On Tuesday 18 August 2015 10:11:31 John Kasunich wrote:

> The test you described would tell you if the spindle axis is not
> parallel to the Z axis (the column).  It would not tell you if the
> column is tilted relative to the table.  I don't know which type of
> error you suspect - I'd think the latter would be more likely.
>
> Update: I went and looked at some pictures of a G704.  It looks
> like the head can be tilted sideways?  So you are worried about
> that adjustment?  If that is the only thing you care about, then
> you could test it with a straight rod in the spindle and an indicator
> on the table.  But again, that test will NOT tell you if the column
> is tilted.
>
> There are three things you might want to check (and each one
> needs to be checked in two planes, XZ and YZ.
>
> A1) Is the spindle axis parallel to the quill travel
>
> A2) Is the quill travel parallel to the Z-axis (column)
>
> A3) Is the column perpendicular to the table
>
> If the quill is locked, you can reduce it to two items:
>
> B1) Is the spindle axis parallel to the Z-axis (column)
>
> B2) Is the column perpendicular to the table
>
> Note that traditional tramming tells you if the spindle axis
> is perpendicular to the table, but that is NOT the same as
> knowing A1-A2-A3, or B1-B2.  Imagine that the column is
> tilted 10 degrees to the left.  You could tram the spindle to
> be perpendicular to the table by tilting the head 10 degrees
> the other way.  But the Z-axis would still be totally screwed
> up.
>
> You either have to measure all three (or two) items, or trust
> the chinese that the castings and machining are correct and
> the only problem is mis-adjustment of the head tilt.  I wouldn't
> trust them and would try to measure everything.
>
>
> I don't see any need to remove the fixture and workpiece and
> indicate to the table.  You should be able to indicate to the top
> surface of the work itself.
>
> You really want to mount an indicator to the spindle, with as much
> reach as you can get (2-4").  Then zero it in one location, turn the
> spindle 90 degrees, take a reading, turn another 90, take a reading,
> turn another 90 take a reading, and turn once more to get back to
> the original position.  That will tell you if the spindle is
> perpendicular to the surface you are indicating to, whether it is
> table, fixture, or workpiece.
>
> What is your biggest collet?
1-1/8" I think, but no precise rod in bigger than the 1/2" A2 stock I 
bought to make #209 nipples out of.
> Can you stick a chunk of 3/4 rod in a 
> collet and then stick a mag-base to it?  Or C-clamp a V-block to it
> and then mag base to the flat surface of the V-block?  Seems like
> you should be able to improvise something.
>
> There is absolutely no need to use a 0.0001 indicator.  If you get
> it trammed to 0.001 over a 4" span (indicator tip 2" from spindle
> axis) you are going to be more than good enough for even the
> finest woodworking project.

The artifact I seem to ge getting could come from either spindle 
miss-aligned to the post, or the post isn't square. The QC sheet says 
the post is well aligned and I haven't laid a wrench on those bolts 
other than to determine that they were 1/8 turn from broke. I wasn't 
able to move them with reasonable wrench force.

I did have the head off to reduce the weight I had to man handle while 
installing the Z ball screw, so that is where I suspect the major error 
is.  The outer sides of the head, and the Z sled, are painted as cast, 
so a miss-match by using them for alignment purposes is well within 
believable.

So I'll go and "get to it" & see if I can learn anything.

[...]

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

------------------------------------------------------------------------------
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to