Dado set cutting oversized

[RndmNmbr]

I've recently purchased my first dado stack. It's a Ridge Carbide 8 inch set. I happily put together my first blade assembly per their instructions with the outside plates and four of the 1/8in chippers to make a 3/4in test cut. My cut came out ~0.050 wide. I was expecting it to cut either dead on or slightly under. I wouldn't have been surprised if it was off 0.005 or so, but over a 32nd seems odd.

I started measuring things last night. It appears that the chippers all have thicknesses from 0.125 to 0.127 at their thickest portions. There is a smoother "hub" area at the center of each chipper that is ~0.002 thinner than the 4 arms. The plates on the ends are of the correct thickness and geometry to generate the proper 1/8in cut. (The plate is 0.110 and the tooth is ~0.142 wide.) Soooo ... to me, this means that I should expect the stack to end up perhaps ~0.006 wider than 3/4in. I measured runout using a dial indicator, and I get around 0.0025 TIR just below the teeth, which seems pretty sane.

So here's where things seem to get weird:
If I place the stack on the bench and measure from the bench to the tip of the outside carbide tooth (using a gage block), the stack height seems to be as I expect - just a tiny bit over 0.750. However, when I put the stack in the saw and firmly tighten the arbor nut, it seems to expand and cut ~0.050 wide! If I set the arbor nut as loose as I dare and run another test cut, it now appears to cut ~0.016 wide.

Does anyone have any idea what is going on? Are the outer plates really "cupping" outward when I clamp the stack on the arbor and throwing things out this much!!?? Since the inner "hubs" of the chippers are thinner than the outer portions, perhaps this is possible? Is this normal for a dado stack?

Any and all input greatly appreciated!

 

[creasman]

Something to check. With stacked dados you have to make sure each blade is aligned such that the teeth don't line up with an adjacent blade. You need to rotate each before you tighten the stack down until they sort of interlock. The blades are designed such that the kerfs should overlap slightly to give a cleaner cut.

 

[Michael Mathews]

also, lay a straightedge across each plate and make sure they're all flat. I'd also double check your arbor for runout.

 

[junquecol]

Get a 5/8" bolt, using washers next to head of bolt and nut, build your dado stack on it. Using dial calipers, measure stack to see what width it is. Because this is a new dado stack, check to make sure the have the outers on correctly, right and left. Look at the teeth. (I took a "Sharpie and marked mine, L & R.) The outer side will be slightly higher profile than inside. FYI, you don't have to "crank down" on arbor nut with a dado set. Normal tight is just fine.

 

[Bill Clemmons]

Here's a simple jig you can make to check the thickness of a dado stack before you put it on the saw. In this pic I was trying to get a perfect fit for a T-track I was putting into another jig. Your set probably came w/ spacers/shims to help get a perfect fit. It's easier to try them on this jig than taking blades on/off the saw arbor.

 

[Joe Scharle]

Or, use a router and be done with it!

 

[RndmNmbr]

Thanks for all the ideas! The 5/8in bolt based setup jig idea sounds like a good one! I will definitely mark the left and right plates and make sure I never miss that! I'll check the outer plates for flat when they are not torqued later today.

I checked the runout already below the teeth on the outer plates, and it's 0.002 to 0.003 total indicated runout. That, coupled with the chippers being _slighly_ over thickness, would lead me to expect perhaps a 0.010 in over width cut. I have been quite careful about the carbide teeth not interfering, so I know it's not that.

I had an idea last night: I wonder if I stick some shims in and then tighten if I get the same effect. If the distortion in the stack is due to how the plates and chippers contact each other under compression, perhaps standing them off a hair on shims may eliminate the expansion problem under compression.

What I'm really after is repeatability ... if it's repeatable, I can deal with it. If it's different every time I tighten it down, I'm concerned. On a milling machine, I use a torque wrench to tighten everything from collets to vises to work around this kind of thing, but I didn't expect to worry about it for this.

Do you guys with Forrest and Freud sets see anything like this?

 

[bobby g]

Have you discussed this with the manufacturer?

 

[RndmNmbr]

Have you discussed this with the manufacturer?

I did reach out to them, but have not heard back.

 

[Roy G]

Since plywood seems to be undersized, having your dado cut oversize seems to magnify the difficulty of getting tight joints. Have you tried a test cut with the dado set up for the plywood thickness? I have a dado set but I have to confess I use a router to cut dadoes.

Roy G

 

[junquecol]

Thanks for all the ideas! The 5/8in bolt based setup jig idea sounds like a good one! I will definitely mark the left and right plates and make sure I never miss that! I'll check the outer plates for flat when they are not torqued later today.

I checked the runout already below the teeth on the outer plates, and it's 0.002 to 0.003 total indicated runout. That, coupled with the chippers being _slighly_ over thickness, would lead me to expect perhaps a 0.010 in over width cut. I have been quite careful about the carbide teeth not interfering, so I know it's not that.

I had an idea last night: I wonder if I stick some shims in and then tighten if I get the same effect. If the distortion in the stack is due to how the plates and chippers contact each other under compression, perhaps standing them off a hair on shims may eliminate the expansion problem under compression.

What I'm really after is repeatability ... if it's repeatable, I can deal with it. If it's different every time I tighten it down, I'm concerned. On a milling machine, I use a torque wrench to tighten everything from collets to vises to work around this kind of thing, but I didn't expect to worry about it for this.

Do you guys with Forrest and Freud sets see anything like this?

I always "build " my stack to get desired width. Never trust the printed chart, as it's only a guestamate IMHO

 

[tvrgeek]

Stumpy Nubs has a video on setting up dados.

But something is very odd, Your blades should fit flat face to face if the teeth are not touching. It is also odd all the chippers are the same size. That is supposed to be one of the very best sets out there. For the flex you mention, I have to think something got bent ( destroyed)

Once you figure this out, make a test board that fits your ply suppliers where you record what combo is needed for each. In other words, make your OWN chart. But I found I can't trust Home Despot one week to the next anyway.

 

[Woodman2k]

I have the 10" Freud which I think is the best set available in that size and as long as I make sure no teeth are touching, then the variance is within a few thou'. My set has 4 -1/8 "chippers plus 2 -1/16" chippers plus a large shim set to match metric dimensions and my included chart is often very close. Make sure your plates are all clean and no tooth to tooth contact.

 

[chris_goris]

runout.