Question
I currently use a cutlist program which nests all my parts together, and I cut and label according to the nested layouts. Parts then get stacked in piles of like parts as they come off the slider. I average about 3 sheets per hour using this method.
A second way of cutting the parts would be to rip all like widths, then crosscut all parts. For example, rip all parts for base cabinet sides and bottoms, then crosscut all to the correct width.
My thoughts:
1) Nesting will offer less scrap.
2) Cutting like parts will save trips around the slider to set the rip fence, saving time.
3) Cutting like parts all at the same time will reduce part to part variation.
Just curious on how you are doing it, if you have tried both ways and settled upon a method, if you have another method, or if you have any comments.
Forum Responses
(Cabinetmaking Forum)
From contributor L:
I don't have a slider, just a standard TS. I cut the strips out of all the sheets and label the ends 1, 2, 3... and same with the paperwork. Then I go back and crosscut the strips. I would rather do it with a CNC, if I could afford one.
If your carcass parts are fairly standardized in width and you are using melamine or even carcass grade plywood, then it would be faster to always rip all the sheet stock first.
Leftovers from your crosscutting operation could be recorded and added to your cutlisting program at the beginning of the next job that uses that same material and thickness.
On the other hand, if you are using a very expensive plywood, then it would make sense to allow your cutlisting program to lay it out for maximum yield. By maximum yield layout I mean that the cutting strategy would make it necessary to alternate between rip and crosscut operations to net the parts from each sheet (nest?).
This also would depend on the output of your particular shop. As a one man shop making a product that varies widely, I use cutlisting strategies that minimize waste and make it necessary to alternate between rip and crosscut (nest?) to net the parts from each sheet.
As contributor L pointed out, it helps to use some sort of numbering system on your layouts and ripping ends to avoid getting lost and confused between rip and crosscut.
Why bother with optimizing? Start with the largest and end with the smallest and don't waste time getting 2 more inches out of scrap. Any un-used larger scraps wait for the next job, and if still not used, they get tossed. Cheaper than moving the stuff around. Plus, it's been paid for by the customer.
I can cut, band and bore a large kitchen in 1 1/2 days using nothing more than my pea-brain.
Before the vertical saw was purchased, I used a standard table saw. It was a physical drain to lift and push each sheet through, plus I wasn't getting good square edges and ends.
Optimized Patterns (what you are referring to as nested) typically averaged about 2-4% more yield. My labor savings from natural optimization were far more valuable than the material savings from optimized patterns, not to mention better throughput overall (the parts reached the secondary machining processes sooner).
Material handling is definitely easier with natural optimization.
Optimizing can save way more than 2-3%. I have been able to beat my program by 2 sheets in a large kitchen. When you are paying $100+ a sheet it can be worth it. But for me to hand optimize a kitchen that size it can take upward to 2 hours, where the computer will do it in less than 10 minutes. So that $200 gets burned up pretty quick and I usually choose to let the machine do it.
Next kitchen I do, I think I will try ripping all, then cross cutting just to see the difference it actually makes. Waste from the cutlist program was 2 wheelbarrows full. We will see what happens when I rip then crosscut.
Contributor B, I agree, I'm just using $27 per sheet melamine, so if I have to buy an extra sheet to rip then crosscut, I could care less as long as it is faster. Now for the couple sheets of veneered that get used for gables and the like, I could really care less about efficiency, as I am looking for maximum yield here.
Although it may be a great investment, you need to have the collateral to invest if you don't have the cash in the bank. I would love to have a CNC, and I could probably buy a low end machine if I really wanted to. But I don't. If I jump in I want one that will work for my future, not the here and now.
| If you are seeing this message, your version of flash needs to be upgraded to view this video. Click here to get the latest version of the Flash player. |
Here is what she made to hang on her bedroom wall. Drew it completely by herself and toolpathed with my pre-saved strategies in Enroute 4.0. She nested it herself in the material that she measured. Of course I resawed the material and planed it for her. It took me months to get to this point. Today's kids that grew up with computers have no trouble with this technology. The same kids would prefer not to work at McDonald's and would probably enjoy working for you.
In these times a CNC is a survival tool. You can't realize the though-put they are capable of until you get one. Even if you are afraid of the technology you can easily find some to run it or learn it in a few months yourself. Old school tools will only increase your costs of production and make it harder for you to survive in the modern market...
Please don't take this the wrong way. I'm saying this as a father. I have a 5 year old daughter that I let start jobs on my CNC, but I make her dress just like any other person that works for me. Work boots, eye protection, and hearing protection is a must, not a choice. Sears sells head phones for kids. They only cost $20.00. I wish you and your daughter all the best.
I have extensive experience running and programming CNC routers as well as CAD drafting experience and although they are awesome machines, I don't currently have the power to run them in my shop or the cash to buy them.
