Edge Quality Issues with CNC Nested-Based Router Cuts

Here's a look at the quality-control problem of fuzzy cut edges in panels cut by a CNC router. December 3, 2010

I've started using E-cabs and outsourcing my CNC cutting work to a local shop with a Thermwood router. They build mostly frameless plywood kitchen cabinets. I do mostly commercial melamine/laminate jobs and wood furniture. I think the quality they get off the router and their edgebander is acceptable for kitchens (prefinished plywood with wood edgebanding tape), but the flatness of the router cut edges I get need to be sanded flat or jointed to get the quality edge I get strait off my sliding table before I apply PVC, laminate, or thick solid wood edging or trim. I think they use the same set of bits to cut just about anything, should I suggest different tooling or is the quality I'm getting them typical and should stick to prepping the edges as I described?

Forum Responses
(CNC Forum)
From contributor I:
I would ask them to use a new tool on your next job and see how the cuts compare. I use one set of tools for melamine and another set for plywood. This seems to help quite a bit with edge quality.

From contributor G:
A saw will always give a smoother edge than a router. When using a router, the core of the material plays an important role in the edge quality. The coarser the core, the rougher the edge will be. Even with a new bit, the edge quality will deteriorate well before the surface edge requires a tool change. We have a very similar problem when using Panolam, as it has a very coarse core. If we're using PVC edgebanding, then we'll frequently need to use an edge sander prior to edgebanding.

From contributor K:
"A saw will always give a smoother edge than a router." I don't agree at all with this. I believe the opposite is true. How do you think a jointer works? It cuts with a knife sharp edge just like a router bit just a different diameter. I sharp bit will make a big difference it the edge quality. Also if the machine is old and has worn ways the cut quality will be poor. The feed speed can be adjusted to get a better cut also.

From contributor Y:
You are outsourcing so time is money to your supplier. He will be running as fast as he can to make money. We can get an excellent finish on plywood (UV prefinished) we just watch our feed speed and use high quality compression cutters.

From contributor S:
"A saw will always give a smoother edge than a router." Then there's something wrong with your bits. We have a large beam saw and a router. The router cuts are superior to a saw blade cut.

From contributor O:
A CNC router should cut at least as good if not better then a saw with proper tooling, and the correct spindle speed and feed rate. The condition of the machine can definitely have an impact. If it is older and not well maintained, cut quality will suffer. If the spindle is in bad shape and the bearings are worn, this could also contribute. It might make sense to visit the shop and discuss your concerns. Unless the machine is really worn out, the remedy should be fairly simple.

From contributor G:
My interpretation of the original post was that he was referring to the smoothness of the particle board core, not the actual edge of the melamine. Are you all saying you can get a smoother cut on the core of course particle board with a router than with a beam saw? I've worked in a shop with a Schelling beam saw for 15 years, and have also used a Masterwood point to point and more recently a Morbidelli doing nested manufacturing. In both cases I saw a smoother cut on the core using the beam saw. The router will definitely give a nicer edge on melamine or laminate edges of the face.

I'm interested in what tooling you’re using to achieve this. At my current employer, when edgebanding doors with PVC, the owner prefers to manually cut them on a vertical panel saw, rather than nesting them on the router, because the rougher core from routing telegraphs through the PVC. We've been using Vortex 3189 compression spirals, and am about to try a coated compression bit from Active Machine and Tool.

I'm very interested to hear how you're getting smoother edges with a router than a saw, as I just haven't seen it in my 15 years. I know that a finer core will give better results, but that's not always an option. As I mentioned, Panolam is probably the worst offender, due to its very coarse core.

From contributor G:
Maybe a related question, but are people that are doing nested routing cutting conventional, or climb cutting?

From contributor U:
Interesting. I am also outsourcing cabinet parts to a local Thermwood shop and they do a great job with melamine, but all of my plywood ends up with raggedy edges that need to be sanded. And the cuts I get from my panel saw are also much better than the CNC shop does for me. I do believe that in my case it's an operator inexperience problem.

