rip fence question

Any measurable bow could cause the work piece to bind and cause any number of problems to your saw, your work, or most imporatantly: you. I keep mine square with a simple framing square, and that seems to do it for me. When I fist set up my fence, I checked some test cuts with with three different squares, and the cuts were true. I fine tune the saw’s table to the blade and the fence follows.

-- Brian, Lebanon PA, If you aren’t having fun doing it, find something else to do.

I was looking through the user manual for my Shop Fox fence and saw this. “Adjust the fence until the back end falls 1/64” away from miter slot so the workpiece clears the opposite end of the saw blade without binding. This is called “lead” (pronounced leed).” I finally quit trying to get the fence in line with the miter slot and just adjusted it until when I ripped a board, it measured the same on both ends. Just so happens that it pretty much lines right up with the slot. What do you guys think about this “1/64” thing?
-JJ

-- Make things with wood.

Why would you purposely want to throw off your cut? I agree you do not want binding, but a table saw is used for a lot more than rough ripping lumber. If your fence is off 1/64”, and you are cutting a dado, you are deliberately causing the back teeth to come up into uncut wood. Lousy edge and inaccurate cut. If you use a fence guided tenoning or splining jig (like for picture frame corner splines,) again, you are throwing in a little more error that will cause problems with the final fit.

Also, with the fence off, you are depending totally on the riving knife or splitter to prevent the drop side from being caught on the upward moving teeth. If you do not have one, or the splitter is not of the best quality, as is the case with most less expensive saws, you are again allowing the wood to contact the back teeth, where catch and lift are the most common cause of kickback.

A couple thousandths is okay, but .015” to .016” is way too much.

Its a poorly thought out solution to a problem that causes more danger than doing it right the first time. Kind of like many of the government solutions to problems.

JMTCW

Go

-- Go http://ncwoodworker.net/pp/showgallery.php?cat=500&ppuser=730

I set mine up for equal distances from teeth at the front and rear of the blade, no toe out. I use the same tooth to make the measurement. I’ve never checked my fence to see if it’s got a bow. <g>

I also use a shop made splitter that’s slightly thicker than the blade kerf. That’s a big part of the equation.

-- - Real men read directions

I agree guys. It didn’t make good sense to me either. It’s kind of like putting a 4’ level on something and it’s only off 1/16”. That might be ok for 4’, but if you’ve got to run that 20’ you’re going to have big problems.
-JJ

-- Make things with wood.

here’s a bit more detail. i ordered a biesemeyer fence to go with my new contractor saw. i set it up using a dial indicator and homemade jig that fitted to the miter slot – the jig has essentially no play in it.

when i measure from a point at the front tooth of the blade to a point at the back of the blade, the face of the fence is wavy – not by much, probably +/- .001 – .002. from the front of the table to the back of the table, though, it’s more like +/- .005, or up to .01” from the highest to lowest point.

in reading about setting up the fence, i see folks that recommend setting the back of the fence open ~.002” or so relative to the blade to prevent binding (as folks above have alluded to). this suggests to me that the error in the fence face that i am describing (e.g., waviness) should be much smaller than this – if it isn’t, there doesn’t seem to be much point in specifying having the back of the fence open by .002 . . .

similarly, the face of the fence is not perpendicular to the table. i’ve checked the face using a good square and feeler gauges, and it’s off by .006 – .007”. i’ve checked the table using a starrett edge, and it’s off at the edge of the main table segment by ~.0015, so that’s not the (primary) source of the lack of squareness.

so i’m well aware of the tendency to fuss over new equipment trying to make it perfect – my question is whether the above tolerances are really problematic, or whether i’m falling into that trap. what are the tolerances that i should expect out of the fence?

Not sure what whould be acceptable. I can tell you that I am the first to reach for the dial calipers or the mic, but just breathing on wood can cause it to swell a .001 I think. All I can say is get a nice long thick piece, jointed and planed to 4s and give it a rip. If it turns out satisfactory to you then it is within tollerance, if not figure out what is wrong. Just my .02 but then I dont even own a TS yet.

-- Jimmy, Oklahoma "It's a dog-eat-dog world, and I'm wearing milkbone underwear!"

When you’re measuring “waves” in a fence face, use something approximately 12” long to run between the indicator and fence.

How often do you rip things the same length as a dial indicator tip? I didn’t think so. <g>

-- - Real men read directions