I still think the best way to make repetitious parts out of plywood is with a CNC, and definitely the safest. Absolutely safer than standing in front of a slider all day. If you are only cutting 20-30 sheets a year then maybe not a good investment, but if you are buying whole units of plywood routinely it is the only way to go. I think many old school guys are afraid of the technology and learning curve. The point with the video was that anyone can learn it. I know several shops with sliders sitting in the corner collecting dust since they bought a CNC. They are relics of a bygone day.
Beam saw versus router: if the saw is single sheet cutting the router still wins (Schelling versus Komo), even just cutting rectangles! The saw wins if there are repeat patterns that can be stack cut. Material optimization, the router generally wins since cut lines don't have to run "through." There are lots of variables in what we do, so judgment must be used to optimize the entire system. Keeping TOC in mind improves throughput.
To answer the original question... First of all, three sheets an hour sounds really bad to me. I think you are pushing the yield too far. Most optimizers have a "yield vs. labor" parameter. I have found that going for the easier to cut patterns (labor saving) makes no difference on large optimizations (10+ sheets) and gives like a 10% to 20% worse yield on smaller runs (under 5 sheets), which is an extra half or whole sheet. I am not guessing on these numbers; the percentage yield is displayed in the optimization report.
I am one of those annoying lean manufacturing zealots, so take this for what it's worth. Unless you are just really slow on the saw, I am guessing the reason you are cutting three sheets per hour is not that it takes that long to cut the parts out. I suspect you are spending a lot of time stacking and sorting the parts while you are cutting. That could easily reduce your productivity by 40% or 50%, especially on a slider where walking around the saw is a big deal.
In my shop we set the optimizer to keep the cabinets nested in order with a maximum of 3 cabinets overlapping a sheet. That means that basically cabinets are coming off the saw together in complete sets with a little mixing of the next two cabinets. This allows us to immediately start assembly because the parts for a given cabinet are all there. So there is never a stack or pile. The parts go straight to the edgebander, then the boring machine, then to final assembly without ever stopping.
In my shop this system requires at least 4 people to run continuously, and with 8 people we can produce 32 cabinets in an 8 hour shift. With 4 people we can finish the average sized kitchen for a condo or smaller home in one shift easily.
I do not know if my system would work for you, but I bet you can improve on how you are handling your parts coming off the saw so as to speed up the cutting.
As for the rip vs. cross cut first vs. full nest? I say eliminate or minimize third phase cuts and rip first. If you rip first you should not have to use the rip fence except once. European shops only use the rip fence for lumber, never for sheet goods. It saves a lot of time if you stay on the slider side of the saw. Also get into the habit of moving the rip fence while the table is in the full out feed position, then pull it back into the locked-for-loading position. Move the flip stops in the same fashion, while the helper is pulling the cut parts. Another tip is to pull the offcut of the first rip (the part on the left of the blade) on to the material stack (it should be just to the left of the saw). Then rotate the "right of the blade" rip so that the good edge is on the cross cut fence.
If you only use the slider (not the rip fence) the first four or five cuts are a lot faster. It goes like this... Head cut (rotate 90), dust cut rip (move the sheet for the first rip using the flip stop that was already set in place before you moved the sheet), rip cut (if there are more rips for 12" upper parts then keep ripping and dropping the rips back onto the sheet stack), crosscut for part lengths, then any third phase cuts for parts as needed.
The rip fence was never used and no piece was ever rotated more than 90 degrees for the next cut. Also all the part rotating can be done after the slider is returned to the starting point where the blade is never in the way! Except for the first rips, all the parts were done on the left side of the blade and everything on the right side was waste. This means you never had to reach across the table except for the first rip. You always pull the offcut onto the slider after the cut is complete. Complete cuts are handed off to the left, not the back of the table. When finishing a part (third phase cuts or end trimming) the rest of the offcut stays on the outfeed table so you can pull it onto the slider when ready. If you are passing finished parts off the outfeed table it interrupts the flow of the cutting pattern. Learning on an American saw where finished parts always go off the back/outfeed table it took me a while to realize that on sliders finished parts come off the slider and go to the left, or over the crosscut fence and onto a holding table where the parts are held for edgebanding.
I could go on but I am not sure if this makes any sense to anyone except me. I will say that there is some debate as to whether to dust cut or head cut first and about some other details. But doing it the way I described one man can cut a sheet faster than a CNC machine and get a better edge for the edgebander. I know that CNC routers can leave a good edge if the right tooling and maintenance is used. I also know that in practice CNC machines leave a less than ideal edge (compression bit lines and fuzzy edges are common issues).
