log in or register to reply here.
Back
Top