Shopsmith Forums

That is what was used before the dial indicators came on the market

I have a dial indicator that I am trying to learn to use. I have been attempting to test run out on an arbor mounted saw blade.

If I hold the drive shaft from turning and spin the arbor on the shaft I get a different reading than if I secure the arbor to the shaft and turn the shaft.

Can someone explain the differences that I seem to be reading.

When I started this effort I was assuming that the blade when mounted was perpendicular to the shaft. Is this approach naive?

If I was working on a typical headstock, what run out should I expect? It appears that I have about .007" or .008". Some arbors/blades are better but none are worse. What is acceptable?

BigSky wrote:I have a dial indicator that I am trying to learn to use. I have been attempting to test run out on an arbor mounted saw blade.

If I hold the drive shaft from turning and spin the arbor on the shaft I get a different reading than if I secure the arbor to the shaft and turn the shaft.

Can someone explain the differences that I seem to be reading.

When I started this effort I was assuming that the blade when mounted was perpendicular to the shaft. Is this approach naive?

If I was working on a typical headstock, what run out should I expect? It appears that I have about .007" or .008". Some arbors/blades are better but none are worse. What is acceptable?

CAUTION!!! Do not attempt to read unless fully rested and wide awake!;)

When the arbor is secure to the shaft, it assumes a position that does not move relatve to the shaft. Thus the indication is the 'wobble' of the surface of the saw blade plus any contribution of off centering of the arbor plus any shaft wobble.


When the arbor is loose on the shaft, any added wobble due to the changing position of the arbor on the shaft is included, but any shaft wobble is not.

It is possible that the force rotating the blade is also deflecting the blade.

The blade will be perpendicular to the shaft if the mating surface on the arbor is also perpendicular to the bore in the arbor. However the blade may(read that as most likely!) not be flat! That is why squaring the table to the blade includes observations of the same blade tooth at both front and rear of the table. A single tooth when the shaft is rotated will define a plane that is perpendicular to the shaft(assuming no shaft wobble).


If that 0.007 to 0.008 or less includes the blade out of flat, then those are not excessive. Keep in mind the workpiece 'sees' the cutting edge of the closest tooth as the dimension being cut to. Visualize a wobble blade!

Big Sky, I measured runout on two saw blades mounted on two different arbors and got readings similar to yours -- .007 and .008. I would have hoped for better, but that's what it is. These were saw blades with 1 1/4" arbor holes.

Here was my set-up:

[ATTACH]12818[/ATTACH]

And before everyone started using dial indicators nobody knew that there was anything to worry about as long as the cut was good.

.