Shopsmith Forums

Along with aligning the SS goes keeping it aligned. Many of the SS's I bought were not properly tightened. While I had to use a rubber mallet to get a couple apart I was often surprised at how many were barely holding together. I have wondered how many were poorly tightened at assembly and how many had been maybe shaken loose by constant movement over rough floors. This was mainly at the ends of the bench tubes and way tubes. I know some were never tightened since there was no setscrew ring mark in the tube. Those were just under-tightened. At the other extreme a couple had been so badly overtightened at the bench tube clamps that the tube was squeezed badly out of shape. On one table trunnion the bolts were so over-tightened that I couldn't believe that the threads had not stripped.
Maybe a new SS should come with a torque wrench and a list of values for each bolt and setscrew... I have long observed that most folks that do not regularly do a lot of nut and bolt work do not have a "feel" for proper tightening. How could they? without a way to measure torque and a list of values you can only learn it by experience.
Of course the other important thing about staying in alignment is the way you treat the SS in using it. Simple stuff like not dropping heavy boards on it. Not using it as a hatchet station...
What are some other considerations to holding alignment that I am not thinking of?

RE Dusty's new fangled level procedure: Is the level very narrow AND placed in the gaps of the tooth sets? How else could it be flat against the blade?

The possible runout is the reason the SS procedure references ONE tooth! It is also the reason one should NOT touch the blade when rotating it to check at front and rear position of the miter gauge.(rotate the back end output shaft)

I prefer using a sanding disk for this(it allows longer front to rear travel). It also needs to be to the same spot on the disk.(eliminates runout contamination) P.S. I like to pick the highest point as the "spot".

JPG40504 wrote:RE Dusty's new fangled level procedure: Is the level very narrow AND placed in the gaps of the tooth sets? How else could it be flat against the blade?

The possible runout is the reason the SS procedure references ONE tooth! It is also the reason one should NOT touch the blade when rotating it to check at front and rear position of the miter gauge.(rotate the back end output shaft)

I prefer using a sanding disk for this(it allows longer front to rear travel). It also needs to be to the same spot on the disk.(eliminates runout contamination) P.S. I like to pick the highest point as the "spot".

Absolutely. No debate at all. When doing what you propose here, you are performing a classic runout test. All that you would typically add is a dial indicator so that you can express degree of runout.

Another quick and dirty check involves a ZCI (zero clearance insert). With the ZCI installed, position the main table as though you were going to utilize the table saw with maximum depth of cut. Using the quill, set the blade so that one face just kisses the edge of the ZCI. If you are using a freshly cut ZCI, quill adjustment won't be necessary.

Whiling rotating the blade slowly (BY HAND), observe the clearance between the blade and the ZCI. The deviation that you witness is the runout.

JPG40504 wrote:RE Dusty's new fangled level procedure: Is the level very narrow AND placed in the gaps of the tooth sets? How else could it be flat against the blade?

There are a couple companies that offer special equipment to do these runout tests and table alignments. One offering is basically a guaranteed flat "what amounts to a sanding disk". This disk is mounted on the arbor and used in the same way as you would use the sanding disk to perform table alignment or runout tests.

How about that, a basic Shopsmith setup is already so equipped. Using the "high spot" serves the same purpose as having a perfectly flat alignment plate. No special fixturing needed.

dusty wrote:Absolutely. No debate at all. When doing what you propose here, you are performing a classic runout test. All that you would typically add is a dial indicator so that you can express degree of runout.

Another quick and dirty check involves a ZCI (zero clearance insert). With the ZCI installed, position the main table as though you were going to utilize the table saw with maximum depth of cut. Using the quill, set the blade so that one face just kisses the edge of the ZCI. If you are using a freshly cut ZCI, quill adjustment won't be necessary.

Whiling rotating the blade slowly (BY HAND), observe the clearance between the blade and the ZCI. The deviation that you witness is the runout.

I feel the need to point out that with an improperly adjusted table, the ZCI cut will be non parallel to the miter gauge slot. Perhaps cutting one and measuring the distance from the opposite ends of the cut and the miter gauge slot COULD determine table (mis)alignment.

This does serve as an excellent method to observe runout!

dusty wrote:There are a couple companies that offer special equipment to do these runout tests and table alignments. One offering is basically a guaranteed flat "what amounts to a sanding disk". This disk is mounted on the arbor and used in the same way as you would use the sanding disk to perform table alignment or runout tests.

How about that, a basic Shopsmith setup is already so equipped. Using the "high spot" serves the same purpose as having a perfectly flat alignment plate. No special fixturing needed.

I neglected to mention that to use the sanding disk, it needs to be clean(no sandpaper).[gets a little gritty if not]

If one does not like using a 'high spot', get a stone, fix it to the table, turn ss on and gently feed the quill so as to grind the disk flat where it 'meets' the stone. Once it is 'shiney' all the way around, quit.

