Replies

1. 
   forestgirl | Oct 01, 2003 07:01pm | #1

   I can think of two possibilities:

   • Your speed square isn't accurate
   • Your insert isn't flat to the table

   forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl   ;-)

2. 
   RANGERP75R | Oct 01, 2003 07:42pm | #2

   ron

      Regular 1/8" kerf or TK? If TK, with or without a stabalizer?  Rip, cross-cut or combo blade? What kind of wood? Stock thickness and what HP motor on the saw? Are you ripping or cross-cutting when you get the 2* degree angle. Have you checked the blade insert to see if it is level with the table. Have you checked he table-top itself for flat? If you were ripping, were you forcing the stock or maintaining proper feed rate?

      You adjusted the fence to parallel, did you check to see if the fence face was 90* degrees to the table if you were ripping?

      You removed the blade to see if anything was pushing against it. Did you check for wobble on the arbor shaft? Did you check the arbor washers or stabalizers to see if they are flat? Run-out? And last, but not least; if ripping did you check to see that the edge against the fence and the down-side of the stock on the table were 90* degrees?

      Just a few questions ( ha..ha..) that might provide info to a solution. The solution is probably a simple detail..

   sarge..jt

     

1. 
   chillywilly | Oct 01, 2003 08:42pm | #3

   My checklist for tonight:

    - is insert level with table?

   - is table flat?

   I used (2) 1/8" crosscutting blades with the same results.  The results are present both ripping and crosscutting.  The motor is 1.5hp.  Feedrate is steady.  Fence is 90* to table.  I'm cutting birch plywood, 3/4" thick.  Speed square...is accurate.  No wobble in the shaft or the washers.

   Somebody out there is gonna solve this for me....maybe it'll be me!

1. 
   jhausch | Oct 01, 2003 08:45pm | #5

   My bet is arbor run-out or blade vibration.  I get the same with my $hitty northern tool contractor's saw.Steelkilt Lives!

1. 
   chillywilly | Oct 01, 2003 09:08pm | #6

   OK...Ignorant question here.....what is "Run Out" and how would it affect a clean cut?

1. 
   jhausch | Oct 01, 2003 11:20pm | #8

   No such thing as an ignorant question:

   As we talk about run out, we mean that the arbor may be slightly bent and causing the end of the blade to wobble.Steelkilt Lives!

2. 
   forestgirl | Oct 02, 2003 02:59am | #12

   IMO, you're confusicating the issue talking about the fence being 90* to the table.  To test the blade for a 90* cut (90* to the table), that's what you want to narrow your focus on.  I'd venture to guess that most of us (those without Starrett squares that is) get the blade as close as possible with whatever square we're using, and then do test cuts to fine-tweak the blade to exactly 90*.

   To do that, cross-cut a piece of wood that's, say, at least 1" thick and 2" wide, then lay the two pieces next to each other as they came off the saw, except roll one piece over to the opposite face.  As the just-sawn ends come together, there should be no space between the two -- they should fit together perfectly in a straight line.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl   ;-)

1. 
   NormInFujino | Oct 02, 2003 06:50pm | #14

   FG: IMO, you're confusicating the issue talking about the fence being 90* to the table.

   Another one for my woodworker's glossary!". . .and only the stump or fishy part of him remained."

   Green Gables: A Contemplative Companion to Fujino Township

1. 
   forestgirl | Oct 02, 2003 07:39pm | #15

   Tee hee.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl   ;-)

2. 
   yotonyb | Oct 03, 2003 04:54am | #16

   Keep 'em clean, keep 'em sharp, keep 'em waxed.  A lotta problems can be prevented/solved just by following those maxims!

   tony b.

3. 
   Steinmetz | Oct 01, 2003 08:44pm | #4

   Ron, are you sure your blade is not hollow ground or, are you inadverently touching any protruding teeth on the blade?

   Hold a lit lamp behind the square to eye ball the fit. ( No light should show if everything's square) Rotate 1/4 turn of the blade and recheck once again . If the material is not too thick, buy and use two stabilizer washers.  The clamp on the angle adjustment should be tight.  Also check if any wood chunks have lodged in the trunnions. Stein.

4. 
   JohnWW | Oct 01, 2003 09:18pm | #7

   If you can't track down the source of the error, just adjust the 90 degree stop to create a square cut and don't worry about it.

   By the way, what type of saw are you working on?

   John

   
   

   ---

   Edited 10/1/2003 2:34:17 PM ET by JohnW

1. 
   Robbie | Oct 01, 2003 11:42pm | #9

   Ron:  The crude explanation for arbor flange runout is "wobble".   That is, if there is wobble in the arbor flange, it will telegraph to the blade.  The blade, instead of slicing cleanly, will wobble through the cut.  Usually, it will show up as blade marks in your cut.

   The other type is runout of the arbor itself.  That would mean that the arbor, instead of being perfectly cylindrical and square to the miter slot, might have a small bend in it.  This too will shape the consistency of the blade's movement. 

