Are My Trunnions Mounted Correctly

[ChrisNeilan]

Can the trunnions be swapped front to back? If so, that may give you some information...

[skou]

I have two sets of trunnions off. They do not appear to be twisted or bent or in any other way deformed. But have you looked at the trunnions? Those are a pretty stout chunk of cast.

Dusty, my southern friend, cast WHAT?

Cast Iron, stable as a rock. (Or, that is what Chevrolet has taught me.) Cast aluminium? (my British friends taught me) not so stable!

BUT, on my REAL antique Model 10, I've got trunnion pins made out of cast iron, and the sockets made out of cast (forged?) aluminium. (Or aluminum, for you Colonials.)

My stuff works just fine, but I don't have the fancy EXPENSIVE diagnostic tools.

steve

[dusty]

If the 4 mounting bosses on the table bottom are 'coplaner' to each other and parallel to the top surface of the table, and the trunions mating surfaces are 'coplaner' when attached to the tie bar, then the table and trunion mounting surfaces all will be 'coplaner' when secured with the screws. This does not create a change in the 'flatness of the table regardless of how much torque is used to tighten the mounting screws.

If however the table mounting bosses or the trunion mating surfaces are not 'coplaner', the table may be pulled out of flat when those surfaces are 'forced' to coincide by tightening the mounting screws. I think the table is the most likely part to deform.

It is difficult to determine this 'coplanerarity' due to the spacing and shape of the trunions etc.. However if the trunions with tiebar etc. are set upon the table mating surface(no screws) one could just look for any gaps. The presence of any gaps indicate where shimming is likely needed. However one must determine if more than one shim is needed to also maintain table top to pivot axis consistency between both pivots. Not doing so will likely create that heeling discussed over on the other thread. I do not know of an easy method to determine that consistency other than trial and error/correction.

I have reason to believe that they may not be perfectly coplanar. The infeed/rear boss may be a tiny, tiny bit short (or the outfeed/front a tiny, tiny bit tall).

Or my test tool (1" MDF) is not flat.

The test tool in question is a piece of 1" MDF that has been cut to span the trunnion bolt bosses on the bottom of the table. This "test tool" rocks ever so slightly on one of the diagonals. A .003" shim transfers the rocking motion to the other diagonal while reducing the movement to nearly nothing. The ideal shim would be .0025" but I don't have one and (after all) this is a wood workers table saw. What's a .0005" deviation going to cause.

This table is headed back to the "flatness test" while being mounted as a drill press table. I am not going to look at my notes until after I have completed the "flatness test".

In theory, what I have done will not alter the previously recorded measurements. The trunnion bolts are barely snug so as to NOT TWIST the table.

[dusty]

It is now 1/2 hour later than the last time I made a definitive statement.

After installing that .003" shim and barely snugging down the trunnion bolts so as to NOT TWIST the table, the new set of readings ARE DIFFERENT.

I gotta slow down and rethink all of what I am doing. The results contradict all good reasoning. The shim is doing nothing more than occupying what was "free space". It can not be the cause of these changes (I don't think).

Side Comments: The trunnion holes have been drilled out so that all four holes are the same size (1/2"). The dollar bill has been installed per Nick's recommendation. This is the table that normally resides on the Crafter's Station and therefore is NOT the table that started all of this.

[beatnik]

Doesn't really matter how stout a chunk they are, the mounting surfaces can still be off. A piece of mdf isn't going to tell you much at those tolerances.

[dusty]

Doesn't really matter how stout a chunk they are, the mounting surfaces can still be off. A piece of mdf isn't going to tell you much at those tolerances.

I don't really disagree but for right now I can do no better. Does it matter at all that when placed on a piece of safety glass there are no gaps equal to or greater than .001". That's how I chose this "test tool".

[beatnik]

Could you use four known accurate spacers and set the removed trunnion/pivot set up from the table upside down on them on your glass ? Do the same with your table setting the mating surfaces on those spacers ? Have you checked everything is parallel ?

This is what you're working on, right ?

http://4.bp.blogspot.com/-sGvDxNkd9qQ/U ... G_0632.JPG

[dusty]

Could you use four known accurate spacers and set the removed trunnion/pivot set up from the table upside down on them on your glass ? Do the same with your table setting the mating surfaces on those spacers ? Have you checked everything is parallel ?

This is what you're working on, right ?

http://4.bp.blogspot.com/-sGvDxNkd9qQ/UX4CDr8T_UI/AAAAAAAAAkw/-ESdvWdaNkQ/s1600/IMG_0632.JPG

Yes, that is what I am dealing with except that I have a 510 table. If I could test that all reference surfaces were parallel (coplanar), I would have this snockered. The main table not being perpendicular to a drill bit (axis of rotation) was the starting point.

[dusty]

I am going nowhere fast. The other table (from the Shorty) has been checked and today's readings are also different than yesterdays. This table checks near perfect (on the diagonals) except for what would be the right hand corners (front and rear) in the saw mode and they are only .006" to .010" off. Everywhere else, on the diagonals, the readings are less than .005".

All I can conclude is that there is a variable involved that I definitely do not understand.

[ChrisNeilan]

Any wide variations in temperature in your shop?