Perfect edge finger joint with one cut statement

[brucehiggins]

I need to edge join thin cedar boards where strength is more important than looks. I came up with this joint using a SINGLE CUT STATEMENT:

6990
Those are 2 boards fresh off the bot pressed together. My next post will have the details. I hope I am not re-discovering the wheel but it was fun figuring out.
Bruce
brusigns.com

[brucehiggins]

Details:
Here's the only cut statement:
CC,1,T,-1,225,45
That cuts an arch of an inch diameter starting at 225 degrees and going to 45 degrees (I have a 48x48 bot and need to join 5' boards so I am doing it at a 45 degree angle). Change that to CC,1,T,-1,270,90 to cut parallel to the x axis. Here's the whole cutting file (it could be set up with prompts for length, depth of cut, etc easily). Btw this was my 1st attempt at coding a cutting file so I know it could be improved. Oh, use a 1/2" straight bit and place the board so the edge facing the x axis gets cut. I am happy to elaborate if there are ?'s. I will try changing it to cut clockwise to see if it cuts down on the cedar splintering (start at 44,44 and use CC,1,T,1,45,225).

'SHOPBOT FILE IN INCHES
IF %(25)=1 THEN GOTO UNIT_ERROR

[beacon14]

Bruce,

That's a neat joint for sure, but frankly I'm not sure it's going to be any stronger than just glueing two straight edges together, assuming you have a jointer or other method to get a clean, straight edge. Since most woodworking glues produce a bond stronger than the wood itself, I'd almost be willing to bet that if you tested both your joint and a basic straight edge joint to the breaking point both would fail along the wood grain next to the joint, not at the joint, especially in a soft, fairly weak wood like cedar.

Now if you were joining the boards end-to-end that would be quite different. I don't mean to rain on your parade and I'm always one to encourage experimentation (and playing with programming statements) but that's basic woodworking theory.

By the way you can save a variable and a few lines of code by using

&passcount=&passcount+1

instead of using "&temp" to store the new value.

[brucehiggins]

David,
You are probably right. My concern with thin cedar (I will be using 5/16") was that I can't clamp it as tight as hardwood and therefore wasn't sure the glue joint would be as strong. I know with this joint it won't break on the glue line. I will try a test like you suggest.
ps. I first coded &passcount=&passcount+1 but then wasn't sure it was okay. I suppose someone will say you can do &passcount+=1 which many languages allow. The programmers guide doesn't have all the info you might need. For instance, can you say &z=&a*-&b or can you use parens to control precedance i.e. &z=&a*(-&b)? Btw, I am a computer programmer at my day job.

[larry_r]

Bruce

Check out this link. About half way down there is a picture of a puzzle joint that Jeff Hamilton discusses. I am certain this or some variation would help you.

http://www.talkshopbot.com/forum/messages/29/21849.html#POST54375

Larry

[beacon14]

Bruce,

Most people overestimate how much clamping pressure is required. With such thin pieces, if the surfaces are true and clean, you can literally tape the joint together with masking tape while the glue dries and it will still end up stronger than the wood itself.

Most common functions will work as expected, including those with variables. There is a file called calculated_functions-15272.pdf in my Program Files\ShopBot\Developer Tools\Docs folder- not sure if it came with the SB3 software or if I downloaded it from somewhere but it has a list of supported functions. If you don't have it I can e-mail it to you (tried to attach it but no joy).

And yes, you can use parentheses to direct the order of calculation of functions.

[bill.young]

Hey Bruce,

I use the CC command to do my wavy scarfs for end joining panels...it's pretty slick how simple it is.

http://seaside5592.blogspot.com/2005/01/new-variation-on-stepped-scarf.html

Bill