Please help with CNC design for tangential knife cutting.

Hi,

I am new to DIY CNC insofar as I haven't built one yet. I have been hanging around the edges of these forums getting an education and feel that it's time to get doing.
I have been basing my build ideas off the Solsylva belt drive table plans but also trawling through everything from Mechmate and Lumenlabs.

Everything I find assumes the use of rotary tools or laser/plasma. If blades are mentioned at all it's for vinyl with swivel mount blades. I would basically like to build a flatbed tangential knife cutter for use on cardboard and architectural chipboard. This is a cardboard material like gameboard backing which is used for modelling. I have seen it referred to as "beaver puke" if that helps :)
The main use of this machine would be for architectural modelling although I would also like the option of putting a router on it. If the knife carrier were a dual mount with pen-plotter that wouldn't hurt either.

So lets say a standard 3 axis router build with an option to mount a tangential cutter assembly. And let us assume about a 30" x 40" cutting surface.

At this point, I know nothing. That being said, let the laundry list of my questions begin.

Would the force involved in dragging a knife through 3 ply (~ 3/16" ) or thicker chipboard require a stronger machine than one would need for a router ? (I have no idea at all what the forces involved are. All I know is that my fingers hurt like hell after a day with the Exacto knife).
Would a belt drive do it or would there be excessive backlash ?
Ditto gas pipe rails. Do I need more rigidity than the Solsylva machine ?
Motors... 425 Oz / in ? I've seen anything between 250 to 600+. What would the benefit be ? Would progressive dulling of the blade cause the machine to chatter if the motors are underpowered ?

My previous architecture school had a Zund 1200 in the shop which was mainly used for stepped terrain models. This was in Switzerland, i.e lots and lots of curvy contour lines in cardboard.
Having seen that thing go, my expectations are now entirely unrealistic. I have to remind myself that gas pipe and skate bearings are not going to get me 600 ipm. But in the interest of argument I would be more than interested in what your impressions would be.

These are the kind of knives I'm talking about.

http://www.smp-sign-systems.com/prod..._Messer_e.html

The main reason why I would like to have a router on this thing as well is because, once it's built, it's first job is probably going to be making it's own vacuum clamp bed to hold the cardboard down.


Thank you all for any and all assistance with this project. I hope that this will evolve into a build thread before long.

jfmonod

Quote:

---

Originally Posted by jfmonod

Would the force involved in dragging a knife through 3 ply (~ 3/16" ) or thicker chipboard require a stronger machine than one would need for a router ? (I have no idea at all what the forces involved are. All I know is that my fingers hurt like hell after a day with the Exacto knife).

---

A quick way to find out how much force you need is to measure it with a fisherman's pocket scale. A MechMate is happy to push around more than 50 pounds [25kg] all day

I would consider a V-Bearing setup instead of the gas pipes in this case, Geralds Mechmate is a DIY'ers Dream machine it would work well for what you want, maybe even my 4x4 would work with mods for the vac table, it is open underneith and easy to impliment, A metal construction for the machine base and bed would be what I would shoot for. Prob. the 425oz or 570 oz would work for you, but on the mechmate you will need larger nema motors.

Joe

joe

I was looking at the Mechmate with envy actually. Truly a dream machine. Your 4 x 4 would be a good build too Joe. There are a number of build threads on the site that actually make it look easy. Of course there's the fact that I'm in architecture and have a limited metal working experience to contend with, but if I have to suck it up and learn how to weld then so be it : )

As I was initally looking at David Steele's belt drive table (see my above lack of metal shop experience), I got some good feedback from him on the advantages and drawbacks. It figures that that one's probably not going to work. It also figures that I already bought the plans and have the thing half built in Inventor : ) Well worth it though, if only for the Inventor tutorial and the fact that it makes it look doable. Of course I went back to the Solsylva site and just noticed that he reworked the design in a RP / leadscrew version with more heft.
I figure that about 10 lbs + of force in the z axis would be good for driving the knife into the material and around 8 lbs lateraly as long as it's sharp. I'm guessing that adding about 50% in the x - y plane to compensate for a dulling knife would be desireable so as not to lose steps. As this is about twice what you can expect from the belt table according to DS it looks like it's going to be RP.
Thanks for the fisherman's scale tip by the way Gerald_D. Never would have crossed my mind.

