Table Saw Alignment

What

The saw arbor needs to be perpendicular to the miter slots.  This ensures that the blade will be parallel to the miter slots.

Before launching into blade alignment, we need to bust a couple of woodworking myths.  The first involves the notion that it’s better to use a flat blade replacement plate than using the blade for alignment.  The fact is, using a flat blade replacement plate is one of those comfortable intuitive ideas which seems right but is actually really bad. Unless you are certain that it’s truly flat then it could actually lead you to misalign your saw.  This isn’t a topic of preference or opinion, it’s a matter of fact. 

It takes a good deal of skill and knowledge to produce a plate which is flat to less than 0.001″ that will stay flat over time.  I haven’t seen any product on the market that could possibly qualify.  No 1/4″ or 3/8″ thick piece of aluminum plate could every qualify no matter how much skill and expertise is applied to it’s manufacture.  Both the material and thickness are completely inadequate.  Buying such a product is a waste of good money no matter what the price.

Not too long ago I purchased a popular blade replacement plate to test it’s accuracy.  It was warped so much that it rocked noticeably on my surface plate.  A customer, who had purchased the same product, sent me his to test.  It was also severely warped.  If I attempted to use this plate as it’s maker insists, there would be no way to determine if a change in reading on the dial indicator was caused by alignment error or warp in the plate.  If I adjusted the saw to minimize the change in reading, then I would be guaranteed to misalign the saw by several thousandths to compensate for warp in the plate.

There is a better method and understanding the geometry is pretty simple.  If you mark a spot on your blade with a felt tip pen, and then rotate the arbor on your saw, that spot will travel in a circular path.  The circle traced out by the spot will lie in a plane. That plane will be perfectly perpendicular to the axis of rotation of the arbor.  It doesn’t matter if the blade is flat or not.  It could be bent like a potato chip and the spot will still trace a circle which lies in a plane perpendicular to the arbor’s axis of rotation.  Arbor runout (from being bent) will not affect the path of the spot.  Flange run-out won’t affect the path of the spot either.  If you always place the stylus of your dial indicator on the spot when you make a measurement, rotating the blade as necessary, then your results will be more reliable than any flat plate can provide.  All measurements (alignment and blade tilt) can and should be done in this manner.  It guarantees accurate results, saves you time (not needing to swap out the blade), and saves you money (not needing to buy the plate).

The second myth crops up less often.  It involves the choice of the alignment reference.  The goal of fence and blade alignment is to ensure that they are parallel to each other.  So, the intuitive comfortable idea would be to make your measurements directly between the blade and the fence.  Advocates of this idea insist that using an intermediate reference (like the miter slot) compounds measurement errors.  So, they will guide their dial indicator jig with the fence and place the stylus tip directly on the surface of the blade.  Unfortunately, this practice often results in a poorly aligned saw.  Why?  Because blades and fences are rarely very flat.  When you see a change in reading on the dial indicator, you won’t have any way to know if it’s due to a fence flatness problem, or alignment error.  The correct procedure is to take all your readings with the dial indicator jig riding in the miter slot and the stylus against the blade (for blade alignment) or the fence (for fence alignment).  The miter slot is machined and much more reliably straight.  You are far better off trusting it as the blade alignment reference because when you see variation, you will know that it’s due to alignment error.  And, when aligning the fence, minimizing the change in reading will yield the best possible results.  As an added bonus, cross cuts using the miter gauge will also be accurate - which is completely ignored when aligning the fence directly to the blade.

Why

Few calssic symptoms of poor blade alignment:

1. Burning of the cut edge
2. Board wandering away from the fence during a rip cut
3. Sawdust being thrown up at the operator
4. Kickback

How

Watch the video…

Or read the text… (and get more info)

Start by marking a spot on the surface of the blade body.  Don’t mark a tooth.  The teeth are deliberately ground at an angle so getting consistent and reliable readings on a dial indicator will be impossible.  The recommendation to mark a tooth comes from an alignment method which involves judging the scraping of a tooth against the end of a stick.  It has no place in the process when using a dial indicator (or in this blog).

Place the stylus of the dial indicator against the blade body on the spot that you marked.   You’ll notice how handy the Stylus Offset Bar is for this procedure.  It allows measurements to be made at the widest part of the exposed blade.  Some tilt the indicator to make measurements at this level but this is a bad practice that produces inaccurate and unreliable results.  Others argue that this is yet another good reason to use a flat blade replacement plate.  I think the Offset Bar is far more convenient, less expensive, and doesn’t introduce error into the alignment process.  It doesn’t matter how wide your measurement is if the object being measured has error.  Rotate the scale so that the large hand points to zero.

Now rotate the blade so that the spot is located at the trailing edge and move the dial indicator so that the stylus is on the spot again.  Any change in reading indicates misalignment.  Correcting the error on a cabinet saw involves loosening the bolts which hold down the table, and shifting the table to minimize the change in reading.  On a contractor’s saw, the trunnion bolts are loosened and the blade carriage is shifted to minimize the change in reading. 

A future update will include photos of the procedure on a cabinet saw and a contractor’s saw.  In the meantime, please refer to your owner’s manual for details.

Here’s a few tips that can help make the process easier.

• Loosen three bolts all the way and leave one snug.  When you shift the table (or blade carriage), pivot around the snug bolt.  This will help you to keep the blade centered in the throat plate and make the process a lot easier. 
• Leave the dial indicator in contact with the blade while you make the adjustment.  Watch the reading as the large hand goes back to zero.  Since the pivot point is not located at the first measurement position, it will take a few iterations to get the blade properly aligned.  So, go back to the leading edge of the blade, rotate the scale so that the large hand points to zero, go to the trailing edge and shift the table (or blade carriage) to eliminate the error.
• You can make fine adjustments by tapping the table (or blade carriage) with a block of wood (or a wooden mallet).  Don’t use a steel hammer as this can cause the cast iron to crack.  For a contractor’s saw, the PALS product can make fine adjustments easier.
• Tighten each bolt a little at a time.  Placing two or three flat washers between the bolt heads and the casting can help minimize squirming as you tighten - especially if you put grease between the washers.  Again, leave the dial indicator stylus against the blade and watch for any change in reading as you tighten.  You can make minor corrections as you go.

Ed Bennett
ejb@tablesawalignment.com