Design consideration for shaper motion and speed

I am looking at a design for a specialized kind of shaper that essentially shaves a piece of material and I want to design the machine to make the cleanest cut possible (no shearing). My main concern is how to determine the optimal speed of the cutter to make the cleanest possible cut. This device is not a production machine, but would only be used in a laboratory setting where the cuts would be made periodically. The thickness of material removed will be very thin, maybe 0.001" or less at a time, so we are just shaving a thin skin off the top of the work piece. (Obviously, once the thickness being shaved goes below 0.001" the machine is turning into a high-precision device, so an ordinary shaper cannot be used.) The behavior is similar to that of a microtome, but microtomes are designed to cut biological tissue which is much softer than the materials I need to cut.

The most common type of material that would be cut would be a polyimide resin, which is basically a kind of hard plastic.

The simplest possible design would just have the material in a holder and the holder would be driven against a fixed blade using a screw. This would produce a slow steady force.

In a traditional shaper the blade is driven by a flywheel. The cutter motion in such case would have a quadratic variation in the speed of the cutter, but of course, the machine can be designed so that the work is only contacted by the blade over a short range of its motion, so that the speed of the cutter would be somewhat constant while it is cutting.

My problem is: how do I determine the optimal speed of the cutter? Do I use a screw and just cut through the material really slowly, or do I have the cutter slam into the work at high speed? What will produce the cleanest cut?

Also, if we consider the way that the cut occurs, the blade encounters the material and the material resists, so the force builds up and the material will tend to compress, but then the cutter breaks through and cuts easily after that. So, there is an initial shock when the cutter first hits the workpiece, but then much less resistance as the cutter continues through the material. How do I design the motion of the cutter in such a way that optimizes for this kind of action?

I am looking at the way microtomes are designed (which tend to be idiosyncratic proprietary designs), but since the material they are cutting is extremely soft, it is only partially relevant.

In danger or re-inventing the wheel here!

If a conventional ball screw drive can't run fast enough then a toothed belt system surely will. When going really fast controlled slow down and acceleration will be less than trivial too.

Dunno how big your workpieces are but if fairly small I'd seriously consider picking up one of those cheap desktop size alloy bed CNC bed router kits and re-building it into a planer style set up with moving bed and fixed uprights for the tool carrier. Pretty much all the gear you need to start playing will be there. Just add a solid baseplate, alloy tooling plate perhaps or one of Thor Labs bread boards, a couple linear rails for the bed to run on, fit a few brackets and its playtime. Maybe add some weight on the bed to help overcome the initial shock when cut starts.

From very, very limited encounter (coffee break chats level) with a similar thin slicing project I suspect cutter profile may be of great importance. As I recall matters there were issues with the thin slice crumpling and folding up as the cut began resuling in somewhat variable thickness of the slice. Cure may have been to start slow.

Clive

jscpm said:

I am looking at a design for a specialized kind of shaper that essentially shaves a piece of material and I want to design the machine to make the cleanest cut possible (no shearing).

Click to expand...

I'm not sure how to interpret your requirements, particularly the 'no shearing' part. It could be that you are insisting that the cutter blade must be precisely at right angles to the line of action of the cut. If that is not a strong requirement, you might be better off allowing the cutter blade to be at quite a pronounced angle to the direction of cut so that the cutting action has a more gradual shearing action. This will require a longer stroke of course, but would tend to reduce the shock as the cutter enters the work. It might produce a slightly different distortion of the slice, but you will get that in any case.

The way I would do this experiment is by making a blade holder I could mount to the head of my cnc milling machine or the carriage of my cnc lathe. First thing would be be to figure out what blade geometry works best. Motion control and rigidity already provided fro nothing.

Then I would make a something with a pair of THK LM guides, a fast lead ball screw and a servo motor. Design the thing so that it is very rigid and the servo is more than torquey enough. You can put this together with $1000 worth of parts on ebay. I am partial to AB DSD-2098-030 servo drives that can be dad for under $200 on ebay all the time. They run on single phase, you can put out 30A peak and configure them for pretty well any servo motor.

Yup, as noted above by several others.

Ballscrew, and a servo motor.

By using off the shelf linear ways, ballscrews and servos, you can pretty
much assemble this very much "cookbook" style, and very modular.

The servo/ballscrew will allow almost instant changes to velocity, anywhere
in the stroke, that you may need
as you delve into this project.

There were two methods used in the past to provide constant speed of the ram in shapers: rack and pinion drive and hydraulic drive.

What is the cross section of the workpiece? If it is rectangular, there should be no obligation to present the cutting edge broadside to one edge of the work, when it could instead be presented to a corner (cutting along a diagonal). This would permit a more gradual build up and release of cutting force, and reduce entry shock to the mechanism.

I would tend to think that very high speed will do a better job of cutting clean, something like a rail gun or a guillotine

Most of the answers seem to be how to drive the blade to allow experimentation, not addressing the question. I suspect that unless you have really detailed specifications on the material, an empirical approach is the only choice. Making a guess at the cutting speed, probably very slow will not work because the plastic will deform. It's tendency to stick to the blade is important and without knowing it, you can have no definite opinion. A horrible example is trying to cut a piece off a block of roofing tar. You cannot slice it with a knife because it sticks to the blade. It might be possible two pull a wire through it, the way we slice cheese. Roofers hit it with an ax. The ax separates the pieces so rapidly that the deformation force exceeds the tensile strength of the tar and it splits ahead of the blade edge. That produces a fractured surface like broken glass, not what you want. Presumably you want a surface with as little damage to the plastic's structure as possible.

Can you use a lubricant or Teflon coat the blade? Attaching the blade to an ultrasonic driver might be useful. So might be submersing it in a lubricant. I don't know, just random ideas.

I would polish the blade with strokes at right angles to the edge to put micro scratches in the cutting direction.

Bill

How big are your samples? If they are small enough, spend some google time with "microtome". Sample sizes can be up to 1"x2" on some models. Typical section thicknesses are .5 micrometers to 100 micrometers. if you do go this route, be very, very, very careful of the blades. They are so sharp that you don't even feel them cutting you if you slip onto them.

For a bigger slicing project, check out Visible Human Project - Wikipedia

Google on vibrating blade microtomes. This vibration is not ultrasonic, and the blade vibrates side to side. <https://en.wikipedia.org/wiki/Vibratome>

don't be too quick to discount an ultra microtome.
the plastic TEM samples are embedded in isn't that soft

this is probably not relavant but I will bring it up anyways. Leather skivers can slice 30 inch wide sheets of leather so thin you can see through them. they are a band saw, running a 2 inch wide band blade that is hollow ground continuously. the frame is very heavy and the blade is under tremendous tension.