Table saw blade set up question

[Bill Bryant]

Is it possible for a saw blade on a good contractor saw (Delta) to remain at 90 degrees to the table and parallel to the miter slot, but move farther or closer to the fence when raised or lowered?

In other words, could an otherwise perfectly adjusted blade be 1.000 from the fence when cranked up one inch, and 1.004 from the fence when cranked up three inches? (This while remaining parallel and 90 degrees to the table top)

If so, how could this be corrected for? Or could it be?

[Art Walker]

Is it possible for a saw blade on a good contractor saw (Delta) to remain at 90 degrees to the table and parallel to the miter slot, but move farther or closer to the fence when raised or lowered?

In other words, could an otherwise perfectly adjusted blade be 1.000 from the fence when cranked up one inch, and 1.004 from the fence when cranked up three inches? (This while remaining parallel and 90 degrees to the table top)

If so, how could this be corrected for? Or could it be?

You should use one of the miter slots as your point of reference on
the table. Measure their deviation from parallelism and just keep that
in mind in later measurements. Use the slot which allows most convenient
measurements...in the general case this will be the left slot for right tilt
blades.

Anyway, assuming you're happy with the heel situation, measure from
miter slot to blade plate near/at the table surface, over the arbor. If this differs at 1" from 3" exposure, and you're really convinced the blade is vertical at both exposures, then sources of change include:
1) runout in the blade itself. repeat the measurement with the blade rotated in intervals of 45 degrees.
2) runout in the arbor flange. repeat the measurement with the blade rotated on the arbor in intervals of 45 degrees.
3) if the arbor axis isn't parallel to the hinge axis of the arbor bracket, then you can imagine having the arbor stay roughly horizontal throughout a range bracket rotation wherein the lateral position of the arbor moves out of the original plane. Hard to believe this casting could be bent though.

0.004 difference between 1" blade and 3" blade is something most people can live with. If it makes a difference in your work,then measure at a few more points between blade peeking out and full height, make a chart of this, and therafter tune fence position or whatever according to blade height. However, at this level you need to stick with one blade and make a witness mark where the flange meets the blade plate to maintain the validity of your calibration chart.

Blade to fence is something you do after you can't improve blade to miter slot invariance any further.

Art

[Art Walker]

If this differs at 1" from 3" exposure, and you're really convinced the blade is vertical at both exposures, then sources of change include:
1) runout in the blade itself.

I should add, I read recently about someone who was using arbor flange and washer with mismatched radii, and was actually turning the blade into a very shallow cone when he cranked the arbor nut home. The usual runout measurements made at constant radius would not detect this pathology, which would definitely contribute to a difference in measurements at different blade heights, as well as bad cutting results.

Art

[Tom Veatch]

Is it possible for a saw blade on a good contractor saw (Delta) to remain at 90 degrees to the table and parallel to the miter slot, but move farther or closer to the fence when raised or lowered?

In other words, could an otherwise perfectly adjusted blade be 1.000 from the fence when cranked up one inch, and 1.004 from the fence when cranked up three inches? (This while remaining parallel and 90 degrees to the table top)

If so, how could this be corrected for? Or could it be?

If you're sure that's what you're seeing, then, at the risk of sounding facetious, it's obviously possible. Mechanisms can certainly be fabricated that will display that sort of movement.

I'm not familiar with that particular saw, but those that I am familiar with adjust the blade height by rotation about an axis that is presumed to be parallel to the axis of the blade arbor, or, IOW, perpendicular to the plane of the blade. I've not made any geometrical analysis to prove or disprove it, but I suppose it's mathematically possible there could be some angular relationship between the height adjustment axis and the spin axis such that there would be two points in the adjustment range that would give the results you describe.

If so, I believe it would be due to manufacturing tolerances in the geometry of the trunnion assembly and doubt there are any adjustments or shimming that could affect that angular relationship.

It is also possible that if the trunnion is against the stop at the 3" blade height position, a slight deflection of the trunnion could be caused by pressure against the stop. You could eliminate that possibility by taking your measurements at a third height setting slightly below the maximum blade elevation.

Have you tried taking those measurements with different blades or with a known flat plate, or with the blade/plate rotated/indexed to different positions on the blade arbor.

If it was me, I don't think I could repeatably adjust the position of the fence to within 1/250" of the target cut width, so I'd call it good.

[Ralph Barhorst]

Is it possible that your fence is not perpendicular to to table top.

[Bill Bryant]

I'll do some more measurements today and post again with more concrete results.

[Bill Bryant]

OK I've done the measurements.

