Table saw parallelism question

[vavet]

I have a ridgid 4516 portable table saw. I just finished building a table for it and decided to check the parallelism of the miter gauge slots to the blade. It’s about 0.020 out from the front to the back of the blade. I followed the instructions in the manual, which calls for loosening the motor mounting screws and adjusting the eccentric screw to dial it in. I got it down to about 0.014” but it was close to the limit of its adjustment in that direction before I even did it.

Is there a maximum amount of “pinch” you can have before kickback really becomes a problem?


I’m not building fine furniture here, but I also don’t want to be working with a dangerous and inaccurate tool.

 

[tarbellb]

You will have witness marks then burn marks long before kickback in my experience.

 

[The Cobbler]

normally you parallel the fence to the blade . the miter gauge is not as critical. but 0.014 is pretty close I'd say .

 

[jg4660]

normally you parallel the fence to the blade . the miter gauge is not as critical. but 0.014 is pretty close I'd say .

^^Agree.

 

[Firebrick43]

I don’t, well sort of. You can have some if the distance from the slot to blade increases as you push through. If it decreases you will never get straight cuts using the miter gauge or a sled. Also you won’t have any Success with tennon jigs, finger joints, or dados. I would never except that unless I was just doing rough carpentry. The fence has to be parallel

 

[McFarmer]

Is the front of the blade closer to the slot on the left side?

Both slots the same ?

 

[dscheidt]

normally you parallel the fence to the blade . the miter gauge is not as critical. but 0.014 is pretty close I'd say .

14 thou is a huge difference. No jig (cross cut sled, tenon jib, etc) will work with that big a difference, if you care about accuracy. Of course, this is a job site saw, and not suitable for precision work, so it might not be that big a deal.

OP: you are measuring to the same tooth on the blade right? Most saw blades are not perfectly flat, and most have alternating left/right teeth.

 

[rsanter]

Good enough
Now set the fence parallel to the blade and dial the miter gauge to the blade as well

 

[larry_g]

What other tests have you performed?

Have you rotated the blade while keeping the indicator steady to prove a flat blade?

If the above fails, check the flange on the arbor, Is it straight and does the indicator show any wobble there?

Are you sliding the indicator along the blade or measuring and then rotating the blade to measure the same tooth?

Indicate on the blade while raising and lowering the blade, is it true?

There are a few variables in a saw that all have to be correct. Don't get to hung up on one ignoring the others. Have you looked for a bit of debris between the blade and the arbor flange? Have you checked this with a different blade?

lg
no neat sig line

lg
no neat sig line

 

[1cargarage]

You don't actually want the fence to be perfectly parallel to the blade. Ideally, the distance between the blade and fence will be slightly smaller at the operator's side.

Also, when referencing the blade during measurements, make sure to have the blade projected as far as possible out of the table (height). Also, be sure to reference from the same blade tooth when measuring from the front and back to ensure readings aren't skewed from blade runout.

 

[BukitCase]

Agree with 1cargarage; and by "slightly smaller at operator end", I have my older PM66 with full table extension, Beismayer fence splayed by .003-.004" further AWAY from the blade at the outfeed side. Measuring with digital gauge, high blade, same tooth used front and rear, and multiple checks.

Cuts come out looking like they're sanded, perfectly parallel, no binding, etc, nearly new TCG blade, I have to look at the blade to know when 3/4 oak ply enters the blade - virtually zero change in sound.

I wouldn't expect that performance with ANY jobsite saw, but I would think that one should have anywhere from zero to a few thou splay at outfeed, otherwise it WILL bind/kick back/burn, etc... Steve

 

[vavet]

Thanks to everyone who has offered suggestions.

Is the front of the blade closer to the slot on the left side?

Both slots the same ?

On the right side, the outfeed end is farther away from the blade. On the left side, the outfeed end is closer to the blade.

14 thou is a huge difference. No jig (cross cut sled, tenon jib, etc) will work with that big a difference, if you care about accuracy. Of course, this is a job site saw, and not suitable for precision work, so it might not be that big a deal.

OP: you are measuring to the same tooth on the blade right? Most saw blades are not perfectly flat, and most have alternating left/right teeth.

I marked a tooth with a sharpie and I am measuring to the same tooth each time with the blade in the full upright position. I am measuring the farthest ends of travel that I can.

What other tests have you performed?

Have you rotated the blade while keeping the indicator steady to prove a flat blade?

If the above fails, check the flange on the arbor, Is it straight and does the indicator show any wobble there?

Are you sliding the indicator along the blade or measuring and then rotating the blade to measure the same tooth?

Indicate on the blade while raising and lowering the blade, is it true?

There are a few variables in a saw that all have to be correct. Don't get to hung up on one ignoring the others. Have you looked for a bit of debris between the blade and the arbor flange? Have you checked this with a different blade?

lg
no neat sig line

lg
no neat sig line

I haven't done any of the above. TBH, this is the first time I've done this with any table saw. I've had this one about 8 years. Perhaps I'm putting the cart before the horse with this test. As suggested above, maybe I should check the fence parallelism before I check the miter slots.

You don't actually want the fence to be perfectly parallel to the blade. Ideally, the distance between the blade and fence will be slightly smaller at the operator's side.

Also, when referencing the blade during measurements, make sure to have the blade projected as far as possible out of the table (height). Also, be sure to reference from the same blade tooth when measuring from the front and back to ensure readings aren't skewed from blade runout.

Agree with 1cargarage; and by "slightly smaller at operator end", I have my older PM66 with full table extension, Beismayer fence splayed by .003-.004" further AWAY from the blade at the outfeed side. Measuring with digital gauge, high blade, same tooth used front and rear, and multiple checks.

