Thread: How do you cut an exact 45deg mitre?

Don't you fine tune the angle until you get an acceptable cut? It might take a few attempts but you will get there.

Hi Jill,

I'm with Wongo. Apart from using a shooting board to fine tune your angles (the best way I think) I manage to get excellent results with the Kapex, by putting the same length rails one on top of the other on the saw table, and making sure that the ends away from the blade are completely flush ( this is essential) then checking the ends at the saw for small burrs or "bends" in the centre of the cuts. Depending on the timber and how hard and fast you've made the cuts, you can occasionally find a small difference (probably about 0.16 degrees or less). As Wongo says, fine tuning like this can take a couple of tries, but you'll get the result you want and shouldn't lose significant stock. If you ckeck all your end cuts this way, you can hone the edges quite nicely. But I'd probably still recommend a shooting board, even though I don't have one m'self!
BTW, do you do a dry clamp up before trying to fix anything. A loose knock up won't necessarily tell you if there is a real issue.
Also, make sure there are no "dags" hanging off the back of the saw cuts. I usually shear these away with a little Festool DTS400 delta sander.
One more thing, I don't swivel the saw from one side to the other. I use the same 45 degree mark for all cuts. I just turn the timber round.
And lastly, make sure that your stock is really square. If the width varies by any amount at all, your corners will suffer and look like they do not fit.


Regards,

Rob

Hi, using a sub fence that is adjustable may be the go. I have the Bosch SCMS and would not dream of getting such accuracy. Even the Kapex would be rather difficult. Remember these machines are not as accurate as a good tablesaw (my opinion anyway).

Also Jill,
How sharp's your blade and which blade are you using?

What are you cutting Jill?
Is it possible to cut both mating pieces one on top of the other, with the top piece flipped upside down? Bit hard to see your pencil marks but as long as both pieces are in line (i.e hard against a straight fence) when flipped they should make a near perfect 90.
There will be a slight difference in the length of each cut face, but that is a separate issue

Good luck.

If the top and bottom are not the same length Or the 2 sides not the same length then when you assemble it square the corners will look off and not meet.

If your not getting 90 degrees with just 2 sides only meeting then its the cut.

Its not beveled on one side is it and your putting that frontside flat for the cut (say side piece) and on its back for the top piece are you? Does this make sense?

I've seen people do this because they think moving the blade from one side to the other is worse.

For large frames, the joints don’t have to be spot on. The wood will be flexible enough, so you can close any "gaps" by using clamps during glue up.

How are you going to join them together, biscuit, loose tenons...?

There's great satisfaction in getting the miters exactly right, with a lot of fiddling. Especially important in cutting segments for segmented turning, with many segments in each circle. Though I haven't tried it yet, an adjustable sub-fence seems promising in fine tuning the angles.

In the meantime, there's a way to cheat and get superlative results. For 4-sided frames, use corner clamps with full-depth gaps along the miter line. Clamp the pieces as best you can. Then run a back saw, or other reasonably rigid tool, along the miter so that it engages both pieces. The kerf has constant width, and the pieces will then fit perfectly. It may take two or three cycles to assure complete engagement, and the pieces should be slightly oversize to allow for reduction.

I've used this trick for the aforementioned turning segments with a band saw, using masking tape for "clamps." In that case, the cuts don't need to be perfectly straight; the pieces fit together perfectly anyway.

Cheers,
Joe

On my old Makita I have a little bit of adjustment in the fence (bolt hole clearance) this allows a tad of fine tuning to get it right, not sure on your machine tho

Another trick you can do to tune the cut is by placing a strip (or 2 or 3 or 4....) of paper at the far end of the fence or close to the blade end of the fence depending on whether the saw is cutting (+45 or -45), use some scrap (flat straight and square) and test by doing 2 cuts then place both pieces on a flat surface, bring both cuts together and check for gaps, any error will be twice that at the saw, and the wider the piece the more the gap shows up, alternatively you can use a square to check in the open 90° by turning one board over


Pete

miter gauge and sander for final adjustment

The cheapest way is as above use a disk sander and mitre gauge, the best and highest quality result used by Picture framers is by trimming your or fine tuning you mitre saw cuts with a guillotine, you can pick up hand operated chinese units quite reasonably priced http://www.macwood.com.au/index.php?...oducts_id=1249

Jill are you hold the wood on the saw table by hand or clamping it down.
Clamping rather than hand pressure may make just enough difference (assuming the opposite pieces are the same length) to get the mitres to close.

For small pieces - a hand saw and a mitre shooting board works just fine. On larger pieces my drop saw is accurate enough.

Many thanks for all your suggestions and ideas; there are many, many years of experience there that will benefit a lot of us.

I will try to answer the queries that have arisen:

1. I have a brand new Oshlun 260mm x 60T saw blade and have just made a zero clearance inset on the Festool Kapex 120 SCMS base, and screwed a 18mm sacrificial fence to the main fence to minimise the risk of a rough cut. The mitre cuts are all beautiful, but on the softwood pine I get furry burrs on the right top edge for some reason, but the sander quickly removes that. It is the right side that is held down tightly by a push stick when cutting. I only cut through with the saw, then switch it off before raising the blade.

2. I intend using biscuits in each joint

3. I liked the idea of cutting both sides together, that way the lengths are identical. I actually clamp them together at one end.

4. I am left handed, and that creates problems using machinery and tools. I have swung the sawblade to the left, and use the left hand to lower and push the blade. There is no room to clamp the piece there. With my right hand I use a 300mm long push stick to press the workpiece hard against the fence and to hold it down while cutting. Seems to work well. I find the Festool clamp is not very secure.

5. I tested the accuracy of the saw cut at 90 deg using the “5-cut Calibration Method” ( see attached pdf file below), and got a error of 0.07 degrees. Can't complain there.

6. Thanks to Wongo for suggesting that I fine tune the position of the sawblade. I actually hadn't thought of this. My thinking was to adjust the positioning of the 45 degree scale pointer so that the preset position cut exactly 45 degrees every time.

7. This morning at the local timber yard I bought 5.4 m of 2 x 1 finger-jointed pine to play with ($9), and got them to run two adjoining edges through the thicknesser so that I could be sure it was the same thickness and width along the whole length. I used up 1.5m on many cuts trying to fine tune the position of the saw blade to give an exact 45 deg cut, as suggested by Wongo. The amount of offset is shown in the attached photo, just a couple of smidgeon widths

8. I then used the remaining pine to make a test frame, getting the sides the exact same length as suggested by LGS above. The mitres joints were still not an exact 90degrees. Measuring the inside diagonals gave a 4mm difference over 550mm length. Using an old mitre corner clamp certainly seemed to enlarge the error. In the end I used a band clamp that seemed to pull the corners together, although not exactly a dead exact 90 degree corner. Will just have to be careful when measuring up the size of the glass to suit.

9. The photos below show the test frame, and the worst joint. I used a dado to cut the inside rabbet 10mm wide by 13mm deep, but one piece, as shown, fell off the end of the tablesaw and landed right on the pointy end of the mitre, as you can see in the photo

That's about it for now, has been a long day, but have certainly made considerable progress after yesterday's effort, thanks to all the suggestions you have provided