Sled and zero clearance questions

[Bill Huber]

I have 2 different questions.

1. I got my sled finished and was testing it for square, I use the 5 cut method with a 6 inch board.
I measured it with my calipers in the inch mode and came up with .2835 on one end of the last cut and .2890 on the other. I then switched to fractions on the caliper and came up with 9/32 (.28125) on each end.
So should I say that is good or do I need to work at getting it closer and just use the inches on the calipers?

Some times these number just get all screwed up in my brain, that is I think I have one. My Dad always told me that when the Good Lord was passing out brains, I thought he said trains and said I had one.... and I do have a train.:D

2. I made a zero clearance insert for the TS and put it in but it sure make a lot of noise when it is running. It is made out of Phenolic and I raised the blade thought it 3 or 4 times with no change. Do I need to file the opening just a little wider or is there something else that I have done wrong.


Thanks....

[Bob Wingard]

Myself, I'd try to improve it .. .. but if you're happy with the joints it produces, that's fine. From your numbers. it sounds like it's "out" by approx. 4 thousandths per running foot .. .. not bad, but not great either. Making it better depends on how difficult it will be to adjust your fence .. .. if you've provided for easy adjustment, I'd go for it .. .. if not .. be happy with what you've accomplished.

[Doug Shepard]

Dont have an answer for #1 but the ZC insert noise is somewhat normal on a new one and should go away in time. One thing you can do though is put a dado set shim or piece of paper on the arbor then re-raise the blade. That will open it up a few thou on one side.

[Bill Huber]

Myself, I'd try to improve it .. .. but if you're happy with the joints it produces, that's fine. From your numbers. it sounds like it's "out" by approx. 4 thousandths per running foot .. .. not bad, but not great either. Making it better depends on how difficult it will be to adjust your fence .. .. if you've provided for easy adjustment, I'd go for it .. .. if not .. be happy with what you've accomplished.


That is what I needed to know, I guess I really need to keep working at it. I want it write.....

Thanks.

[Bill Huber]

Dont have an answer for #1 but the ZC insert noise is somewhat normal on a new one and should go away in time. One thing you can do though is put a dado set shim or piece of paper on the arbor then re-raise the blade. That will open it up a few thou on one side.


I knew someone would have the answer, I will stick a shim in it and brign it up again.

Thanks.....

[Stephen Pereira]

Forgive my ignorance but what is the 5 cut method?

[Bill Huber]

Forgive my ignorance but what is the 5 cut method?

Well you take a square board and cut one side, then rotate the board clockwise and put side one against the fence. Make cut 2 and rotate the board clockwise and put that side against. Then make cut 3 and turn the board and then cut 4 and turn the board. At this point you are back to side 1 being next to the blade, cut a 1/4 inch or so off and measure each end.


http://www.thewoodshop.20m.com/calibrate_sled1.htm

http://www.thewoodshop.20m.com/five_cut_method_swf.htm

[Phil Thien]

Myself, I'd try to improve it .. .. but if you're happy with the joints it produces, that's fine. From your numbers. it sounds like it's "out" by approx. 4 thousandths per running foot .. .. not bad, but not great either. Making it better depends on how difficult it will be to adjust your fence .. .. if you've provided for easy adjustment, I'd go for it .. .. if not .. be happy with what you've accomplished.

Isn't he off by .00275 (two-thousandths) per foot?

(.2890-.2835)/(6*4)

That isn't bad. About as thick as a thin piece of paper over a foot. AND, you have to take into consideration that you have five chances to screw the process up. That is why I prefer just using a larger workpiece and a good, trusted square to set my sled up.

[Andrew Williams]

I would adjust and try again. It is not hard to get much closer to square than you have gotten. You will appreciate the difference when making mortise and tenon joints.

[frank shic]

Bill, I would leave your crosscutting sled just the way it is right now. You've managed to get it pretty accurate which is no easy feat with a sled as I've learned from personal experience. My exaktor sliding table is off by 0.005 but you know what? I haven't lost any sleep over it yet! Your zero clearance insert may be rattling if it doesn't fit snuggly in the throat plate. The solution would be to shim underneath or on the sides to get it fitting tight.

[Gary Keedwell]

I agree..don't beat yourself in the head too much. There are too many variables to deal with. Your sled might be as good as you can get because your miter slots could account for some of it.

Gary K.

[Gary Keedwell]

Isn't he off by .00275 (two-thousandths) per foot?

(.2890-.2835)/(6*4)

That isn't bad. About as thick as a thin piece of paper over a foot. AND, you have to take into consideration that you have five chances to screw the process up. That is why I prefer just using a larger workpiece and a good, trusted square to set my sled up.

