Why Will This Not Work?

dusty · Post by **dusty** » Fri Aug 15, 2014 9:31 am

for verifying table flatness?

If I zero the dial gauge on a known flat surface and then move it around systematically on the Main Table top, will the observed variations be an accurate representation of the tables' flatness?

[ATTACH]25925[/ATTACH]

[ATTACH]25926[/ATTACH]

[ATTACH]25927[/ATTACH]

[ATTACH]25928[/ATTACH]

If not...why not?

I would have given this table a clean bill of health except for the fact that this is the table that caused my concerns with the Shorty when set up as a drill press. This lingering issue has been expressed in a different thread and is yet to be resolved.

beatnik · Post by **beatnik** » Fri Aug 15, 2014 10:00 am

No.

You need a surface plate to start with so you have something that is known flat. Set three supports of exact height on the plate and lay the table upside down on those three points. Then you can check the surface from underneath with an indicator.

beatnik · Post by **beatnik** » Fri Aug 15, 2014 10:01 am

https://www.youtube.com/watch?v=w35b12YJC1s

dusty · Post by **dusty** » Fri Aug 15, 2014 10:08 am

Thank you for the reply. Yes, I know that I can do this with a lot of extras that I don't have. Extras like a surface plate and calibrated pyramids. I'm trying to get a reasonably accurate verification of flatness without an investment.

Maybe you are telling me that that can not be done. If so, I appreciate that also.

beatnik · Post by **beatnik** » Fri Aug 15, 2014 10:19 am

I think what you have is only indicating that inch or so of area. Have you tried a good straight edge to see if it shows anything obvious ?

billmayo · Post by **billmayo** » Fri Aug 15, 2014 10:20 am

I found my framing square and a feeler gauge was all that I needed to check out the flatness of any Shopsmith table or joiner. Close counts for my woodworking needs. I did switch to using the Total Shop steel main tables when I was going to do any metal working with the Shopsmith. I still have 2 more of these very heavy Total Shop steel main tables laying in my shop that I never got around to using.

JPG · Post by **JPG** » Fri Aug 15, 2014 1:10 pm

The block and dial indicator makes a good 'hole' detector.

The surface being 'measured' is an unreliable reference surface.

I agree a good known straight edge is adequate. Backlight to detect, feeler gauge to measure.

It might be interesting to observe the area 'above' the mounting screw bosses as the mounting screws are tightened(one at a time).

wa2crk · Post by **wa2crk** » Fri Aug 15, 2014 2:02 pm

I use a known to be straight 4' level and a flashlight to check for table flatness. I chased my tail for a while when I tried using a 3' aluminum rule that my wife uses for her sewing until I realized that it was bowed about 1/8" at the middle.
Bill V

dusty · Post by **dusty** » Fri Aug 15, 2014 2:46 pm

Let me start here by saying that I do not have a major problem with any of my tables. What I have is a bunch of seemingly contradictory information that has me confused and challenged.

The dial indicator, moved systematically all over the table top shows no significant variations. All points equal to the same point are equal to one another. Yes?

The straight edge and flash light and feeler gauge are good. At least I think that is what has been said.

Well, I used those. Moving Starrett 36" Straight Edge to all of the horizontal and vertical lines in the attached drawing and measuring the gap reveals no gap greater than .005".

BUT, if I relocate the straight edge to the diagonals and make those same measurements (with a feeler gauge), I have gaps in some locations as great as .012"-.014".

How can all points be planar in the horizontal and vertical while at the same time indicate low spots along the diagonal.

[ATTACH]25929[/ATTACH]

BuckeyeDennis · Post by **BuckeyeDennis** » Fri Aug 15, 2014 2:53 pm

dusty wrote:Let me start here by saying that I do not have a major problem with any of my tables. What I have is a bunch of seemingly contradictory information that has me confused and challenged.

The dial indicator, moved systematically all over the table top shows no significant variations. All points equal to the same point are equal to one another. Yes?

The straight edge and flash light and feeler gauge are good. At least I think that is what has been said.

Well, I used those. Moving Starrett 36" Straight Edge to all of the horizontal and vertical lines in the attached drawing and measuring the gap reveals no gap greater than .005".

BUT, if I relocate the straight edge to the diagonals and make those same measurements (with a feeler gauge), I have gaps in some locations as great as .012"-.014".

How can all points be planar in the horizontal and vertical while at the same time indicate low spots along the diagonal.

[ATTACH]25929[/ATTACH]

I believe that a simple torsional twist (of the entire table) could do that. In which case the center would be high when measuring across one of the diagonals, and low when measuring across the other diagonal.