Cuts come out looking like they're sanded, perfectly parallel, no binding, etc, nearly new TCG blade, I have to look at the blade to know when 3/4 oak ply enters the blade - virtually zero change in sound.

I wouldn't expect that performance with ANY jobsite saw, but I would think that one should have anywhere from zero to a few thou splay at outfeed, otherwise it WILL bind/kick back/burn, etc... Steve

Most of my cutting with this saw is for shop fixtures to be used for other things. I would like to build a couple pieces for the house, but they'll be functional, not decorative - a mudroom shelf that has a function and size unique to us. I also want to build a corner shelf for our breakfast room to be used to keep games and other commonly used items out of sight but easily accessible. It also has a unique footprint need that we've not been able to find available commercially. I know this is not the fine machine we would all want if we were earning a living with every day. I'm not cutting dovetails. My plan is to use my Kreg jig to assemble the pieces I've mentioned. I just need to have a safe tool and a reasonably accurate tool.

I think the next step, unless I get a better suggestion, is to check the alignment of the fence instead of the miter slots.

 

[rlitman]

...Is there a maximum amount of “pinch” you can have before kickback really becomes a problem? ...

Yes. Measure how much the tip of your carbide teeth extend past the plate of the blade. Bad things happen when the wood being cut touches the plate of the blade and not just the teeth.

normally you parallel the fence to the blade . the miter gauge is not as critical. but 0.014 is pretty close I'd say .

The correct procedure is to first parallel the arbor (blade) to the miter slots. THEN parallel the fence to the blade (or slots). On a tabletop saw where fence parallelism is a moving target, this procedure isn't that critical, but on a cabinet saw, taking your time here and doing it right really pays off.

You don't actually want the fence to be perfectly parallel to the blade. Ideally, the distance between the blade and fence will be slightly smaller at the operator's side...

This is dead wrong.

 

[JohnM45]

When I got into wood working quite a few years ago, I remember reading a lot of articles on setup. I always enjoyed Fine Woodworking magazine and Popular Woodworking - many of their "setup" articles suggested the fence to be set .001 to .003 inches wider at the outfeed side of the blade. I've always kept mine around .001 and don't have much issue at all.

Depending on how much precision you want will depend on the setting. Also, some saws aren't going to hold settings very well over time. I have an early-2000's Delta Unisaw with Biesemeyer fence and its rock solid...set it, lock it down and you're good to go for a long time.

 

[RTM]

I agree with those who say set the blade parallel to the table’s slots, then set the fence. The slots are fixed, the fence is adjustable, relative to the table.

 

[CallumRD1]

The mitre slots are the absolute reference point of a table saw. They are fixed and cannot be adjusted. You then true the blade relative to the slots and the fence relative to the slots.

 

[Showkey]

Post #9 from Larry G has it right.

Tuning a table saw is a complete process in the proper sequence. One adjustment will effect others.

Often: Starting with blade runout.....example if blade runout is .005” or 010” measuring off the blade has a built in tolerance error.

There are dozens of instructions and videos on the topic.

Google “tuning the table saw”

If you really want the meter gauge cuts to be accurate build a sled. ( especially on low cost contractor saw).

 

[JohnM45]

The mitre slots are the absolute reference point of a table saw. They are fixed and cannot be adjusted. You then true the blade relative to the slots and the fence relative to the slots.

The mitre slots are references, but I'd never adjust my fence to the slot because they have nothing to do with each other. My fence and my blade have a lot to do with each other.

Yes, make the slots and blade as close to exactly parallel as possible, but adjust the fence to the blade.

In a perfect world if you make the slot and blade parallel, you could adjust the fence to the slot. But at the end of the day, the blade has the last say in the matter....so adjust the fence to the blade.

 

[1cargarage]

There is indeed a sequence for proper table saw setup. Here is a link to a comprehensive article from WOOD magazine just to throw something published into all this conjecture.

https://www.woodmagazine.com/woodworking-tips/techniques/outfitting-woodworking-shop/tune-up-your-tablesaw

 

[Hytekrednek]

I have always been taught to align everything to the miter slots. Align the blade to the slots, and align the fence to the slots. Slots are not adjustable, the fence and arbor are. Also, 2 ways to skin this cat... I prefer to have my fence perfectly aligned to the slots with no run-out of the fence. Others like to have about 4 thou of fence run-out. Reason I prefer a perfectly aligned fence is any amount of run-out can result in a slight angled cut. It may only be 1/4 of a degree, but it will be there in many cases. For high precision work I suggest no fence run-out. For all other work 4 thou of run-out from front to rear may be best for you.
If in doubt, try both ways.

I use a Delta Unisaw 3 horse with 52" unifence with aftermarket fence rail.

 

[Hytekrednek]

also, make sure your fence is straight, your blade is not warped, and the miter slots are straight and tight before you adjust your saw. Before you do anything... UNPLUG IT FIRST!!!

 

[rlitman]

...Reason I prefer a perfectly aligned fence is any amount of run-out can result in a slight angled cut...

Not quite. Having the fence aligned at an angle to the blade could create a cove cut, but it will not cause the piece you are cutting to taper along it's length.

 

[The Cobbler]

many people are talking setups for precision saws, which this is not. keep that in mind!
many of the jobsite saws don't have the adjust-ability that precision table saws do.

 

[er3456df]

Not quite. Having the fence aligned at an angle to the blade could create a cove cut, but it will not cause the piece you are cutting to taper along it's length.

He meant bevel angle, not taper. You're right that it'll be a small amount of cove/arc, but it'll look like a bevel angle when you fit the parts together.