[Hack]

I have been trying to find a self centring fixture for my cnc router. Basically, I want all material to be clamped in the center of both the x and y axis simultaneously and it must be repeatable no matter the size of the piece. My table has a cut area of 24 x 36. My material is wooden cabinet doors and drawer fronts. This doesn't necessarily have to be a clamp or vice thing (though it would be nice), but I am more interrested in the repeatable positioning capabilities

Accuracy needs to be +/- .03 inches or better. Does anyone have any ideas on where to buy this or how to design this. Was thinking some type of rack and pinion but concerned about dust. Could it be made with some sort of stepper setup and adjusted / jogged via software (turbocnc) by using a threaded rod with right hand threads on one end and left hand threads on another? My doors are only 3/4" thick so low profile would be nice.

The idea is to home the router to center of the table, have the self centering table and engrave designs onto the cabinet doors, repeat.

Thank you

Hack

[miljnor]

Thats a tall order (when you say any size). what are your working dimensions?

if your talking about 24 x 36 down to say 1 x 1 your going to have to make something yourself (probably). There arn't any GOOD fixtures that can span that range but Hey I work with metals not wood so I could be wrong.

[CRFultz]

I was wondering why you don't use a alignment pin at the bottom left and then do a offset at half of the material. If the door is 12" on the x and 20" on the y, then center will be 6 and 10...I probably don't understand the question.
Chuck

[jderou]

Have you considered a gear rack centering fixture? I've got example drawings if you need them.

[ger21]

Why can't you reference from the corner, 0,0 for example. That's how most commercial routers in commercial shops work.

[turmite]

His problem is the many different sizes of both drawer fronts as well as door sizes within a set of kitchen cabinets. If he uses one corner he has to write an offset for each piece that is different in size. With a known center, he can do the work from that location without having to , measure, jog or do offsets.

The rack system doesn't sound too bad. Let us know how this goes.

Mike

[ger21]

Sorry, I missed the part about the engraving. Do you really need an accuracy of 1/32" for an engraving? Why not lay out a grid on the table to line up the doors, maybe get some stick on tape measures for the dimensions.

[Hack]

The problem is definitely the fact that a person would have several different sizes to contend with. Normally from 6" x 12" to a full 24" x 36". There are times when doors and drawer fronts are bigger and smaller, but that is rare. And in those instances, one could deal with them in other ways.

Alot of the European Style cabinets only use a 1/8" reveal between the doors. With that tight of reveal and doors so close to each other, I think a good eyes would notice much more than a 1/32" variance. Thus the accuracy.

jderou

I was origionally thinking a gear racked system, but have not found one to buy or build. Pics would be great - Either post them here or send to [email protected]. This is a highspeed connection, so don't worry to much about file size.

The grid on the table idea sounds ok, but it is up to the operator to assure that everything is square and perfect. Which is fine, but really isn't much faster than just relocating the machine zero based on part size.

Thanks all, please keep the ideas coming.

[JFettig]

I dont really get why you cant have zero zero at front left? just have a stop there so you push it all the way down and all the way left and you have a zero, program it accordingly. Thats exaclty how I have done most of mine, set screw in the middle of the vice jaw zero'd on both sides of it and the back of the jaw.

Jon

[mvaughn]

Not knowing anything about cabinetry I'll assume that most doors and drawer faces are made with standard sizes in mine.

Why not create a set of Dies that nest in each other. Have the first die/jig pocketed to fit your largest door in the center of it and clamp or use pins to hold it in the center of your table. Then you can just drop door panel into it and it's centered.

Next pocket the next door size down into another die/jig that is cut to the size and fits in your first die/jig.

Continue on down until you have a set of jigs that hold each of your doors and drawer faces.

Kinda like those Russian nesting dolls.

I can draw a picture if my explaination doesn't make any sense. Hopefully I didn't misunderstand the question.

[jderou]

You can also have a ballscrew/acme screw driven system. You would have to screws, one right hand thread, one left hand. Attach them in the center (maybe with a gearbox between them to attach crank/motor), mount some sort of linear guide for the nuts, attach stops to the nuts.

