|Home » Forums » Cabinetmaking » Message||Login|
You are not logged in. Consider these WOODWEB Member advantages:
scaling up production of non-rectangular parts5/18
Good morning all,
I was hoping someone could give me some advice on the advantages of going from flat-table CNC to beam saw/point-to-point. I know there is lots of data on this for conventional cabinet parts, but what about more complex parts with more tool changes?
We currently have a conventional 5x10 flat table 3-axis router. We are making an unusual product that is comprised of about 19 sheets of laminate clad 3/4" and 1/2" plywood, about 110 cut plywood parts in all. The product is unusual in that maybe 80% of the parts are non-rectangular, and all of them require tool changes: About 20% of the parts only require one tool change (perimeter cut followed by a drilling operation, for example), but the remaining parts need as many as five different tools.
How does the relatively high complexity of this sort of task affect the choice of machinery and process for large scale production? Let's say 1,000 sheets per month scale. How about 5,000?
Right now, before refining the toolpathing to squeeze every bit of efficiency out of it, we are averaging 95 seconds per part to machine start to finish, including the time to load/offload the router. (based on 19 sheets and 110 finished parts) So this is from the moment the first sheet is loaded to the moment the 110th machined part comes off the cnc bed and is placed in a pile.
I can imagine the benefit of stack cutting on the beam saw all the rectangular parts that only have a single tool operation after it's perimeter cut. We would make a jig for the router table so you could load as many as 8 identical parts at a time and pendulum machine those parts. The router would work continuously until the pile is gone with the operator pulling drilled parts off and loading blanks at the same time. But if there are tool changes, then I bet the pendulum efficiency probably vanishes pretty quickly. For multiple changes, unless that tool change happens in the matter of a few seconds (which it doesn't), I'm guessing I might as well just do it from start to finish on the CNC.
I can simulate different scenarios and put hard numbers on it, but am wondering what process a very large scale manufacturer might apply to this task? Is there no better approach to scale up than adding more CNC routers and run them in parallel? Maybe tweak design to get as many parts as possible to simple rectangles and reduce tool changes to a bare minimum?
I sounds like you are saying each part has five tool changes? Each SHEET should have 5 tool changes. If you can't nest it with one tool change per tool you should look at your software.
If I misunderstood please ignore my reply.
Many of the parts have between 3 and 5 tools required to machine them completely. So doing it as we are now with a flat table router start to finish, it is per sheet as you suggest. But if it's beam saw to point-to-point, I am assuming the change happens with every part. Am I missing something?
another way to state my question and a lot more briefly is:
Is it true that machining parts nested on full sheets on a flat table router is sometimes faster, everything considered, than incorporating a beam saw, regardless of scale? It seems to me this is obviously true but maybe I am missing something since I don't have experience with mass production.
I think it depends on the parts. Our CNC unloads itself. That helps a lot. A beam saw operation means you are going to handle the parts a lot.
The questions I would ask.
Does the router have the capacity to do what I need without affecting my other business?
Does the Router operator have other tasks be can be performing while the nest runs (like edge boring and banding)?
If it has the capacity and you are not inefficient with labor by having idle time then it is the way to go. If you can't meet the capacity or it would disrupt other business I would consider other options.
The router operator's time is already allocated fully to other tasks while the router is cutting. The bottleneck is likely going to be cnc time. Currently actual machining time, including perimeter cutting on the router (since I don't have a beam saw) is 120 minutes for 19 sheets. load/unload should add about another 90 seconds between run files after we setup for sorting off the CNC bed, which we haven't done yet.
From your comments Bill it sounds like it really is the case that for some projects (heavy on complex non-rectangular shapes with multiple different tool operations per part), a nested flat table router cannot be improved upon. No matter what the scale. The only way to scale is to have multiple production cells, not some different process.
This has been beaten to death here. What machine are you running? Software doesn't matter. If you don't have a enough machine you cant run fast enough. Tooling is a major player also of course. I have 4 very large CNC machines and 2 beam saws. I machine 3-5000 sheets per week on the routers. Complex parts not simple cabinet stuff. Good tooling and a heavy machine= speed. Ive worked on ptp machines before so i know and understand them well. It would be impossible for me to reach my production level i have now without nesting. I dont care that the saw can cut 8 sheets of 3/4 at a time. The parts not done then it has to get moved to another machine. We do have 2nd OPS on parts. I can put pods on my tables. You cant put a table where there's only pods. The major players are good at making deals on ptp machines. Look at the used and auction markets. They're flooded with these machines 8:1 ptp vs heavy iron. A good system quality tooling heavy machine and big balls.....success.
Do you have an on-board tool changer and boring bars?
Yes, 10 position tool changer.
A boring block would not improve times, nothing is standard repetitive spacing. A hole here, a hole there, different every part and a variety of different diameters. To machine the whole assembly I use nine out of my ten tool positions, though it's probably an average of 3 or 4 tool changes per sheet if I had to guess.
My machine has 2 spindles and can load the next tool while one is cutting effectively cutting the tool change time from 20 seconds to less than 5. Makes a big difference in cases like this.
If you use the spindles independently of each other in the boring block and assign each with a different diameter boring bit you could eliminate a few tools and tool changes on the main spindle.
I'm with the other Gary on this one...If you have more than one diameter of drill on an sheet, the boring block will save tool changes, which are time.
Can't imagine life without a drill block, if for no other reason than setup time.
I keep about 10 different sizes in the block.
Are you saying you don't have a boring block? They change drill spindles almost instantly. How many different router bits does the pattern require, non- boring operations. I have a PTP used for one job. I would not get another! Next new machine will be a self load/unload router. PTP is for sale.