YES Dusty this IS the 'starting point' upon which all else(except those 'independent' adjustments I previously referred to) is dependant!;)

ALL that has been said heretofore re blade flatness/table position has ASSUMED little or no runout of the quill itself!

dusty wrote:I think the beginning of this procedure is with the blade. There are ........

I accept .2 degrees as an acceptable deviation. If blade orientation doesn't pass this test with flying colors, STOP. There is a problem that must be corrected.

The Single Point of Reference: I consider the face of the blade to be the single point of reference for all other table alignments. There might be debate on this but that is my personal viewpoint.

dusty...if my trig is correct, a .2 angular deviation would equate to a little over 1/64", actually .0175" "runout". Having no clue what is deemed "acceptable runout" on a blade, would .0175" be considered reasonable? Seems high.

---y---
x......
x....
x...
x..
x.

tan angle = opposite [y] / adjacent [x]
so
tan .2 degrees = runout [y] / 5" [x] (this assumes a 10" diameter blade)
so
runout = tan .2 degrees * 5"
so
runout = .0175"

(apologies for the pictorial attempt at my triangle above.)

mickyd wrote:dusty...if my trig is correct, a .2 angular deviation would equate to a little over 1/64", actually .0175" "runout". Having no clue what is deemed "acceptable runout" on a blade, would .0175" be considered reasonable? Seems high.

---y---
x......
x....
x...
x..
x.

tan angle = opposite [y] / adjacent [x]
so
tan .2 degrees = runout [y] / 5" [x] (this assumes a 10" diameter blade)
so
runout = tan .2 degrees * 5"
so
runout = .0175"

(apologies for the pictorial attempt at my triangle above.)

Taking this a step further: Ignoring actual accuracy(which could increase the measurement inaccuracy conclusion below), the resolution of these angular measurers (wixey, beall etc.) is .1degree. Assuming an angle can be up to +- .05 degree above or below the indication, they have a built in potential error of +- .00436 in. Thus a device indicating .1 degree may be ignoring a runout difference of up to .00872!

We ARE splitting hairs here, but I think perhaps a dial indicator is better suited for runout measurements rather than an angle measurer.

Then again, how close is close enough? I do not think less than 10 thousandths of an inch is too much inaccuracy for the task of setting the table to the 'blade'.

Well now, I am really in a pickle.

If I can't get past deciding how much saw blade wobble I can accept, there is no way that I can go on to perform a complete table alignment.

Or is there?

When we determine and mark the high spot on the sanding disk or we color code a specific tooth on the saw blade and that point is used for "the point of reference" during table alignment, that is just what we are doing.

We are ignoring run out.

I trust that you will correct me if I am wrong but there is no where in Shopsmith documentation that the method for measuring runout is detailed.

Furthermore, if one decides to improve run out performance on a particular Mark V, just what can be done. I believe that the only recourse is to change bearings in the quill or replace the quill.

I am ignoring the fact that you could have a bent blade, or improperly installed blade or an excessively worn arbor or saw dust between the blade and arbor - things like that.

Now, that having all been said, I think it is safe to begin doing the table alignment. Besides, my Wixey (attached to the saw blade) reads 0 degrees. Can you get better than that?

dusty wrote: I am ignoring the fact that you could have a bent blade, or improperly installed blade or an excessively worn arbor or saw dust between the blade and arbor - things like that.

I don't get how you get to ignore those possibilities...
I'm going to drive my pickup to town and ignore the fact that I lost the keys, the fuel tank is empty and all 4 tires are flat.
I know what you are saying but I can see non mechanic operators jumping to the quill without "beginning at the beginning" and checking for all other possibilities first.
There are a number of quill run-out indicators. One simple one is a long drill bit and looking at tip travel.
A little run-out is not the death knell to an accurate cut. The blade is just cutting like a wobble dado but still cutting. The table being at exactly 90° to the quill is far more critical to a saw cut.
The simple way for a newbie to get started (without getting all in a panic because he has now become paranoid that the whole world is now .001 out of alignment at all possible points) is to be sure the miter gauge slots are at 90° to the quill and then just pick up a board and cut it off. Check it with a square, look and see if the cut looks good and if it does go ahead and do some woodworking.
-
On a side note: I was cutting some pieces that I wanted to match existing rough-sawn lumber so I used a steel blade and added a lot of set to one tooth each way to leave a rough cut. I was a little surprised at just how fast I had to feed the wood to get the desired rough cut. Now for the dumb part... I didn't mark which blade it was so it is now "in the stack" somewhere. It's not really much of a problem since I only use carbide for almost everything but it annoys me that I was dumb enough to not mark it heavy.