Actually, if he is making 5 cuts and then measuring...wouldn't you have to divide by 5...then the sled would be off by only .0008. It is a accumulative thing.
Gary K.

[Art Mann]

The 5 cut method is good because it does amplify the error. In this case, I think it is going to be very difficult to improve on the current adjustment. You could make it worse. What you should do is repeat the experiment a couple of times. I bet you will find that the repeatability error of your results will be larger than the original error you measured. You are not dealing with machine tools.

[glenn bradley]

2. I made a zero clearance insert for the TS and put it in but it sure make a lot of noise when it is running. It is made out of Phenolic and I raised the blade thought it 3 or 4 times with no change. Do I need to file the opening just a little wider or is there something else that I have done wrong.


Thanks....

What is holding your ZCI in place? Single screw, spring tab, gravity? If gravity I would confirm that it is evenly supported and then perhaps use some tape (or something fancier if you like) to shim the side for a tighter fit.

P.s. Remember to divide your 5-cut error by 4; that's your error per pass.

[Phil Thien]

Actually, if he is making 5 cuts and then measuring...wouldn't you have to divide by 5...then the sled would be off by only .0008. It is a accumulative thing.
Gary K.

I think my math is correct, he is off approx. .003" over 12".

I do stand by my previous statement that I'd rather my Starrett square tell me my workpiece is square rather than using a caliper on a cut-off.

FWIW, I can EASILY see .003" over 12" with my Starrett. I can see .001" over 12". I really see no need for the five cut method when my Starrett allows me to quickly and directly check my workpiece, and I never need better than .001".

I guess the five-cut method would eliminate the chance that my Starrett is inaccurate. But I check it w/ other squares from time to time.

If you think .001" is hard to see w/ a square, try this: Make a square cut (I know, chicken and the egg) on the end of a board. Hold your square so the blade is on the end grain. Hold up to the light, check for light passing gaps (should be none).

Now, put a .001" feeler gauge between the wood and the blade and hold it up to the light again. That .001" gap passes quite a bit of light, doesn't it?

I don't need any more accurate than that.

[Bob Feeser]

Bill,
.2890 - .2835 = .0055 That is off by 5 1/2 thousandths. For critical applications, getting cuts within a thousandth is ideal. The fact that you are using a sled, and the 5 cut method, you are getting great results because if I am reading this correctly, you are getting a single pass, accurate to one thousandth.
Considering the following;

It's not that the sled is getting cuts that far off, it is because the cuts are amplified by referencing off of the previous cut x 4, before making your 5 cut for the test piece. That then indicates to me it is possible that you are getting cuts in a single pass from your sled, that are accurate to within a thousandth of an inch. Isn't that all you need? "Strive for perfection, settle for excellence" I always say. When I am working on a project, and start getting anally critical of myself, I calm it down by adopting a German accent, and start walking around, calling myself, "Dr Goodenough". :) :) :) Absolute perfection is a trap, ever elusive, always just outside of our grasp. If the defects give it character, then does that mean that little defects give it little character. :confused: Oh boy I am full of them today.Some of the top table saw fence manufacturers only guarantee to 1/64th accuracy. I get ridiculed by friends for going to the thousandth level on my table saw. Considering that you are using the 5 cut method, which is new to me by the way, if I read it right, you are amplifying any errors by using the prior reference to make the next cut. If it passes that test, you know it is right. Correct me if I am wrong, but isn't that like whispering something in someones ear, then they whisper it into someone elses ear, and by the time it gets around the room it is comical to hear how it comes out. It is never even close to the original message. ;)

A little shop talk: The 5 cut method seems to be ideal for truing up your settings. When doing projects, I purposely avoid that amplification, by referencing the jointed side against the saw fence, then cutting on the opposite side, thereby creating two perfectly parallel pieces. Then I use the newly cut sides to make my crosscuts.