CNC is the perfect solution for a one or two man shop that does custom work. But it does take a certain amount of gross sales to make it pay. A lot of small shops have a hard time getting enough work to make it pay. Employees are expensive, that is a fact, but CNC machines do not run for free. Tooling, maintenance, repairs, spoil boards, electricity and cleaning up the machine all have a much higher cost than a typical Euro shop using dowels.
A head cut is a dust cut of the 4 ft. side. Third phase cuts are... when parts require an extra cut to get the width correct from a rip. Something like that!
I am a very tech savvy guy and CNC machines test the limit of my knowledge. What would the little girl do if there was an error in the G-code, a bit breaks (or any of a dozen other things go wrong like walking parts) and the remaining parts have to be re-optimized on the machine controller? How about compensating for material thickness variation across multiple sheets (the CNC's Achilles heel)? CNC machines and the software to run them are very sophisticated and require more skilled rather than less skilled employees.
I probably have the most amazing story here about how CNC transformed my business. I made a ton of money and never even bought one! I outsourced the machining. I ended up having a faster turnaround than the shops with the CNC machine! Today I use European machines and dowel construction because it is faster and more profitable (for me, not for everybody). I will probably buy a CNC feed-through boring machine like the one Gannomat makes to speed up the boring procedure in a year or so.
Anyway, there are a lot of ways to make your business more profitable. Focusing on machines and new technology is one of the least effective ways to do it. Instead look at the processes between the machines. Once the material is delivered a given cabinet spends 90% of its time sitting around and 10% of its time getting worked. Buying expensive machines to reduce the 10% makes no sense when it is usually a lot easier and less expensive to reduce the 90% part. Look at what happens before and after parts are machined; that is where profits are lost.
I have been saving up for a slider saw, line bore, edgebander, etc after trying to build kitchens with a Powermatic and a Festool. Biggest problem is squaring the panels with typical cabinet saw. Realized that the inefficiencies would not let me make a living with my current setup and went back to installing.
Analyzed my installation technique and figured out how to streamline all functions that took too much time. Really tried to have ongoing improvement on technique and behaviors/speed. Sometimes it meant new tools, sometimes a new methodology. Looking internally at all the things that you can control and acting on them/trying them out can yield amazing results.
I have doubled and sometimes tripled what I make in a day by doing very simple things and always looking for a better way. When I get to put my shop together I would like to come buy you lunch and see how I should be using my slider.
Well, actually I am visiting here in Texas (came to go to the AWFS) for another month.
I would love to hear your ideas and experiences on installing. I had just about given up on installing before I sold my shop. I was still somewhat convinced that "No one could install my cabinets as well as me." But the truth is I never made money installing, and there are very good install crews out there.
As for the cost for my company's (New CNC) machine, let's just say less than a new pickup. The machine runs under $5.50 per hour on a 5 year lease. By the way, after 5 years, the per hour goes to $0. (Try to hire a guy for that! $5.50 for the first 5 years, then free for life.)
Now let's assume your fancy CNC will cut 1 sheet every 20 minutes (I know they can do it faster, but let's use 20). That's a total of 134 hours per year your machine will run, or ~11.5 hours per month.
Assume a monthly payment of $500, which is probably low, and you have $43.50 per hour.
The CNC needs to be running all the time to get it down to $5.50 per hour. I know you can increase capacity because now you have the extra speed, but you have to have the sales to increase the capacity, and with the current economy most of us don't have the sales to justify the cost. Maybe one day, but not right now.
150+ thousand dollar CNC machines are a different story. They can cut complicated patterns and do all the boring (with dedicated x/y lines boring heads) in 10 to 15 minutes. That is impressive.
We tried outsourcing parts but due to the custom nature of our shop, ever changing materials made this difficult. So we are buying one of those 150k routers. We have many reasons; most have been noted in the previous posts, but for us there is one more. It is becoming increasingly harder to find skilled men who can produce at this rate and quality. We need to free up our skilled men for more complicated higher value work than basic cabinets.
Second reason is I already draw all our job in CV and it is machine ready. We feel we can cut out the time from when I give a guy a set of drawings and cut list and he gets it to the saw with a full understanding of the job. If it is drawn right it will be cut and labeled right... no confusion or miscommunication.
I have a slider, 5 station edgebander, three boring machines (one three head , and two 2 head vertical boring machines). Also a hinge machine and dowel gun.
We don't have a case clamp, so we use confirmats to close up the dowelled joints. We also use a hot glue gun to secure the backs in the dado and hold the case square while the dowels dry.
It takes us 30 to 45 minutes to make one cabinet and we can make a cabinet every 10 to 15 minutes. We make our doors on a different line and they are started around 18 hours before so they will meet the cabinets for final assembly. There are usually 6 to 8 guys on the case work line, which include final assembly and stretch wrapping.
CNC is quite possibly a great option for you, but I suspect you can speed up the operation without adding much machinery. Have you ever taken any courses on Lean manufacturing?