   The way to measure runout is with a dial indicator.  This is a device that measures very small differences in shapes and distance.  You can buy a dial indicator and build a jig around it OR buy a jig that includes the dial indicator, like the TS Aligner.   THey are indespensible in tuning machinery.

   Having said that, after you've checked your throat plate, etc. ,if you haven;t played around with "dynamic testing", then I would.  Find a nice piece of thick, true hardwood.  Crosscut it in half using your miter guage.  Then flip one of the pieces 180 degrees.  If there is error in the angle, it will be doubled when you flip it and you should be able to determine how to adjust your blade tilt.   This may help you to compensate.

1. 
   chillywilly | Oct 01, 2003 11:58pm | #10

   Thanks for the explanation.  I know I can compensate for the problem but I want to eliminate it altogether.  It's really a pain in the a55 when the square says 90* and the cut is 88*!!!

   I gets even more frustrating when you're working with angles different than 90 or 45!

5. 
   thmsjohnston | Oct 02, 2003 02:26am | #11

   Ron,

   I have suffered from the same problem with my Unisaw occasionally. What I found was a build up of sawdust in the tilt mechanism ( worm drive) at the end of the travel was not allowing the arbor to fully seat in the 90 degree position. When I thoroughly cleaned out the threads and gears in went to 90 degrees fine.

   Earl 

1. 
   Steinmetz | Oct 02, 2003 04:16am | #13

   Earl (and all)   BINGO! IMpacted sawdust in the wormgear .Stein.

6. 
   mike4244 | Oct 03, 2003 08:29am | #17

   Ron, buy an engineer square. They are not exspensive, a 4" square will do. Are you squaring from insert to the tooth or blade blank? Do not square to the tooth, only to the blank. Make sure insert is flush with table top and the top is clean.

   mike

7. 
   larry309 | Oct 03, 2003 03:32pm | #18

   Ron;

   I just went thru the same problem with my Delta Contractors saw, model 444.  It took me a day to invert and lay the saw down and clean it good and solve the problem.  The small allen set screw in the top was set for 90 deg. but the blade was out of 90 from 2 to 5 degrees.  When I turned the saw over and cleaned it, I found two collars on the tilt adjustment shaft that had moved over the last ten years.  I adjusted the backlash out the tilt and raise/lower shafts and then set the collars to the ends of the tilt adjustment shaft, I reset the set screw for the 90 degree stop and then I ran the saw 6-8 times and checked the blade angle and it stayed at 90 each time.  Maybe this will help.

   larry

1. 
   trialnut | Oct 04, 2003 05:22pm | #19

   I would like to offer what I feel is a more accurate way of setting your saw to square. Rather than setting the blade square to the table, which has many factors that can influence the results, try the following. Get the flattest piece of wood you can find. A piece of stock about 8" wide x 16" long and 2" thick works well. Joint one edge. Now rip the board, flat side down, taking just enough off to cleanup the other edge. Take the board and stand it up on the fresh sawn edge and check the flat side for squareness to the table. Adjust as needed. When you get it adjusted for square this way it takes out the other variables and is more accurate because you are working with longer surfaces. Now set the angle scale to read 90°.

   RichThe Professional Termite

1. 
   johnhardy | Oct 06, 2003 04:39am | #20

   I've found another method that I feel is a little easier and more accurate ... but similar to your method.

   What I do is rip a board that is flat to about 2.5 inches wide (a little less than the maximum depth the saw blade will cut). Then I use the miter gauge or my sled to cut this board while it is on edge ... creating two pieces of the board. Take both pieces and place them on the saw table, closing up the kerf. You should see no light, even if the cut isn't at 90 degrees. Now take one side and flip the board. Put the cut back together. Look for light. If you see some, the cut wasn't at 90 degrees. Note that this method doubles the error, so any correction you make needs to take that into account.

   When you can make a cut like this and see no light after flipping one side, the cut is absolutely at 90 degrees.

   This is how I set up my SCMS as well as my table saw.

   John

1. 
   trialnut | Oct 06, 2003 12:54pm | #21

   John

   The drawback I see with your method is if the piece has any casehardening or other tension in it to make it warp even slightly when ripping then you can be chasing your tail for a while. You can do basically the same thing with my method by doing two pieces and standing them next to each other. I like the wide board stood on edge because the longer the surface the more the error shows. I use this same method to set the jointer fence and a similar method to set the disk sander. For the sander I take two plywood squares, stack them together and sand one edge flat. Then stand them up on this flat edge against the miter fence and sand the verticle edge flat. Take the two pieces and still standing on edge open them up like a book on a flat surface. Butt the two freshly sanded verticle edges together and look for light between them. There are a lot of different ways to set things square and you just have to try several to see what you like best. Whatever works for you is the right way.

   Rich

   The Professional Termite

   Edited 10/6/2003 6:01:58 AM ET by Trialnut