Pertaining to the zero clearance insert, I personally am annoyed when using a new insert, that has not been sweetened up, by ever so slightly cleaning the new edge with a piece of fine sand paper wrapped around a paint stirring stick. The miniscule thickness you are taking off, makes all of the difference between super quiet, and irritating noise. If that few thousandths that you are opening up the blade trough is a factor, you must be cutting some very narrow pieces :eek: to fall in that thousandth gap between the blade and the insert. :) Maybe you are concerned about keeping the saw dust mostly under the table. It might be interesting to note that Forrest WWII blades, in the instructions, they tell you to skew the fence setting so the back edge of the blade is wide by .001 to allow for the teeth kissing the exiting workpiece. Why I brought that up is because the kiss of the blade against your workpiece, or the .002 kiss on the waste stock if you do what FWWII instructs you is going to create more airborne dust than that. (The thousandth, or up to it that is naturally occuring, plus the .001 adjustment equals .002 or less) The back of the blade is moving upward when moving over that area, sending the dust from that .001 shaving into the air. Phew, boy are we getting critical here. :)

By the way is it a cross cut sled you are referring to? If you are using a cross cut sled and getting those kind of results, I want to know what method you are using to secure the runners to the bottom. Try as I may, using wood for runners, I am getting seasonal variations, and invariably have to hold the sled, at a slight skew to make sure it presses against the edge that is the totally accurate one. Next time around, I think I will start experimenting with some phenolic material to make the runners on the bottom of the sled that glide in the miter slots. I would be curious as to how others have solved the "moving" runner under a table saw sled problem.

[Art Mann]

I calculate 0.003"/foot also. You don't divide by 5. You divide by 4. The first cut isn't in reference to anything. The first cut is just to get a straight edge from which to reference the second cut. Otherwise it would be the 4 cut method.The OP should have used 1 foot square or larger matrial to start with to get better accuracy of the measurement.

I need to point out something about the 5 cut metod that few people think about. Repeatability. Several months ago, I did the 5 cut method on my sled. Actually I did the exact same test 5 times in a row. I took great care making the cuts and used 15" square hardboard rather than plywood because it is more consistent. I used a good set of digital calipers accurate to 0.0005". I am not bragging, but I have done a lot of precision work and am probably at least as consistent as the average guy at technique. The results I got varied by 0.004"/foot over the 5 trials. I think if others did the 5 cut method several times, they would see a similar variation. That is probably about the best accuracy you can expect with this method. There is a limit to what you can do with equipment made for machining wood.

[David DeCristoforo]

"... back to side 1 being next to the blade, cut a 1/4 inch or so off and measure each end...."

Measuring is always good. But you can get a better idea of how much a measurable error will relate to "real world" conditions by cutting the 1/4" falloff piece in half and comparing the two ends. Sometimes when you can see and feel the diff, it can tell you more than a caliper or ("god forbid!) a tape measure reading.

[Gary Keedwell]

.0008 a cut or .001 a cut what's the difference? If you can make 4 cuts and measure diagonally and be off only ten thousands, your doing great. A person trying to do better then that...needs to get a life.:)
Gary K.

[Art Mann]

.0008 a cut or .001 a cut what's the difference? If you can make 4 cuts and measure diagonally and be off only ten thousands, your doing great. A person trying to do better then that...needs to get a life.:)
Gary K.
:confused:
Are you talking about the 5 cut method? You don't measure any diagonals with this method. You measure the width of narrow strips you rip on both ends. Maybe I am misreading your comment.

[Bill Huber]

Worked on the sled today, like 4 or 5 hours the best I could get was .004 to .001 this is over a 1 foot board, 5 cut method 5 different boards.

Board one, 5 cutes and checked and had .002 end to end.
Board two, 5 cutes and checked and had .004 end to end.
Board three, 5 cutes and checked and had .001 end to end.
Board four, 5 cutes and checked and had .002 end to end.
Board Five, 5 cutes and checked and had .001 end to end.
I was trying very hard to put the same pressures on the sled in the same way each time. So I think that is just about as good as I can get it.

The ZCI. I put a shim behind the blade and raised up up 2 or 3 time and then took it out and that did it, nice an quite

Thanks for all the comments.....

[Gary Keedwell]

I admire your tenacity.;) :)

Gary K.

[Eugene A. Manzo III]

But when does woodworking turn into machining ?:confused:

[Bob Wingard]

The 0.001" and 0.002" seem very repeatable .. .. the 0.004" could have just been a spec of sawdust in your way. Granted, it took awhile, but aren't you proud of what you were able to accomplish ??

Congratulations !! !! !!

[Bob Wingard]

But when does woodworking turn into machining ?:confused:


When you start cutting dados in a Bridgeport, as did my late brother. He built an entire deck this way, but, granted, it WAS before he had ANY woodworking tools !! !! !!

[Gary Keedwell]

When you start cutting dados in a Bridgeport, as did my late brother. He built an entire deck this way, but, granted, it WAS before he had ANY woodworking tools !! !! !!

:D :D I've milled my share of dado's in a Bridgeport. Does a clean and accurate job. lol:o

Gary K.