From contributor S:
"He was referring to the smoothness of the particle board core." Yes, but a router will cut it smoother if it's a climb cut, not conventional because conventional pulls the core.

From contributor O:
From the original posting it sounds to me like the whole edge quality is suffering, which leads me to believe that the machine condition, tooling, and cutting strategy could all be contributors. I am confident it probably won't take much to improve the quality. Will it be enough to satisfy is another question. He will have to work with his supplier to make that determination.

Acceptable edge quality depends on who you're asking. For me it is when the prospective buyer says it's good enough, and it's cost effective. Cost effective is the most important consideration. CNC routing is one way to get there, but not always the best cost solution. Nested base manufacturing is a great strategy but it can also be the bottle neck in many shops. In many companies, using nested base and traditional panel processing techniques are required to get the most cost effective results. The choice depends on the project.
Contributor S is getting great results from CNC routing and it does not surprise me. Contributor G appears to get better quality on some materials with a saw, but suggests that nested base would be the preferred strategy.

Here are some things you can try: I am not sure what software is being used but most of them will allow you to leave a cleanup pass. Rough cut the perimeters leaving .02" - .05" and use a finish pass. This should improve the edge quality with any cutter. Try a strait flute cutter instead of a compression spiral. Onsrud makes a solid carbide, three flute strait cutter that I have used in various types of wood and plastics with great success. A strait two flute is also worth trying. I would climb and conventional cut in any tests. You never know. Investigate other cutter geometry. One time I was cutting soft aluminum. I tried all the recommended cutters with poor results. I added misting without any improvement. I ended up getting superb results from an O-flute cutter designed for plastics without misting. Go figure. Feed rate and spindle speed have an impact. You will just have to run tests. Material hold down and machine condition are also important. I still contend that a CNC router should achieve equal or better results than a saw. Will it be more cost effective? For me that is the most important question.

From the original questioner:
I'm using E-cabs, so besides from knowing how to design what I want to build no other decisions are made on my end besides what material and then it get's cut. I don't know much about CNC programming or the controllers so I don't know of any advantages or limitations of the Thermwood/eCabinets combo. Most of my jobs are small, under 20 sheets, and since the shops system of just loading the file and material and hitting start works for them, I don't think it would be cost effective for me to pay for all that time and money testing cutters and programming or in the shops interest since what is does currently works for them. The MDF core cuts better than the particle board core, but the plywood is the worst, and the quality control is the most demanding for it (almost always high grade veneer core, not prefinished). I think a 3/8 spiral cutter is what's being used for most the cutting, and a fuzz or fibers is typical on most cross grain cuts.

From contributor L:
This sounds like the issue has to do with tooling and cut speed/direction more than anything. I rough size my panels on a sliding table saw, and then trim all to size on a point to point machine. When the tooling is sharp my panels come out great, edgebanding goes on much better than a panel cut with the saw. One main factor is that the compression bit cuts the edge of the panel perfectly at 90 degrees and there is no scoring blade (which is ever so slightly wider than the main blade).

I get such great cuts off the CNC machine that a salesman selling an edgebander recommend not getting premill on the bander after seeing my panels. With premill on an edgebander the quality of banding is perfect, as is uses dual diamond tools to trim the edge of the panel before applying the banding. I use a conventional cut on panel products. I am only taking of 2 mm per edge, which also squares the panels perfectly.

From contributor G:
From my experience, routing 2mm off a precut panel gives much different results than cutting nested parts.

From contributor E:
We used to have issues with plywood fuzzing very badly. We fixed the issue by changing to ccw tooling to cut the plywood. Our nesting solution does not allow changing direction of cut on the fly, so we opted to change tooling on the fly.

From contributor Y:
We can get smooth cuts on our CNC (nested) and no scoring saw line so the banding is always tight on both sides. Another thing you could try is thicker edge banding. PVC tape comes in two or three thicknesses or try 1mm tape. 1mm is more durable and will not let the core telegraph through. Slow the router down, use 1/2" compression cutter, climb feed and make sure they have enough vacuum on the parts.