Knee mill table aligment?

[PekkaNF]

I have a knee mill. Main column has square guide for knee. The mill is not really small.

How do I verify that the table is square to column?

It looks like mill cuts a little off. I can tram the spindle to the table very square and still slab bolted to table will off a little even after careful finishing cuts.

I have some 400 and 500 mm long paralles that check straight. I have one good referece to check them up.

Problem is a "try sqare" I have one small very good square, but is is only 50*70 mm. Then I have some 150-250 mm sqares that are close to proper size but they all are a little off when I check them on parallel, back to back.

There are ways to adjust carpenters square by punching dots to expand metal. This would make them a bit better, but probably does not make stability any good.
http://lumberjocks.com/topics/449

Any good advice?

Pekka

[trapper]

Might be easier if you look in the Archieves plenty of recent answers on this topic

[PekkaNF]

Care to elaborate?

I did search on several words in knee/mill/aligment and I can find plenty on tramming, but tramming I can do. I'm missing information on knee/table aligment methology. I have dreamed up something, but before trying my luck on something weigting several hundred kilos.

I think I need to start from column guides on, see that they are straight

I have two avenues:
1) Could borrow a engineers framed level and work from that on.

2) Foreget leveling it perfectly, and use something else to square it. Engineeing square forexample, just to see biggest shortcoming.

It's Sunday, I'm comparing some 500-600 mm long paralles to know straight edge and squares. Paralles are stright to my purpose, but only smallest square is trully square to grade#1, DIN something and Mitutoyo I trust.

Monday I could shop something or get a Friend to get me out of trouble. But today I'm cross checcking sqares until I have someting to get intila impression that is the table just drooping a little on it's nose or also every and each way.

* Little update: Preliminary test shows that table is a little nose up and column rail is pretty straight. Table seems to be quite a bit inclined lenghtways, left (heavier side) down. Therefore I checkked if the weight makes it rock and it does 0,05 mm / 500 mm distance- Least gibs are tight. I have to clock remainin axis and see that nothing is simply not loose.

There is no obivious easy adjustment to correct that tilt, I want to make sure I'm doing the right thing.

Pekka

[PekkaNF]

I took some pictures yesterday.

Yup, table tilts just a little bit, I'll clock later today where is that clearance, looks like right amount, but I'd like to know where the paly is and how much.

I ordered two bigger Mitutoyo squares grade #2, that should be close enough.

Ok, some pics

This is just aproximation to find out biggest source of problems and to make sure I don't overlook something easy and obivious.

Pekka

[PekkaNF]

Bit more crankking table left and right, fore and aft, while DTI is mounted on knee. Hardly any movement on fore and aft. Knee tilts just 0,03 mm when cranked full left to right. I did lean on table on my full weight and DTI nedle hardly moved.

When table is moved from full aft to fore it raises just 0,03 mm (parallel on top of the table) DTI stylus on parallel and base on column rail (base rail).

Now I have to establish that axis are square. That is the hard part. Second test with better square, then I'll ask one friend to check with square spirit level and make the adjustement plan. Last part is to retram the spindle and take it to test drive.

How does it sounds? Does anybody see any gross mistake here?

Pekka

[philf]

Pekka,

Two books you may find helpful are Machine Tool Reconditioning by Connelly and Testing Machine Tools by Schlesinger. PDFs of both plus other interesting titles can be found on http://totallyscrewedmachineshop.com/documents/documents.html

Phil.

[trapper]

Dont know if this would work as there will be a lot of better trusted ways-such as "traming"once you have established the table is level,A quick way prior to clocking would be to use a precision spirit level-not a cheap version the type you would use for leveling a lathe out if you have established the bed is true with a DTI you could then put the precision level on the table-they can be zeroed in via the adjusting screws on side of the body-zero it in then put the level on the head maybe holding it in place with a magnet-those levels are precise--or couple blocks on table one to left-one to right of spindle say foot either way.A two long paralells long ways sitting on blocks with a gap say 3ins between them back to front (or whatever)you can drop the head onto the paralells so the spindle is sitting on the paralells

[PekkaNF]

Thank you, that's a quite a bit of bedside reading. Had a 5 min look on Schlesinger paper and that looks like exactly what I am looking for. Probably T-slot quality is not up that standard....but measurement and methology looks like what I need. And it is metric.

Pekka

[sparky961]

There are some great old books at that link you posted, Phil.  Thanks for that.