Two (or more) issues with setting accurate angle for dado stack

[Michael Jasper]

Hi -

So I'm trying to set up a dado stack to cut a 10 deg angle (for a ladder shelf). I've run into a couple issues.

The first comes down to needing a true reference angle. I set my incra V120 miter gauge to 10 deg . I also set a T-bevel gauge to the same angle, using a metal protractor (the one in the photo below).

https://m.media-amazon.com/images/I/61fTUDGje+L._AC_SL1500_.jpg

The two are very close but not the same - probably a degree off. It isn't obvious to me which is the more accurate. I have high quality squares to use for my reference right angle - but I'm not sure what to use for an arbitrary angle like this. What do most people consider the best reference or protractor?

The second problem - once choosing one of the above to set the angle it appears that one side of the dado stack is at the chosen angle but the other side is at a slightly different angle. This is a Freud dado blade on a Dewalt DWE7491RS Table saw. I was careful to make sure the chippers and outside blades are offset from each other. What would cause one outside blade to end up at a different angle?

[Cary Falk]

My thought is if you are sure that the 90 degree is set correctly on the V120 then the 10 degree on the V120 would be the most accurate. I'm not sure that the 10 degree is ultra critical on the shelf unless you are going to display marbles. As for the dado blade, maybe there is something on one of the blades or maybe a warp. Once it spins up and starts to cut, the dado in the wood won't know but might be slightly wider than you thought because of the wobble. I would take the dado blade off and restack it and see what happens. I only check square with one blade during the intial setup of the saw and never check at the blade after that, only the wood, unless I find a problem with the final cut.

[Mel Fulks]

I remember years ago having that problem. I think it turned out that the pieces had to have some thin shims between them .

[Bill Dufour]

Get a ground angle block set. Even a cheap set will be more accurate then a under $100 gauge. A cheap Amazon set says +- 20 seconds, Mitoyo says +- 40 seconds while Starttet says +- 2 seconds.
Bill D

[Edward Weber]

The second problem - once choosing one of the above to set the angle it appears that one side of the dado stack is at the chosen angle but the other side is at a slightly different angle. This is a Freud dado blade on a Dewalt DWE7491RS Table saw. I was careful to make sure the chippers and outside blades are offset from each other. What would cause one outside blade to end up at a different angle?

Make a test cut.
While the dado stack may measure slightly different while sitting static, the cut has to have two parallel walls.
Measure the cut angle and go from there.

[andrew whicker]

I use a gage called MiterSet that has pins and T slot. You put the pins at the angle you want and then set your sliding miter of choice to the pins.

It's pretty handy. I use that as my 'true reference'.

Of course, that's assuming your table saw slot is parallel to your blade.

[Bill Dufour]

One degree is 6o minutes of 60 seconds each so one degree is 3,600 seconds. Measuring to +-40 seconds seems plenty accurate for wood to me.
BILL D

[Cameron Wood]

I've been happy with an inexpensive digital angle finder- advertised to .3˚ accuracy.

Those little metal ones are OK for checking the bevel angle of a chisel.

Another approach is to cut an angle block with a miter saw or equal. It shouldn't really matter if the cuts are 10˚ or 9.82˚ as long as they are all the same.

[Wes Grass]

Are the blades not parallel on the arbor, or are the sides of the groove you cut not parallel? If it's the groove something has to be moving as you're cutting, and I'd think you'd only see this if you draw the work back over the dado instead of thru and remove it behind the blade.

Miter gauge bar fits the table slot?

[Edward Weber]

Are the blades not parallel on the arbor, or are the sides of the groove you cut not parallel? If it's the groove something has to be moving as you're cutting, and I'd think you'd only see this if you draw the work back over the dado instead of thru and remove it behind the blade.

Miter gauge bar fits the table slot?

This is what I was getting at.
Once the wood passes through the blade, both sides will be parallel unless something has moved during the cut.
The cut may be at the incorrect angle and/or width, but the sides should be parallel regardless of the dado setup.

