Bent arbor?

[jspelbring]

OK, I went to the dark side last year (Sears 22114). I even gave it an Incra TSIII. A couple of months ago, I was ripping through some thick oak, and managed to stop the saw and trip the breaker. After that, I noticed that the cuts didn't "feel" right. I also noticed that the blade seemed to "double cut" - that is, when the work piece made it to the back of the blade, it made a slightly wider kerf - and would spit more sawdust at me.

So, I switched blades - no help, so I assume that I didn't bend the blade. I finally got an alignment tool (dial indicator on a miter slide). I aligned the table to the blade, and that seemed to help a little. However, I can place the depth gauge anywhere an the blade, spin the blade, and get .010 variance on the dial.

To make a long story short (I know, too late). From you veteran wood workers out there, does this sound like I bent the arbor, or am I just doing something wrong with my attempts at alignment?

Any and all tips, thoughts, opinions would be most greatly appreciated.

-Jon

[LCHIEN]

just a quick thought. Speaking at least on theory,
a bent arbor would have maximum variation (on the dial gauge reading) on the outside and minimal variation as you got to the center.

You didn't overtighten your arbor nut and warp the blade, did you?

[cgallery]

First, are you sure you are reading the gauge correctly? .01" is one-hundreth (1/100th an inch). When the saw is slowing down you would absolutely see the blade wobbling.

If the gauge is telling you .001" (that is two thousandths), then you have very little combined runout (excellent).

If you are getting .01", then UNPLUG the saw.

Remove the blade nut and the washers and blade and clean everything thoroughly with laquer thinner. Clean the arbor flange and threads with a toothbrush, then inspect them for any bumps or ridges or foreign material.

Now reassemble everything, checking each component step by step for any foreign material or machining problems.

LIGHTLY tighten the arbor nut, and perform your test again. Any better?

If not, switch to another blade. Any different? If not, go purchase a brand new blade. The Freud Diablos are laser cut and have a pretty flat plate. Try again. Any better?

If not, you are probably getting to the point of an arbor problem.

Acceptable combined runout is debated. I like to keep it to .005" or less. Some guys say anything in the thousandths (.009" or less) is okay. Less is better, though, that can't be debated.

[jspelbring]

Thanks!

I'll give those steps a try. Yes, it actually reads .010, and yes, when the blade is slowing, I can see a slight wobble. I've got the (unopened) blade that came with the saw, so I'll check with that (the one that I've been using is a WWII).

[jspelbring]

Bummer

Well, I got out my handy-dandy dial indicator and alignment tools. The good news is that I have everything aligned correctly (within less than 1,000/in), the bad news is that the arbor is indeed bent.

I guess it's a call to Sears to see what can be done. <sigh>

[eddy merckx]

You can bend an arbor just cutting wood? Is that really possible? You'd think a steel shaft would require a hard blow to bend, like if something heavy was dropped onto the saw with the blade extended. If that didn't happen, it sure sounds like a warranty issue.

[Knottscott]

An out of spec arbor is not a common occurence but is far more likely than a bent arbor.

Blade runout is more likely to be at it's max when slowing than when at speed. It's not uncommon for a saw to give a little shudder on shut down. To eliminate the possibility of the blade contributing, try measuring the arbor flange when rotating it by hand or even at speed just to compare. I'd also double check pulley alignment, blade alignment, belt tension, belt placement on the pulley, and possibly even the bearings.

[Tom Miller]

Have you (can you) measure the runout directly on the arbor shaft, or have you only measured on the blade? Another possibility is that the blade washers/stabilizers are not holding the blade exactly perpendicular.

If you can't measure directly at the arbor, consider rotating the washers 180 and rechecking, or some similar approach to rule them out.

Regards,
Tom