milling accuracy problems

I've new to metalworking and haven't really done much until I bought a mini mill (Seig X2 variety) about a month ago. I've been having a
number of accuracy/positioning problems with it though that I narrowed down to this simple example:
I put a drill chuck in the spindle and a piece of flat scrap stainless clamped to the table atop a 123 block. I lock the X & Y tables so they can't move. I load a #1 center drill into the chuck and drill a shallow hole (around 1/16 deep). I then raise the head, put a wiggler with a needle point in, turn the machine on and get the wiggler spinning straight by hand. I lower the needle wiggler down to the hole and it is very obviously not lined up with the hole I just drilled (by about .015 to the left). The X & Y slides were locked the whole time and I didn't lock the Z axis at any time nor engage the fine adjustment at any time.
It doesn't seem to be an issue with play in the vertial column. I tried moving it up an down and playing with it a lot while testing like above but it doesn't move by anywhere near .015.
Is it possible for a wiggler to be broken or defective in such a way that can be made to spin straight, but it is not spinning coaxial with the spindle? Is there anything else it could be? To my untrained eye the whole wiggler doesn't look right. The body of it wobbles back and forth as if not concentric (machine spindle is fine though) but the tip can be made to spin perfectly straight and it was like that when I ordered it brand new.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
There could be several of reasons for this, but what comes immediately to mind is that the spindle axis is not at right angles to the machine table top.
Since the wiggler is much longer than the centre drill this angularity manifests itself as an offset error in the X or Y axes, or both. A quick sketch would illustrate this geometric relationship. (BTW, the body of the wiggler may wiggle all it wants, this does not impair the accuracy of this instrument. Of interest is the point of the wiggler which should be adjusted until no lateral movement is noted. The spindle should be rotating at about 1000 rpm for wiggler use).
To check for spindle perpendicularity you must "sweep the table" with a dial indicator, proceeding as follows:
1) mount a dial indicator on the spindle such that the indicator point sweeps a circle as large as possible over the table top. (you will need to read a book or have others explain what this entails). If you are new to machine tools UNPLUG THE MACHINE for what follows.
2) lower the spindle so that the indicator just touches the table and deflects say .010". Lock all axes.
3) rotate the spindle BY HAND so that the indicator point sweeps a complete circle, and note the indicator deflections (called "readings") at the principal points on the X and Y table axes, ie. at locations X, -X, Y, -Y. Think of a gun target with the vertical line the Y axis and the horizontal line the X axis. Imagine this target lying on the table. The indicator point follows the outermost target circle.
4) now the hard part: you need to adjust the spindle head (the box which carries the spindle and quill, if any) until the dial indicator shows no movement along the X and Y axes. This condition would denote that the spindle is perpendicular to the table top. No indicator needle movement is an ideal, in reality .0005" indicator movement over 6" circle diameter would be a very good initial adjustment.
5) unfortunately this is not the complete story, but should go a long way towards correcting the error you noted, provided the machine was made by a reputable manufacturer.
If you have any specific questions on this please post them here for an answer.
Welcome to the world's second oldest profession, and happy metal mangling.
Wolfgang
retooled wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Well, I use a wiggler and center drills when I'm slant drilling, and even though they are very different lengths, they are both axial on my Bridgeport.
I am not familiar with the Sieg. When you said "you raised the head" is it possible to lose registration there?
GWE
snipped-for-privacy@hotmail.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

    Is the column a round column, or one with dovetails? If it is round, there is nothing to keep the head from rotating around the column as it is raised and lowered -- thus introducing error such as this. You need to re-establish the zeros whenever you do this.
    If it is a column with dovetails, then it may not be perfectly vertical relative to the X-Y table.

    No! The design of the wiggler eliminates all errors relative to the spindle.

    The body may wobble if the chuck is not concentric with the spindle. Or if the chuck is not properly gripping the wiggler's body. But the wiggler point still should be true.
    Good Luck,         DoN.
--
Email: < snipped-for-privacy@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
There was a thread on the Grizzly-Harbor Freight group a while back on alignment of the minimill <http://groups.yahoo.com/group/GrizHFMinimill/ . The meat of it is that you need to square the column to the table first by traversing an indicator mounted on the head up the side of a square or good ground angle plate mounted on the mill table. Do this first with the indicator moving up the side of the square perpendicular to the long axis of the table, adjust by tilting the column. Then move it up the face parallel to the long axis, adjust (if necessary) by shimming the column where it attaches to the base. Then you have to check that the spindle is parallel to the mill column - here you mount the indicator in the spindle and check the run out when slowly rotating the indicator from side to side and front to back across flats mounted on the table. If the spindle is out fore and aft (towards or away from the column) it must be shimmed where the spindle box joins the part that rides up the dovetailed column - this requires removing the head from the column. If it is out laterally (tilted about the y-axis) slightly loosen the 4 bolts mounting the spindle box to the casting riding up and down the column, remount the assembly to the column, and tap laterally while testing with the indicator until the spindle is square to the table. Then remove and tighten the bolts, replace and recheck. A real pain in the butt, but you only have to do it once...Joel in Florida =============retooled wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks to all those who replied to this thread. I found the problem: the mill head was not parallel with the column it slides up and down on. Simply tightening the gib where the mill head mounts onto the column brought this error down from .015 to within .002. It seems I'll have to carry out the process described by Joel below to shim it.
I should have trammed this thing properly to begin with. That was in the back of my mind while investigating this problem, but I assumed that would only point out error in the adjustment at the base that allows the whole column to lean over (which couldn't have caused this particular error). It didn't occur to me that the head itself could be out of parallel with the column it rides on.
Thanks again for the recommendations.
snipped-for-privacy@aol.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.