Table Saw Sled Accuracy - Help Please!

[Charlie Woody]

I tried William Ng's "5 cuts to a perfect cross cut sled" http://www.youtube.com/watch?v=UbG-n--L ... ature=plcp

Just come in from the workshop after using this method to square up my sled fence. Unfortunately I did not get the result I was hoping for!

My last measurements (all in mm) were A 40.4 - B 40.3 / 4 654 x 1143 = 0.04. However just looking at the work piece I could see it was not square so I measure the angles which were 89.7, 89.1, 89.7, 90.5 = 359! I don't know where I have "lost" 1 degree!!!

Can somebody help please?

 

[baldpate]

However just looking at the work piece I could see it was not square so I measure the angles which were 89.7, 89.1, 89.7, 90.5 = 359! I don't know where I have "lost" 1 degree!!!

Hi Charlie,

I'm not much of a woodworker, but I am a trained mathematician. I can only think of two ways in which the internal angles of a four-sided plane figure can add up to less than 360 degrees:
a) one or more of the four sides is not straight - if the sides are bowed inward, the total internal angle will be less than 360.
b) your angle measuring device is inaccurate (or you've misread it, but that's another story!).

Is the 'work piece' you refer to the test piece which is cut 5 times to test and correct the fence angle? The method which Ng offers seems to be critically dependent on several things:
- the sled must run a exactly the same angle to the blade on every cut : any slop in the fit of the runners in the slot(s) will result in randomness in every cut, with unpredictable results.
- the fence must be absolutely flat : in particular, if it is in the slightest bit convex on the blade side, you might not be presenting the work piece to the blade at the same angle on every cut.
Either of these might lead to a cut which isn't exactly straight, although I suspect you would have 'felt' this as you made the cuts.

All a bit theoretical, but food for thought.

Chris

 

[devonwoody]

I assume you are not holding the timber sample being cut by hand when passing the blade.

You cannot hold it firm enough in my opinion that way.

 

[Charlie Woody]

Hi Guys

Thanks for your help. It was the fence that was not flat enough which was causing the problems. Now fixed and whilst it is not 100% it is in what William Ng calls the twilight zone, so acceptable.

I still have to come up with a solution for clamping the workpiece, so will give that some thought in the coming days.

Anyway thanks again.

 

[MickCheese]

How about something like this for work holding?

Axminster sell them.

Mick

 

[devonwoody]

I myself prefer to have a stop on the end attached to the fence with a temporary clamp and hand held then seems sufficient.

 

[mickthetree]

Personally I always make a cut, then nudge the work piece over a fraction of a hairs bredth and take another pass. The initial cut is never normally bang on, but the second cut cleans it up perfectly.

If cutting a piece to length against a stop block I just take it slower, but not too slow to cause burning and the cut goes through plently straight enough. I guess being held against the stop block and with a hand is enough to keep it firm.

Also when squaring a cross cut fench I cant see where all the complication is. I take a piece of ply 6" wide with a very good straight edge which is held against the fence. The fence screwed on at one end only and held the other end with a clamp. I set the fence to pretty much 90 degrees then make tiny adjustments to the fence with a hammer. Keep taking more cuts and make adjustments till the ply registers as perfect 90 degrees with a very good, well trusted square (I have a moore and wright 6").

Once set perfectly I tighten the clamp even tighter, turn over and screw along the lenght.

You can then add another sacrificial face of ply to the fence if you like, but I dont bother.

HTH