Powermatic shaper: trouble shimming spindle square to table
I got a used “Powermatic 27 super” shaper. Spindle runout checked out good: within 1/1000″ at base and 2/1000″ at top.
Clamping a starrett straight edge between a couple spacers at the base of the spindle reveals that the spindle is significantly out of square with the table. This adjustment is apparently made by shimming the four contact points that bolt the spindle assembly to the underside of the table. I find that when I shim the spindle square in one line, it throws it off in another, and I basically chase the spindle in a circle, constantly throwing it out of square one way when getting it square the other. It seems like an impossible task.
I just spent five hundred dollars on a Freeborn cope and pattern cutter set, and it cuts a super sloppy loose joint because the spindle is out of square.
Am I missing something here? Is there any hope?
The table does seem reasonably flat in the 8″ or so radius around the spindle.
Any suggestions would be very much appreciated.
The powermatic tech. I spoke with was of no help at all. “Just shim it” was all I could get out of him.
Replies
To know we are starting in the same place, the standard procedure for this is to tram the table with the end of the clamped straightedge as the spindle is rotated and check that the clearance between the end of the straightedge and the table stays constant through a full rotation with a feeler gauge. This measures the squareness of the axis of rotation of the spindle to the plane of the table, assuming the table is flat.
Assuming that the 4 support points are arranged in a square, the shimming is done using the two diagonals of the square as perpendicular (and hence independent ) axis of rotation for the perpendicularity of the axis of rotation to the plane of the table.
Tram back and forth 180 degrees on one diagonal between opposite supports and shim those two supports until the clearance is the same at both ends.
Then do the second diagonal, hopefully at right angle to the first. Note that the table top may not be flat and that the second uniform clearance along this line may not be the same as the first uniform clearance along the first line. Formally, the axis of rotation of the spindle is a true datum, an idealized line of geometry; but the table top is a physical datum feature and can't be trusted as a true flat plane datum.
You can check the squareness of the spindle itself to the table--datum feature to datum feature--with a cylindrical square.
If this procedure does not square the spindle to the table, something is distorting the components as the bolts are tightened.
This forum post is now archived. Commenting has been disabled