1. Using a dial indicator I confirmed that the runout on my saw blade about 1/4" below the teeth is inside .001. Nice job Ridge Carbide!

2. The blade is dead on parallel with the left miter slot and within .002 of parallel of the right miter slot.

3. The blade is off .004 of parallel with the left miter slot when tilted to 45 degrees.

4. When raising the 90-degree blade from .75 above the table to 3.00 above the table, the blade moves left .010. (That's as much range as my tools will measure accurately.) Using a little guesstimating, I'd say the blade moves left .015 over the full range of height adjustment.

5. I took off the zero clearance insert and sighted down the slot made by the blade. There it is. The slot has a slight but clearly visible arc. The center of the slot is to the right; the ends are visibly to the left.

6. I remember a lot of burning when making the zero clearance slot. Now I can see why. As I raised the spinning blade through the ZCI and into a scrap 2x4 clamped above it, the blade moved up AND OVER .015.

7. I can see, looking at the mechanism, that there is no way this can be shimmed or adjusted out. It's there to stay.

Does this seem at all reasonable on a new Delta table saw? Would this be inside manufacturing tolerances or would it be a warranty issue?

[Mike Marcade]

Try your measurement of the blade moving left as you raise it from .75" to 3.00" again without the zero clearance insert.

If you use the screws in the sides of the insert to hold it tight in the throat opening, it can put the blade in a bind.

I noticed on my saw, that I would measure my blade parallel to the miter slot without the insert and it would be dead on. Then after I put the insert in the blade would be off by .010". I just loosed the screws on the sides so the insert floats a little more in the opening and it works well.

[Bill Bryant]

Try your measurement of the blade moving left as you raise it from .75" to 3.00" again without the zero clearance insert.

I took all my measurements with the ZCI removed. I re-read my post and see why you thought it was on there during measurements. But it wasn't.

[Mike Marcade]

I would definitely go to Delta on this one. They should replace this under warranty.

[Tom Veatch]

I wouldn't sweat 1/250", but 1/64" is, IMO, too much. Talk to Delta.

[Bill Wyko]

I'm the one that had trouble with my arbor mechanism misalignment. It was caused by the motor mount hitting the fence rail at around 40 degrees. you might make sure nothing is touching or binding like mine was.

[Bill Bryant]

I'm the one that had trouble with my arbor mechanism misalignment. It was caused by the motor mount hitting the fence rail at around 40 degrees. you might make sure nothing is touching or binding like mine was.

Nothing is touching or binding. I can raise and lower the blade while keeping the contact point of the dial indicator on the blade. It's one smooth motion--the blade plane goes left from .000 to .010 and back right to .000 without a bump.

[Richard Dragin]

I'll guess that the trunions are not sitting at zero when mounted to the table. You could still ninety the blade to the table but as you move it up and down the blade would shift sidways. You could shim it and experiment till you figure it out but I'd just ask for an exchange under warrantee, good luck.

[Art Walker]

Nothing is touching or binding. I can raise and lower the blade while keeping the contact point of the dial indicator on the blade. It's one smooth motion--the blade plane goes left from .000 to .010 and back right to .000 without a bump.

I'm with the other guys: if the saw is in warranty period you should pursue replacement. 0.015" (your extrapolation to the blade's first emergence from the table) is more than half the difference between blade plate width and kerf width for many blades, and because of this it will be hard to maintain alignment of the blade and a splitter of correct thickness for such a blade as the blade elevation is changed as appropriate for workpieces of different thickness. And this makes the migrating blade plane an arguable safety issue for the manufacturer, on top of its being an impediment to precision work. Not saying there aren't a lot of misaligned saws in the field, but there's a difference between misaligned and unalignable to standards appropriate to your level of diligence. Unfortunately I don't think Duginske addresses lateral migration of the blade plane in Mastering Woodworking Machines, but his statements there on thresholds of accepability for other classes of residual misalignment may be useful to you in asserting a particular number as an appropriate quality control limit exceeded by your saw...without which you may not even get them to look up an internally documented QC spec.

Art

[Bill Bryant]

I phoned the main Delta/Porter Cable service center in Denver and was told that the problem was that the rods connecting my trunnions were twisted, making the trunnions out of alignment with each other.

Try as I may, I can't see how this would cause the saw blade to move toward the fence and heel in while being lowered.

This isn't making sense to me.

[Chuck Lenz]

I phoned the main Delta/Porter Cable service center in Denver and was told that the problem was that the rods connecting my trunnions were twisted, making the trunnions out of alignment with each other.

Try as I may, I can't see how this would cause the saw blade to move toward the fence and heel in while being lowered.

