Replies

1. 
   HowardAcheson | Sep 21, 2011 02:04pm | #1

   Are you sure it's not a problem with the arbor face?  Could be some gunk or metal chip that is "cocking" the blade.   

   Also, are you sure it's not the blade that is out of flat?

   Be sure you are not over-tightening the blade.  That can warp the blade.  The arbor nut should be hand tight with just a touch of additional tightening with the wrench.

1. 
   todpod | Sep 21, 2011 03:28pm | #2

   thanks

   Thanks for the good advice Howie.  I inspected the arbor face with a magnifying glass.  There are no visual issues.  I even tried using clamping washers from a different saw...  same results.  I tried two different blades, plus a very thick moulding blade ...same results.  I find it hard to believe that that shaft could be bent, so I tried hard to disprove it ....to no avail....

   Todd

2. 
   Steamboater | Jan 02, 2012 08:19pm | #3

   I'm dealing with same problem

   My blade is sq. to both the fence & table & I've removed heeling as best I can. But with a dial-indicator  measuring every 90 degrees around the circumference of a 10" blade (I tried 3 blades--a $130 Forrest Chopmaster & 2 Freuds--the thicker ones not the thin ones--I get readings of .015" difference. So I figure the wobble is twice that (.030") for the total side to side wobble. But I don't think it is a bent shaft because if it were I'd expect the blade to also move up and down and it does not. (But maybe I'm wrong.) I posted  this problem all over the 'net today so hopefully someone will steer me straight. All mating parts are clean with no burrs, so I'm thinking it is a manufacturing defect located where the washer presses against the arbor. If anyone has any input it would be appreciated. I'm thinking about honing  the arbor lightly to see if I can do a DIY fix job, but this is new to me so smart $ is against me.

1. 
   RalphBarker | Jan 03, 2012 11:51am | #4

   bent vs. unflat/untrue

   I would agree - if the shaft is bent, it would move through a conical space, moving the blade up and down as well as side to side. So, I'd guess that the flange may be out of true. I wouldn't hone the arbor, though, since that was likely machined on a lathe, and is probably true. Check it with your dial indicator, though, to be sure. Then, add what I'm calling the flange (the inside washer) and the nut (finger tight), and see  what the surface of the flange looks like on your dial indicator. If it is the flange that is out of true, I'd consider taking it to a machine shop with a metal lathe or mill. Or, buy a new one (flange, not RAS ;-)  )

1. 
   Steamboater | Jan 03, 2012 07:14pm | #5

   Non perfect Delta arbor & flange

   Thanks Ralph, I've taken measurements with a dial-indicator with the blades set-up both with and without the Delta flange in place taking measurements at the gullets of  all 3 blades. None of the measurements are dead-on, leading me to believe neither the face of the arbor nor the flange are square to the blade. The problem is worse with the flange in place. So I don't believe at this point that any of it was machined to very close tolerances.

3. 
   oldwindowguy | Jan 05, 2012 10:16am | #6

   You can't really diagnose the condition of a machine spindle or arbor by observing a tool mounted on it.  The only way to determine if a spindle is bent is to remove it from the machine, mount it between precision centers, and run a dial indicator against it.  Even measurements taken directly off the spindle while it is still mounted in the machine won't distinguish between problems in the spindle and problems in the bearings.

   I'd like to save you that effort.  I've seen bent spindles on metalworking machines, but never on a woodworking machine going from small saws to large double end tennoners.  I won't say they don't exist, but my experience (4 decades in the millwork industry) is that wobbling tools can be traced to bearings and/or shoulders (saw flanges being understood as wide shoulders).

   You don't say how large a saw this is or if the primary flange on the saw arbor is loose on the arbor,  pinned to it and turned, or turned directly from it.  0.015" runout on a 10" to 12" saw mounted between 2 loose saw collars is normal.  That is the configuration of the Delta saws I've worked with.  More accurate control requires an arbor with an integral machined flange at least 1/4 the diameter of the saw.  I've seen runout controlled under 0.005" on larger saws like that.

   There were several comments about saw collars.  Better saw collars--precision bore, wider so the bore is longer, relieved and precision turned face--are the most cost effective upgrade for a lot of saws.  I've turned them for a couple of my own saws, and it was well worth the effort.

4. 
   HowardAcheson | Jan 05, 2012 11:59am | #7

   Let me ask a couple of basic questions.  How does the tool cut?  What are you using it for?  Is it your "go to" precision tool or is it mainly use to as  cut off tool?

5. 
   Steamboater | Jan 28, 2012 02:36pm | #8

   .030" run-out on 10" Delta Model 10, 33-990 radial-arm saw

   Howie, Up 'til now it hasn't been an issue since all I've used the ras for is framing & cutting western red cedar siding for my home. It did just fine for all that. But now I started looking a bit closer at my saws' accuracy since I'm doing cabinet work & who knows what else after that. I called Delta & was told their tolerance on a 10" blade is .015" run-out. To me that sounds like a lot. I was also told, even though others have advised against it, that I can indeed do my own work on the face of that 7/8" fixed hex arbor w. a 1200 grit or so diamond stone by simply honing the high  spots. Wish me luck. I'll prbly post myi results in a bit--good or bad.