No announcement yet.

TS3650 alignment

  • Filter
  • Time
  • Show
Clear All
new posts

  • TS3650 alignment

    I recently purchased a TS3650 table saw and I’m in the process of aligning the saw. I’m trying to find information on the maximum acceptable run out between the fence faces and miter slots.

    Current state of alignment:
    • The run out between the left miter slot and the blade is within 0.001
    • Using a feeler gage the maximum gap between the table end and a 6 foot level is 0.007 (located at the mid rail bolt).
    • Using the same technique, the front rail mating surface was measured flat to within 0.0025.
    • The table and front rail mating surfaces were stoned to remove burrs.
    • A full face 0.005 shim was installed between the table and rail at the low point.

    To determine the fence alignment, a series of 3 ‘fence to miter slot’ trials were conducted. For each trial the left and right fence faces were measured from both the left and right miter slots with the following results:
    • The left fence face run out relative to the left miter slot was consistently 0.017 (ccw)
    • The right fence face run out relative to the left miter slot was consistently 0.009 (ccw)
    • The left fence face run out relative to the right miter slot was 0.013 Ave (ccw) (0.0115, 0.014, 0.0135)
    • The right fence face run out relative to the right miter slot was 0.0026 Ave (ccw) (0.002, 0.0035, 0.0025)

    Given the variables associated with repositioning the fence and measurement jig, the maximum trail-to-trail run out variation of 0.0025 (0.014 – 0.0115) seems reasonable.

    However, based on these measurements it appears that:
    • The miter slots are out of parallel approximately 0.005 (difference between the right fence face run outs 0.0064 (0.009 – 0.0026) averaged with the difference between the left fence face run outs 0.004 (0.017- 0.013)
    • The fence faces are out of parallel by approximately 0.009 (difference between the left and right fence run outs relative to the left miter slot 0.008 (0.017 – 0.009) averaged with the difference between the left and right faces relative to the right miter slot 10.4 (0.013 – 0.0026).

    Are these run outs within normal manufacturing tolerances and/or acceptable saw operation requirements? If not, how do I go about correcting for miter slot and fence face parallelism?

    Any help would be greatly appreciated.


  • #2
    Re: TS3650 alignment

    Were your measurements to the miter slots to the inside edge or outside edge? You will probably find a variation .
    Fence side parallelism can be corrected with the use of an auxiliary fence. Make sure you have perfectly flat material and do not over-torque the attachment screws. Shim as required.
    For your miter gauge to slide, it will need some clearance as the edges of the miter slots are not polished. The clearance for slippage gives you the option of a left or right pressure bias when using the miter slot. This will eliminate over half of the .004 difference.
    Align your blade based on the side you most commonly use the miter slot (ie, if you use the miter guage to the left of the blade, align the blade to the left miter slot. If you use the right miter slot for tennoning fixtures, etc, align the blade to the right miter slot). Align the blade to either the inner or outer edge of the miter slot, and remember this bias when you use the miter gauge/etc.. Then align the side of the fence you most commonly use to the blade. For reference, my current blade has less than a .002 runout at the rim, I have it aligned to within .001 to the same tooth on the blade rotated front and back to the inside edge of the right miter slot (I use a tennoning jig a lot) and my auxiliary fence (wood) had about .007 variation along its length and about .004 top to bottom on the day I measured it. With the high temp/humidity today, I doubt the fence is still within those parameters. I recheck and reset the fence when moving to the left side of the blade on the occasion that that occurs. With the slop in the factory miter guage, I use it indescriminately between sides as the slop is more than the difference between the slots.

    Bottom line, it sounds like you have an exceptionally close-toleranced table saw for the $500-$600 range. I doubt you will better those tolerances with a $1000+ saw.

    Just a guess that you are either have a metal working machinist background, or you are an engineer in the mechanical/aerospace or design discipline. Due to the properties of wood products when exposed to different humidity conditions, tolerances in the .001 to .0001 range are not achievable on a repetitive basis unless you have a climate controlled environment and the time to let the wood acclimate to that environment (can be up to a year or more with a dense thick wood).

    Last edited by Gofor; 08-25-2007, 09:01 PM.
    Practicing at practical wood working


    • #3
      Re: TS3650 alignment

      Gofer, Thank you for the reply.

      Yes, I'm a recently retired PWA Mech Eng. I'm told there's an invisible tattoo on my forehead that everybody except another engineer can see (or in your case sense).

      I fully appreciate your comments on the dimensional stability of wood with temperature, humidity and aging (and the saw for that matter with temperature and wear). The max variability I saw was 0.0025 which I thought was very reasonable. Each of the three fence trials, while independent, were done within an hour of each other. I'm sure, particularly here in FA, that that spread would be different tomorrow.

      When I did the measurements, I tried to accommodate the miter slot width variation by biasing the seating pressure against the miter slot face closest to the fence. I would cross cut and tendon similarly. Your advice to bias the alignment, to the miter slot and side normally used most, makes sense to me. And I will cut an aux fence to zero out the fence parallelism and squareness as you suggested. Going in, my biggest concern was the overall spread between fence RO's, being 0.015 across the table (~ a 64th) and not having a calibration (read that, experience) to know its implication with respect to the saw operation (that is, binding or burning). Your shared numbers are most helpful. Thanks again.