This isn't making sense to me.
What is their answer to fix it ?

[Bill Bryant]

Best I could understand, I'm supposed to loosen the rods connecting the trunnions, make them parallel again, and then the blade should then stay in the same plane when being raised and lowered.

First problem: I have no means of knowing that I've taken the twist out of the rods short of devising a way to turn the saw upside down and put a surface plate across the rods. Youch!

Second: I can't see what a twisted trunnion connection has to do with my problem. It might do something to the relation between the blade's being parallel to the miter slot at 90 degrees and also being parallel at 45 degrees (I'm off at 45 degrees by about .004 I think). But it seems to me that the only thing affecting whether the blade stays in the same plane when being raised and lowered is whether the axis the blade spins around and the axis of the pivot that raises the blade are parallel. The only movement when raising and lowering the saw blade is not a trunnion related movement at all, so I can't see how trunnion alignment would have anything to do with it.

[Rob Will]

But it seems to me that the only thing affecting whether the blade stays in the same plane when being raised and lowered is whether the axis the blade spins around and the axis of the pivot that raises the blade are parallel. The only movement when raising and lowering the saw blade is not a trunnion related movement at all, so I can't see how trunnion alignment would have anything to do with it.

I think that is the correct analysis of the problem.
It would be interesting to know if the axis are parallel.

Rob

[Mike Marcade]

The spot on the table where the trunion bolts up should set the distance between the table and the rods. Everything else shoud be set up in the grind of the gearing, backlash, etc. This really isn't user adjustable in my opinion. Does Delta expect you to shim the gearing mechanism like you were setting up a rear axle on a 4 X 4 or something?

If it were me, Delta would be getting the thing back whether they wanted it or not! :D

[Richard Dragin]

To check if the rods are parallel is easy. Remove the blade and lay a piece of glass on them, something about 8"x10" should fit (but check first). It may not be intuative to understand how the off set could cause your problem but untill you check it I wouldn't doubt the service guys.

[Bill Bryant]

This morning I spoke at length with a top guy at Delta's service center in Denver, and he was dead certain that my problem could be traced back to trunnion alignment.

I can't for the life of me see how this could be, but I guess I'll take the Biesemeyer off, take off the cast iron wings, pull the saw stand off, turn the saw over, put a plate on the rods, etc. etc.

Oh, well. I guess I've always wanted to take a saw all the way down to parts and rebuild it. I just hadn't planned on doing it this Thanksgiving.

I'll report back in a few days.

(BTW I just found another thread here, a rather long one, where someone had my same kind of problem with a top dollar Powermatic 2000. When I read through that thread, it was even more clear to me that the issue is a problem with non-parallel blade spin axis and height adjustment pivot axis.)

[Bill Bryant]

I took my Biesemeyer fence and extension table off, took the motor, blade, and cast iron wings off, took the saw off its stand, and turned the saw over on a thick piece of cardboard. Quite the job.

The rods connecting the trunnions are definitely not in the same plane. I don't have a piece of glass or a small surface plate at home, so I used a ZCI that doesn't rock at all on a flat cast iron wing. I know this isn't an extremely high precision reference, but in this case it was enough to discover that the rods are WAY off.

The front trunnion bracket has thick washers between it and the table top.

I need to go get a huge metric socket to loosen the nuts (21mm? 22mm?) and set things back straight. (Those puppies are on there TIGHT. A large crescent wrench was useless.)

I examined the mechanism for almost an hour, and I still can't see how putting these rods back in alignment is going to fix my problem. All the same, I'll do it just to be sure. But I fear that I'll need to remove the washers under the front trunnion bracket, or replace them with washers of another thickness, or my 45-degree cuts will be off.

I'm disappointed that this saw was this far out, out of the box. Maybe it's time for a General or Powermatic cabinet saw.

[Richard Dragin]

Bill,
Did you try placing the ZCI on the rods through the back of the saw before flipping it over? Also,
did you loosen the bolts from the trunions to the top before attempting to adjust the rods? Once to get the rods planar you will have to realign the blade to the miter slot which has to be done with the saw in the upright position.

[Bill Bryant]

I bought a 24mm socket and had to jump on the handle to break free the blue thread locker put on by the Delta factory to make sure the terribly out of line rods would stay that way. I couldn't have applied enough force with the saw upright to loosen these nuts. They were almost welded on (holding the missalignment in place against all but air-driven impact-wrench type force).

I used the very flat 8" x 8" bottom of my General International tenoning jig to set the rods in the same plane, tightened the nuts back to arm strength (not impact wrench) tightness, and rechecked the rods after tightening.

