Workholding is sort of a pain for me. Upon getting my MPCNC running I wanted to make chips so I sunk a few threaded inserts into my spoilboard and figured I’d make something more robust later.
Well, that later is upon me and I am considering the various options, particularly trying to optimize for convenience and being able to swap work pieces out quickly. I’m trying to make ~20 of the same part for a project and I’m imagining that swapping out the stock each time will take nearly as long as milling it, so I’ve been avoiding moving forward.
I want to throw a wacky idea to the forum so someone smarter than myself could tell me why it wouldn’t work, or wouldn’t be practical.
What if I embedded an array of permanent magnets into my spoilboard, and milled up a few adjustable wooden step clamps with a recess for another magnet. If I were able to find magnets with the right balance of strength such that they’d hold the work firmly but I’d still be able to remove them myself without too much fuss, would this not be a convenient system? Apart from calibrating distances and magnet strength to ensure a good hold, what might I be missing?
Alternatively, pieces could have short screws sunk into the bottom to align with magnets directly so that clamps wouldn’t be necessary.
If price weren’t a consideration, perhaps electromagnets or electropermanent magnets could be considered - though from what I’ve seen they’d be rather large and unwieldy, and potentially dangerous if you crash into them.
Appreciate any and all responses, as well as any general advice people can give regarding workholding and which systems they’ve had success with. My preference is to clamp down against the spoilboard rather than from the edges towards the center of the work, because I’m often working with thin stock and cutting rather close to the edges to maximize yield.
Some of the mag switches are pretty dang strong. Maybe you could print a clamp to go around it, but then the spoilboard would need a significant amount of metal.
Not sure how well that would go when you eventually run into it.
I would worry that magnets would need to be too close to the surface on the spoilboard to be effective, and would be more of a collision risk for the bit.
Appreciate all the replies, absolutely some good points that I hadn’t considered, particularly regarding horizontal (shear) force vs vertical force. I’m thinking that perhaps using some type of anti-slip material (rubber, neoprene, sandpaper?) underneath the edges of the work or the clamps could help to increase friction between the surfaces reduce the risk that there is lateral movement during milling.
I am a little reticent of hitting permanent magnets embedded into the spoilboard, but my z position setting seems to be working pretty well with the tiny touchplate, so I rarely end up cutting more than a tenth or two of a mm into the spoilboard. I guess I need to figure out if I can embed them about a 1-1.5mm deep and still get a reasonable amount of pull without spending a crazy amount of money.
Alternatively, there is no reason why the magnets couldn’t be touching directly, so long as the diameter of the clamp magnet is less than or equal to that of the embedded one - consider the following diagrams:
There is probably some optimal point bridging magnet size and strength and the density of magnets in the array on the spoilboard; I’m operating under the assumption that given a decent array size, say at least 20-30 magnets, I could find a way to engage multiple at a time to increase pull force (many weak clamps = a few strong clamps). I have a dust shoe that uses tiny and cheap 8mm diameter x 3mm thickness magnets and with 6 of them engaged it’s a really strong hold.
The nice thing is that unlike t-tracks or a vacuum, the CNC itself can nearly do all of the work for you (cutting magnet holes and cutting clamp pieces). Replacing the spoil board would not be trivial but also not particularly hard if you screw magnets in rather than gluing.
PS. those letterpress quoins are pretty cool. I really love simple and ingenious mechanical designs, and would definitely consider something like a magnet-mounted quoin. Thanks for sharing
If you’ve got a lot of identical pieces to do, a little time invested in easy clamping will definitely pay off. What about a semi-permanent L- or C-shaped jig simply screwed to the spoilboard that wraps 2 or 3 sides of the blank and then a couple of threaded studs to hold a 4th edge with wingnuts? If the sides were undercut (like half a dovetail slot) that could manage the hold-down force.
You’re spot on here, this is exactly what I’m planning for my current project that requires a bunch of repeated runs, in fact, I’m just going to build a sled out of some cheap plywood that I bolt on to my spoilboard and allows me easy alignment and clamping of the blanks, and just recycle or scrap the sled once I’m done.
But as is always the case, any time I need to do something I can’t help but want to solve the problem in the abstract, in some novel way, and with a healthy dose of over-engineering Despite now having read plenty on the theory of why and how magnets work, I still can’t help but feel like they’re actually just magic, so I might actually give this a shot and update this thread as I prototype.
Table saw surfaces are usually cast iron, and very massive, so a heavy magnet can grip pretty well. I don’t know how that would compare to gripping against a smaller mass magnet (negative for less mass, positive for also being a magnet).
But if a feather board comes loose, there is a human right there, with their hands on the wood. So they don’t have to work 100% of the time. The human can also adjust how much force they push into the featherboard to ensure success. The CNC is dumber than that.
My gut says that if you are expecting to just pull them off with your hands, they won’t be strong enough to hold the work down. Not without significant mechanical advantage, anyway.
Many of the table saw magnets use a screw to pull the magnet away from the cast iron, which means they can design a clamping force stronger than a standard issue human can pull up on. That seems like a good design choice.
The ones I’ve owned are also significantly bigger than I would be able to use to cut 1/2" material.
At the end of the day, my guess is that you will have some fun trying, but find that a few dozen well places T nuts and some cheap 1/4-20 hardware will work better and be significantly cheaper than a ton of magnets. But I may be wrong.