[Bill Huber]

By the way is it a cross cut sled you are referring to? If you are using a cross cut sled and getting those kind of results, I want to know what method you are using to secure the runners to the bottom. Try as I may, using wood for runners, I am getting seasonal variations, and invariably have to hold the sled, at a slight skew to make sure it presses against the edge that is the totally accurate one. Next time around, I think I will start experimenting with some phenolic material to make the runners on the bottom of the sled that glide in the miter slots. I would be curious as to how others have solved the "moving" runner under a table saw sled problem.

Bob, this is what I did.....

I started by getting the blade square to the slot with a digital dial indicator on a super bar with the MasterGage and plate.
Then used the MasterGage and dial indicator to get the fence square, I looked out on this one, dead on in 3 tries .000 end to end.

I just used some Red Oak I had and planed to fit a little less then the height and really close to the width. Then with 400 sand paper using the top of the saw as a flat surface, sanded the width to just fit, they would drop in I did not have to press them in.

Cut my MDF to the size I wanted and and moved the fence over so the MDF was positioned where I wanted it to be.

I cut strips of paper and place under the rails until I get it high enough to clear the top of the saw by about 1/6 of an inch or so.
Then I cut paper strips and put on the side of the rail to make it tight. I read that it should be put on the same side of each rail but I shimmed it on the end side of each rail and it was tight, I had to push the rails down into the slot.
I put glue on the rails and then put the MDF against the fence and lowered it on to the rails. Put 50 lb.s of stuff on top and let set for a day.

Then I pried the rails up, took out the shims and tested the fit, just to tight. I started pushing it back and forth and it started to loosen up a little, then I waxed it really good and that is it.

Now a year from now I wonder just how loose it will get. I wish I had used some other type of material for the rails.

[Jeff Raymond]

.0008 a cut or .001 a cut what's the difference? If you can make 4 cuts and measure diagonally and be off only ten thousands, your doing great. A person trying to do better then that...needs to get a life.:)
Gary K.


I'd go ahead and make something with your machine and if it looks good it is good and move on to the next thing!

[Art Mann]

Worked on the sled today, like 4 or 5 hours the best I could get was .004 to .001 this is over a 1 foot board, 5 cut method 5 different boards.

Board one, 5 cutes and checked and had .002 end to end.
Board two, 5 cutes and checked and had .004 end to end.
Board three, 5 cutes and checked and had .001 end to end.
Board four, 5 cutes and checked and had .002 end to end.
Board Five, 5 cutes and checked and had .001 end to end.
I was trying very hard to put the same pressures on the sled in the same way each time. So I think that is just about as good as I can get it.

The ZCI. I put a shim behind the blade and raised up up 2 or 3 time and then took it out and that did it, nice an quite

Thanks for all the comments.....

Excellent! You certainly can't improve on that.

[Bob Feeser]

Bob, this is what I did.....

I started by getting the blade square to the slot with a digital dial indicator on a super bar with the MasterGage and plate.
Then used the MasterGage and dial indicator to get the fence square, I looked out on this one, dead on in 3 tries .000 end to end.

I just used some Red Oak I had and planed to fit a little less then the height and really close to the width. Then with 400 sand paper using the top of the saw as a flat surface, sanded the width to just fit, they would drop in I did not have to press them in.

Cut my MDF to the size I wanted and and moved the fence over so the MDF was positioned where I wanted it to be.

I cut strips of paper and place under the rails until I get it high enough to clear the top of the saw by about 1/6 of an inch or so.
Then I cut paper strips and put on the side of the rail to make it tight. I read that it should be put on the same side of each rail but I shimmed it on the end side of each rail and it was tight, I had to push the rails down into the slot.
I put glue on the rails and then put the MDF against the fence and lowered it on to the rails. Put 50 lb.s of stuff on top and let set for a day.

Then I pried the rails up, took out the shims and tested the fit, just to tight. I started pushing it back and forth and it started to loosen up a little, then I waxed it really good and that is it.

Now a year from now I wonder just how loose it will get. I wish I had used some other type of material for the rails.

Bill,
Thanks for your reply. I was getting great results when I built the sled, am getting good results still, but I have to torque the sled while I am pushing it, to keep it toward one edge that is accurate. The problem is that the miter trac that fits into the slot, shrunk a little bit over time, and now their is a little slop in the fit. I know Norm on the NYW oversized his a bit, and kept it that way, starting out with a squeaky fit. I was wondering if I could get some phenolic plastic, or something even metal that would not change with the seasons? Sounds like you have a great system going for you.
Bob