Here is a drawing of a simple gear rack system. The pinion would be stationary (pinned at center).

[ger21]

Not knowing anything about cabinetry I'll assume that most doors and drawer faces are made with standard sizes in mine.

In our shop, doors and drawers can be virtually any size.

[ger21]

I don't know how you're generating your g-code, but it shouldn't take more than a minute or 2 to do each different size. I'd use 0,0 and program for each size part.

[anoel]

I'm with the guys suggesting that you offset from one corner with a set of stops. A simple method would be to zero in on the corner of say a Y-12" x X-24" door. Then reset the coordinates of the Y and X to -6 and -12 respectively and that'll put you axis dead on the center of the piece. and you know that the corner that you offset from is -6 and -12 to get zero for your next part. I've done this for a couple of engravings that I've needed to repeat that required the engraving datum to be in the center rather than part-left+bottom.

[buscht]

The self-centering jig is a good idea, but I too would use the software version of this.

Let's say that you always want the engraving to be in the center of the part.

Write the engraving program as a subroutine or subprogram.

Write a parametric main program that asks for the X and Y dimension of the part.
Simply have the main program call up the subprogram at X/2 and Y/2.

Keep the part gaging fixed at 0,0.

Overall, this method is just as fast as the self-centering jig, and much more flexible.

The only downside would be its not 100% mistake proof, because the operator has 2 dimensions to enter.
Trent

[mvaughn]

Here is a drawing of a simple gear rack system. The pinion would be stationary (pinned at center).

That's a great idea... a common use for the dual rack and pinion are paper alignment trays in printers.

[NEATman]

Instead of rack and pinion, you could use a belt or cable. As long as the pulleys are on thier side (the axis of rotation is vertical) you could attach a stop block to each side of the belt, so that they move in opposite directions. You would need one of these set-ups in each direction.

If your parts are always rectangular, you could clamp across the corners. I realize that everything will not be the same aspect ratio, but to take care of that, make the self-centering jig pivot about it's center. For instance, in the rack and pinion arrangement, you could use the pinion shaft as a pivot for the entire clamp. Then all you would have to do is make the part paralell to the one axis of travel.

If you followed that one, I have another idea that would center it in both directions with one adjustment. Arrange the rack and pinion or belt system on an angle, preferably the average of the aspect ratio of the parts you are making. Each moving end of the clamp would have a pivot pin. Atop these pivot pins would be a bar, and at each end of the bar would be a pin that would contact the part to be engraved. When you tighten the single angled clamp, each of the pins at the ends of the equal length bars would touch a different side of the square or rectangular part. The angled clamp would center it along it's axis, and the equal length pivoting bars would center it perpindicular to the clamping direction.

Let me know if this is not described well enough, and I could probably model it.

Keith

[Hack]

The self-centering jig is a good idea, but I too would use the software version of this.

Let's say that you always want the engraving to be in the center of the part.

Write the engraving program as a subroutine or subprogram.

Write a parametric main program that asks for the X and Y dimension of the part.
Simply have the main program call up the subprogram at X/2 and Y/2.

Keep the part gaging fixed at 0,0.

Overall, this method is just as fast as the self-centering jig, and much more flexible.

The only downside would be its not 100% mistake proof, because the operator has 2 dimensions to enter.


Actually the software idea would work well. Maybe even better than the centering jig. I use my router for other things and with the jig, I am afraid of installing, uninstalling, installing...well you get the point. Also, it would allow unlimited part size as long as it fit on the table. I think I like it. If the operator can't enter 2 dimensions and be trusted......especially since that would be me. Yes mistakes will happen, but I got to beleive they would be rare.

I could easily set up the corner fo my table for 0,0 both for the router and for the part. But how do I do the code to ask for the inputs. I didn't know it was possible. Still really new to g-code.

[Kmed]

Well what cnc control are you using. That will determine what subroutine / programs and macros or variables you can run.