Dressing a Milling Machine table

[Hank.W.E]

Hi Everyone,
I've done some reading about dressing a milling machine table. Does anyone ever have a good reason to do a light pass with a flycutter on their table? What reasons would this be?

My table seems to travel slightly upwards as I move it on the Y axis towards the column. When I measure a cut path, from the the spindle, on work bolted to the table, this error is diminished significantly. When I measure the cut path on the work, on a granite surface, the error is glaring. This makes me think I should machine my table.

If the problem was "nod" then a cut path made on the X axis would be "deeper" on the side facing the operator. This isn't the case.

Is there anything else that can solve my problem? Machining the table is the last resort, and I'd love to hear other possible solutions.
thanks in advance.

[Richard King]

You don't say what sort of mill a Bridgeport or a Chinese mill? Knee Mill?. I would never cut the top of the table. First take a mic and measure the table it should be the same thickness bottom to top on front way to back way, if it isn't find a local grind shop and have them grind the top . If you cut the top out of square because the knee is messed up and if you bore usng the knee crank your bore will be bad. I would guess the back of the knee is worn out of square or the gib on the column is loose making the table top low in front.
Do you have a blade square you can set on top of the table and crank up the knee? Testing the knee travel with a dial indicator as it moves up. We make the Bridgeport knee high .0002 to .0005 on purpose when we make them rebuild. The table should mic the same all the way around and so should the saddle. But those cheap Chinese machines can be made bass ackwards.

[Hank.W.E]

It's not a knee mill. It's an IH Clone
https://www.machinetoolswarehouse.com/x ... 16133.html

[Harold_V]

You most likely have an issue with the column not being perfectly erect. Very common for that type of machine. Shimming the base will resolve the issue.
Do not take a cut on the table, not for any reason. A tooling plate will not address the error. If the column leans, you lose orientation when moving the head, even when the head moves in a dovetail. You also lose orientation when tools of different lengths are used. It's important that you correct column error before attempting other corrections. It's just as important as the knee orientation is on a knee type machine.

Harold

[Richard King]

He says when he moves the Y in and out the indicator moves. I am assuming he puts a indicator stem in a drill chuck or collet and moves the table (X) left and right and Saddle (Y) in and out. You could shim under the Z or column a 1/2" and when he moves the X & Y in it will be the same. He needs to check the base ways and work up. Need to check the table to see if the top and bottom are parallel and it's the same thickness. My friend has this same exact style of machine and we took his table and saddle off put them on granite surface plate and used a surface gauge and checked both and they were a .008 off in the X and Y .
Email me and I will give you his email address and you can hear it from him.
After he has the saddle and table right he would need to set a precision square on the table and move the head up the column and shim the bottom of column square and then when that was done he will need to sweep the table top to check how the spindle head is aligned to the column. He also should lay the square on top of the table and check the X & Y squareness too. My Email is Richard@handscraping.com I can also send you a page out of a book called "Testing Machine Tools" so you can see new machine specs.

Another good quote is "it's better to sit there and look like a fool , then to open your mouth and prove it"

[Hank.W.E]

Yes, the indicator is in a spindle mounted collet. Tommorow, I'll double check the base ways against the movement if the saddle; try to determine if the ways slope up towards the column, or if the saddle itself is bucking up as it moves.

[Davo J]

I wrote this reply the other day when there where no replies, but had to check my manual for test measurements so some of it may over lap what has been said already.

Hi,
I went through a few Chinese mills similar to the Grizzly 3617 (HM52 over here) with table problems.
The first one I got had a tapered table, the second one had a warped table as well as a lot of other problems.

I would start out with measuring all 4 corners of you table and see if they are all the same.
Next check I would do, would be to have a dial indicator in the spindle and run the table back and forth under it in the Y axis.

To check for twist you need to set up the indicator to reach each end of the table, mine was easy as I could rotate the ram, but yours is different machine, but it could still be done easy enough with some thought a bit of steel and some clamps. What you do is move the table front to back (or visa versa) and right down the + or - readings (if any) on the indicator and do the same with the other end and compare the readings. If one side is spot on and the other climbs you have a twisted table.

If you find the table has the same measurements on all 4 corners but the dial indicator moves when the table moved back and forth it would be the saddle not machined right.
If this works out to be the case, nothing needs to be done to your table, but the saddle needs scraping on one side to bring it down. This is not a small job as you will need to scape in a new tapered gib, which is not a job to start off scraping with.
The specs that come with my machine would allow 0.04 per 200mm in flatness, thats around 0.001575 in 8 inches.

I went through your posts to see what machine you have, and see it's not very old. If you find it a lot more than 0.04mm per 200mm (0.001575 in 8 inches) I would be going back on warranty. If not you need to learn to work with it and know where the low and high spots are.

On my test report the allowable parralism (spelling ?) between the work flatness and table is 0.05/200mm in both longitude and cross movement, but mine is supposedly checked at 0.01 and 0.03mm. I have checked and it is a better than that and I think the test reports are just all printed the same.

Dave
PS
I will go to the other computer scan and put up a copy of my test report. I know it's a different machine, but it may give you an idea of what the Chinese will allow as good enough.

[Davo J]

Here is the test report

Dave

[Harold_V]

You'd be best served to post that report in a larger format. I'd suggest no smaller than 800 pixels in width, with 1024 more desirable. The new software will provide the required thumbnail, yet allow a large enough report to be read when clicked. As it stands, it's pretty hard to read.

Harold

[Davo J]

Done

Dave