Rod twisting is now ruled out, and it's time to reassemble the saw, realign the trunnions (I've always had a PALS system on there so this is a snap) and find out what has changed.

[Mike Marcade]

I want to know how this goes. :D

[Dale Lesak]

Bill, I'm sorry I didn't catch this post sooner. You didn't need to do all the removing and flipping stuff. Just take the blade off and from the top side lay a flat plate across the two rods. plate glass, steel plate any thing that is flat. If the plate rocks they are out of adjustment. The best way to loosen the nuts is with a air or electric impact wrench. you need to tweak the rear trunnion until the plate is flat on both rods No rocking what so ever. Now this is where the impact really helps tighten the nuts back up. then you try to do this with a wrench you are pulling it back out of adjustment. with the impact you are twisting the nut not pulling on it. recheck, if your lucky you wont have to do it again.

[Bill Bryant]

you need to tweak the rear trunnion until the plate is flat on both rods No rocking what so ever. Now this is where the impact really helps tighten the nuts back up. then you try to do this with a wrench you are pulling it back out of adjustment. with the impact you are twisting the nut not pulling on it. recheck, if your lucky you wont have to do it again.

All I had was a wrench so I had to make do. It took some trial and error, but the end result is rods that are dead on.

I still don't see how this is going to solve my original problem, but we'll see in a bit when I get the saw turned back upright.

[Dale Lesak]

Bill, recheck from the top after you get it flipped and set where it will be working. One reason for the misalinement is your working with two different pivot points. I had mine set up perfect with the miter slot and could not make a 45º cut without it binding. been there and done that. now that's the first adjustment I check before anything else. Also after you get it back together check that nothing is hitting when you tilt the blade. Mine was the bottom of the extenshion table, the top of the motor was hitting it at full tilt and and the frount was still able to move and twisted it out of adjustment. Hogged out the bottom of the table for clearance and fixed that.

[Bill Bryant]

Bill, recheck from the top after you get it flipped and set where it will be working. One reason for the misalinement is your working with two different pivot points. I had mine set up perfect with the miter slot and could not make a 45º cut without it binding. been there and done that. now that's the first adjustment I check before anything else. Also after you get it back together check that nothing is hitting when you tilt the blade. Mine was the bottom of the extenshion table, the top of the motor was hitting it at full tilt and and the frount was still able to move and twisted it out of adjustment. Hogged out the bottom of the table for clearance and fixed that.

The issue has never been what happens when I tilt the blade. The issue is what happens to a perfectly 90-degree blade, perfectly aligned to the miter slot, when I lower it from 3" to 1".

I just got enough of the saw back together to use PALS, dial indicator, machinist squares, etc. to line up the 90-degree saw blade (now riding on trunnions connected by planar rods) inside .001 with the miter slot. Sure enough (rats! and drat! and double drat!) when I lower the blade (NOT TILT IT!) it moves toward the fence and the back of the blade moves toward the fence more than the front.

This is exactly what I had before. This is exactly what I talked to the Delta service manager about and was told it had to be those trunnion rods.

Right now I'm thinking phooey on Delta.

Now I'm going back down into the basement to take a rubber hammer to my head.

[Mike Marcade]

I think your "exercise in futilty" prescribed by Delta just confirms that the problem is in the lifting mechanism not the rods that connect the trunions.

[David Epperson]

Sure enough it moves toward the fence and the back of the blade moves toward the fence more than the front.

This is exactly what I had before. This is exactly what I talked to the Delta service manager about and was told it had to be those trunnion rods.


OK. This part is key, and hasn't been mentioned before that I saw. But this indicates that your raise/lower pivot trunnion is not parallel with the table top or perpendicular to the miter slot. You need to recheck your shims to correct this.

[Mike Marcade]

If this saw is brand new, I would not even mess with this stuff. Delta should fix this or give you a new saw.

[David Epperson]

If this saw is brand new, I would not even mess with this stuff. Delta should fix this or give you a new saw.
True enough. But when he does get it worked out, I think I know who the new "Go to" guy for saw alignment is going to be. :D

[Mike Marcade]

True enough. But when he does get it worked out, I think I know who the new "Go to" guy for saw alignment is going to be. :D

Yeah just look for the guy with the dent in his head that matches the rubber mallet! :D

[David Epperson]

Yeah just look for the guy with the dent in his head that matches the rubber mallet! :D
At one point in time or another - That would be most of us. :D

[Bill Bryant]

Yeah just look for the guy with the dent in his head that matches the rubber mallet! :D

I'm easy to spot now.