Q.
Could I make a beaded face frame on a CNC machine? I'd like to start with stock which has been jointed on one face and planed to final thickness, but left a bit wide so the machine can make the final trim cut to width.
Here's what I'd like the machine to do:
1. Cut the stock to length.
2. Trim the stock to finished width.
3. Cut the correct size haunch for the rails.
4. Tennon the rails.
5. Mortise the stiles.
6. Route the bead.
7. Mortise for butt hinges.
I want each component to be fixtured once and remain there for the entire operation. Typically, I make my stiles and top rail 1 3/4", my bottom rail 2", and any mid rails 1 1/4". I have various machines set up now to do this and don't know if it would be more work or less to do it by CNC.
Forum Repsonses
Yes, a CNC machine can do it, hands off, start to finish.
How much volume do you have? If you are spending 80 or 100 "man hours" a week performing those operations, you definitely should look into automation. However, a typical CNC P2P or router may not be the best CNC answer. If you are a custom, build-to-order shop, you have a great many sizes and configurations of pieces to machine.
If you go with a router, you will want some or all of the following options: Drill block, right angle aggregate, saw, C-axis, and a table as long as your longest part. You may need more than one spindle, or a constant horsepower spindle capable of low RPM operation to swing the shaper-like cutters. You will also want to set up ported vacuum stations to index the blanks. Your programming will be special, and could be efficiently served by a custom macro, or parametric program that either takes data as you go, or from a database, which means you would queue the parts in a specified order.
I have not met anyone doing face-frame on their router yet, but that doesn't mean it can't be done. You can beat CNC with dedicated machines and manual labor (or limited automation), but the CNC can do a host of other profitable tasks, such as cutting out the panels for the rest of the job. The pricing for CNC machines ranges from less than a good sliding table saw to a half million dollars, plus. Perhaps a review of how much volume, and how much time you are currently investing would provide a base line for comparison.
I also do a lot of curved raised panel doors, which I feel a CNC machine could really help with, although I don't know how much of a pain it would be to program.
1) I know Auto Cad 2000 2D. Is it necessary to learn 3D?
2) What is a right angle aggregate?
3) I know what the X,Y and Z axis are, but what is a C axis?
4) My current budget is around 100K. Is it feasible to purchase a "stripped down" high quality machine now and add the various bells and whistles later?
5) What brands should I look into?
2) A right angle aggregate is a special tool group, which combines a gearbox to change the spindle (vertical) rotation, into horizontal rotation, as for use with a saw blade. There are literally hundreds of different kinds of aggregates. Some allow the user to adjust the angle anywhere in between 0 and 90 degrees from the spindle. Some that have a built-in vertical cushion allow the tool to "float" across an uneven work surface, yet maintain a constant depth of cut.
3) A "C" axis is a term stolen from the metal turning industry, and as it applies to our case (at least as far as KOMO and Biesse list it), represents a special attachment to the spindle which is used in conjunction with an "aggregate" tool. The purpose of the "C" axis is to control the rotation of the aggregate. Think of a saw attachment. Sometimes you want it to go down the X axis, other times the Y axis. Most saw attachments are either fixed or 90 degree selections. If you have a C axis and saw aggregate, you can go at any angle. Some adjustable angle spindles, in which you put a router tool, are used in producing some very detailed square cornered MDF doors that cannot be made without the aggregate and "C" axis.
4) As far as stripped down and build later, it is cheaper to get it the way you want it from the factory.
5) I have limited knowledge of what kind of machines are capable of your products. The one I am buying is more than twice your stated budget. However, another model by Biesse called a Rover 23, has a tool changer, can handle aggregates, has a drill head (13 vertical and 6 horizontal), and has a saw. I think my quote for that machine was around $140,000. It was a cantilevered design, and had a 4' by 9' nominal universal table. Consequently, you can do nested sheet stock work, or use the neat little pods for stand off work, such as the raised panel. You would have to devise special pods for the face frame.
I have seen Biesse machines and Komo machines, and Autorout machines, I can say that either of the first two will get you there. I do not know if Autorout is an active entity anymore, but I do know the guy who built them aspired to compete with Komo's equipment, and if he finds a few million dollars, he just might.
There has been a lot of discussion about programming. The systems we offer do not require programming for these components. They use parametric programs that are already developed. You need to specify the sizes you want, but do not need to create programs yourself.
We have not done beaded rails, so we would need to develop this using your design. It should be possible.
To make this work, you will need an aggregate horizontal router head, but should not need a "C" axis. The aggregate can be picked up by the main head, but processing will be faster if you use a separate "piggy back" router equipped with the aggregate. This second approach will cost a bit more but should pay itself off quickly when you start running. We have also developed special extruding dies that make low cost holding fixtures for these components.
First, theirs will not give me an integral bead, so I'd be stuck applying the individual bead strips by hand.
Second, because I consider it a mortal sin to make two or more framed cabinets and join them in the field, my face frames can get pretty large and complicated.
We encountered one problem--the doors were not square or exact size. The customer did not want to machine the perimeter, and if the door was more than about ten-thousandths off, or if the operator loaded them slightly off, the bead looked pretty bad.
We ended up using a touch sensor on the machine head to automatically measure each door before cutting. We then "adjusted" the program automatically, so that it would cut an acceptable bead.
This approach did add about about 15-20 seconds to each cycle for the measuring, but allowed us to cut an acceptable bead on somewhat random sized doors. This system is in production today and seems to be working OK.