[Tom M King]

Or the blade gets out of perfect alignment with the fence as the arbor tilts.

The Wixey digital cubes have been accurate enough for anything I do like that. Zero it on the saw top, and stick it to the blade with the built in magnets. Worked easily and perfectly for a pyramid roof.

[Michael Jasper]

Lots of great input - thanks!
Didn't know about angle gauge block sets - there's an igaging set that's probably plenty accurate for my needs.

To answer one of the questions - I had been pulling the wood back over the dado stack rather than removing it after pushing through. Bad technique?
I get that the groove should still be parallel even if the blade is at a different angle...although in this case, the groove is wider than the stack so I have to do it in multiple passes. Probably the different angle of the two plates is not related to the difference between the protractor and miter gauge.

As far as the need for accuracy rather than precision - I think part of the issue, which I realize was a mistake, is I calculated the shelf depths based on the planned angle and cut them, rather than assembling the frame and then measuring the actual depth from front to back. So as a result, setting the uprights at the correct angle was important since I based depth calculations on that angle.

I struggled with the right order to go about this whole thing, and how to get the dado grooves at the right angle, in the right location, matching on front and rear and then mirrored on both sides. I'm half way through the whole thing. Just trying to eliminate possible sources of error.

[Wes Grass]

If there's slop in the miter bar to slot, pulling it back might change the angle slightly. The heavy side of the piece having more drag on the table.

If the blade isn't parallel to the miter slot, the bottom of the groove will still be flat, but the sides will be elliptical. But in a shallow cut like I'm thinking you're doing, it would never be noticed.

[Michael Jasper]

Thanks Wes, I need to think through what you said to visualize all of that. What did you mean by the "heavy side of the piece having more drag"? I should have said that I'm doing half lap joints for stretchers between the front and rear legs of the ladder shelf. The rear support is vertical so 90 deg angles, but the front upright is at an 80 degree angle so I need to make the half laps 10 degrees off. I've been able to adjust the miter bar pretty well to remove slop - but I'm sure it's not perfect.

[Wes Grass]

Ah, for some reason I was thinking you're making a fancy ladder, and you're cutting notches for the rungs. In that case you'd be moving a long piece across the table for each notch, and on a small saw like that, any saw really, the weight imbalance is going to require holding the piece down which makes it drag on the table in a non-uniform manner. So my theory was any play or flex in the miter gauge could tweak the angle slightly.

'Shelves' ... duh. Already cut to width so you have to match these to them. How committed are you to an exact size? Maybe narrow the entire thing a little and trim the shelves to match?

Here's a crazy idea I had for a book shelf some years back. The front 'legs' are arched so the top shelf is shallower than the bottom. Never really figured out how to put it together, let alone how I was going to make the parts.

492193

[Michael Jasper]

Huh - cool design!

I think I'm okay with shelves now - I'll just need to cut a few more pieces for the glue up to make them deeper.
To be honest, I also made the mistake of making what I thought was a scale drawing and determining shelf depth from that - but realized that my angles on the drawing probably weren't accurate enough to do it that way.

Seems like the right way to do this would have been to create the two side frames with the lap joints and only then determine the depth of shelves needed. The shelf I'm building is something similar to this.
492194

[Bill Dufour]

I have learned to design things so they are not flush at the joints. Either inset or sticking out maybe 1/16-1/8". If you miss by a little no one will see the difference. If you try for flush the human eye can see a mismatch of 1/1,000". If you build it perfectly flush the wood will continue to dry and pull back at least 1/1,000" in a decade or less.
BilL D

[Steve Demuth]

It's extremely simple to cut yourself a template at the particular angle you need, and use that rather than a general purpose tool. For a ten degree angle, lay out a piece of masonite or thin plywood that is 10" by 1.76 (1 and 25/32) with square corners, then carefully cut it in half on a diagonal using a sharp utility knife. You can do the same for any other angle by computing the short dimension as tan (angle) * 10".

[Michael Jasper]

Ah, another approach I hadn't thought of - that makes a lot of sense. Should have posted before I started cutting.