Now the hurdle is building the tangential cutter head. For all the stuff on this site, I haven't found anyone attempting that. I was figuring that this could be a 2.5D & 3D machine with a 6" z for routing. When in 2.5D the plotter cutter head could have dual piston or pneumatic actuator assemblies to toggle the knife and pen up and down independently of the z-axis screw. The knife would be on the main z-axis with a +Y offset for the pen.

This is going to be fun.

these type of machines are used in the corrugated sample making business....

we run a Data Tech. M3 (DT was recently bought by Gerber and now goes by Gerber Innovations.

we have a pneaumatic driven reciprocating knife tool. one with a .097" stroke for corrugated and another with a .25" stroke for cutting foam up to 3" thick.

Getting a COT reciprocating knife tool that can be adapted to your machine is going to be the key.

Most of what I've seen out there is proprietary built and would not be available or adaptable.

Hi Matt,
Thanks for the reply. Glad to see you know what I'm driving at. I had access to a Zund 1200 machine way back when. We used it for terrain models for architecture as well as for the more whacked out designs that were a pain to cut by hand. I had an idea that some reciprocating action would be involved for cutting chipboard, homeosote and other like materials. I have been looking for these cutter heads but as you said they're proprietary. The knives themselves are readily available, but designing the head that goes around them may well be outside my league. I'm in architecture and not mechanical design. Following the Solsylva plans might be about my speed, but that's about as far as it goes :) Any ideas would be greatly appreciated.

I'm looking for a COT (commercial off the shelf) solution and suspect that one may be out there somewhere.

This is a great app. (www.laminadesign.com/) It creates automatic cutting paths for curved surfaces (and whatever whacked out designs you want), compensates for material thickness autogenerates egde tabs, eye holes ... etc and numbers and lays out parts.

Any luck finding COT cutter heads ? None on my end. I'm going to have to make one.

JF

My boss and I had MasterCAM demo their reciprocating knife system. It's impressive and much more robust than the Data Tech. foam tool.

We run the Data-Tech 16 hours a day, 5 days a week and 95% of our cutting is in foam from 1" to 3" thick for packaging applications, and are burning out the pneaumatic motors left and right. They have a new current monitoring, constand tourque electric motor tool (knocked off Kongsburg?) in developement, but it hasn't hit the street yet.

I'm hoping that we by the MasterCAM, as we can combine knife cutting and routing (with tool changes) in the same program. But business has recently follen off and I suspect it will get put on hold.

MatserCAM apprears to use a Zund reciprocating knife tool, that they have mounted in a tangential control cartridge system of their own design. This adaptor doesn't really look that complicated, and if you don't need to swap out cartridges for other types of cutters (drag knives, roller wheels, etc...)

I'm curious about your application. Can't picture in my mind how it could be cost affective. Have you considered subbing the work out to a compitent foam fabricator?

MasterCAM sounds like a sweet deal.

As for cost effectiveness, it's not. I'm not thinking about production at this point. I'm a Graduate student in architecture at the University of South Florida in Tampa. Students have a habit of pressing the local laser-cutting shop into service to create model parts for the more bizarre designs. (Usually out of chipboard or basswood). One guy who graduated last year build a 4 x 8 cnc out of scrap steel for about 400 $ total and ran a router and plasma torch on it for his thesis final (not precise in any way but it got the point across). The school has recently received a grant for an in-house laser cutter so we're going to be getting into design fabrication in a big way. I was looking at multiplying the tooling possibilities as a research project with the idea of building a cutter / plotter combo based on a Solsylva RP design. This is mainly to have the option of avoiding the laser burn on the workpieces.
I am not going for high volume production or anything like that. I am also not looking at spending more than 1500 $ (considering that the DIY Solsylva thing costs around 600 - 800$ as is.) I figured that if pushing a router around was feasible, then putting a motor on a knife would be worth looking into. What this is going to take is another question. I've looked into the Zund cutter heads and the cost of one is pretty much my whole budget. This is why I'm reinventing the wheel, starting with teaching myself Inventor.
Yesterday I couldn't spell "Mechanical Engineer" now I are one. : )

I are one too.... :^)

Good luck with the project....

All the Zund cutters I see on their web sight look like "girly girl" tools.

I need a "manly man" knife that can cut 3" foam all day.

Quote:

---

MasterCAM sounds like a sweet deal.

---

for $110K it darn well better be!