So I’ve been looking at doing a foam ripper for @DougJoseph 's kinematic mount, using stuff that I have.
So far the hardware list:
280 DC motor
DC motor gear aet (tach says the motor turns about 15000 RPM in free air, so I’m going with a 2:1 gear reduction for better torque at lower speed) gears fit 4mm friction fit rod and have bushings to fit a 2mm motor shaft. 10, 20, 30, 40, 50 tooth gears in the set. I plan to use the 20T and 40T gears.
PWM motor speed controller
0.020" rod as a needle
1/16" copper tube as a needle guide.
1/2" brass strip for reinforcement (will solder the tube to it.)
4mm steel rod
MR104ZZ bearings (4mm ID × 10mm OD)
So the idea will be to feed the motor controller 12V probably same power supply as the NEJE laser, then run the motor at appropriate speed to generate about 6k RPM at the driven gear.
I will use a drill (nominally for cleaning a 3D lrinter nozzle) to make a small hole in the driven gear about 4mm from the gear center. This will give an 8mm throw to the needle, good for the 1/4" dollar store foam board.
The needle guide will be the copper tube, soldered to the 1/2" brass strip. This will also give the needle guida a heat sink. I may further attach a heat sink to the end of the brass strip if it gets warm.
Doug… nowhere in your docs do you specify the radius of the bolt circle for the kinematic mount, but I reverse engineered it as 21mm. I hope that’s correct, but it’s parametric, so that should be OK. It’s easier to make my own mount than to remix an STL for this…
I am having a bit of trouble arranging the components in such a way as to be able to 3D print a mount. I wpild.like to make it a single piece if possible, but I think it will end up 2 piece if supports are unacceptable. (And to me… they are.)
I forgot where I put the bearings, I need to check them again for fit on the 4mm rod. IIRC they are a friction fit. I believe that I needed to ice the rods in order to fit them inside the beadings. I hope I’m right.
It occured to me that in order to use the gear as a crankshaft, I cannot have the axle go all the way through into bearings on both sides. The axle will need to end at the gear face so that I can have the needle driven in a complete circle.
So the revised plan is to hold the 4mm shaft with 2 bearings at the end, then have the gear at the other end, with the needle held through the gear with a “Z” bend.
I’m starting to lean towards the idea of making the foam ripper body in 2 parts, but I have one more idea to try first.
Well, it seems to be at least feasable. I tried something that I wasn’t really sure about for mounting the guide tube, which needs to be changed, and I dont have a mount for the speed control at all with this iteration.
Also, it extends below the LR3 core by more than I want (though less than the long endmills that I use) so maybe I should shift the motor upwards.
Found the bearings… they will slide on the 4mm rod, so I will need to figure a way to keep the axle from wandering. I wonder if I could file a groove into the rod for an E clip… maybe chuck the rod into the drill press and hit it with the dremel cut-off wheel… I have to cut the rod to length anyway.
Needle holder. I am not satisfied with this clamped holding arrangement, so I have something better for the next print. It will use some M3 acrews. I think it will be fine screwed into the plastic though. Not my finest hour with the soldering iron, but it’s functional.
I have the grooved 4mm shaft in the picture, but it has some burrs on the edges so I need to smooth it out a bit.
I have extra, so it was OK. The tube is still visibly centered on the brass. It is well.within tolerance. The wire will need to flex as it will be driven by the gear. Some lubrication will be necessary. It’s all “.proof of comcept” at this stage.
I used a.120W iron because my butane torch is out of gas.
The grooved rod seems to be OK, but I cut it a couple mm too long, so I.need to shorten it. There’s no room.to get the needle past. at least it isn’t too short!
So finally got back to this. I had to cut a couple of M2.5 screws to attach the motor, and melt a pinhole through the driven gear to drive the needle. About 9mm of throw seemed adequate to me.
This is running at the lowest speed setting that it will keep moving at. My tach battery died, so I will get the max RPM after I get another 9V battery from the dollar store. (It is, of course much faster, I think it should top out about 8000 RPM, a little better than the target 6000.) The motor itself seems to free spin 1 gear about 18000 RPM, so I figure a small loss from drag.
I didn’t have the gear placement finalized for this test, so the one bearing is not fully pressed into place. (It is now.) It sits flush with the plastic once it’s fully seated.
Almost everything here was stuff I had around the house. A 280 motor, DC PWM speed control (intended as an LED dimmer, actually) the plastic gear set, the .020" rod, the copper tube and brass flat stock, solder… just needed a 3D printed frame to hold it all.
Edit: I added the 3 screw holes to secure the brass plate, but when I put it in, they seemed unnecessary. The spool of yellow PLA there has had some go through my last 5 printers now…
It’s a fun mechanism to play with, isn’t it, Dan? Yours will be one of the lightest weight designs I’ve seen… it’ll be interesting to see how it holds up to a half-hour cut job.
Early on, I used DC motors out of a junked HP inkjet printer. It made a fine needle cutter but, when I started printing needle cutter bodies, after prolonged usage it’d warm up and start softening the plastic on my printed flywheels… the motor shaft would heat up from within the “bowels” of the motor itself. Heat is the enemy here… anywhere metal contacts plastic will need to be dealt with. Traditionally, with most needle cutters, the friction-heating problem area is where the needle enters the needle guide… and most often we mitigated the problem by fixing a wooden plate to the printed cutter platform and mounted the guide to that. Here’s one of my later cutters with a laser-cut stack of craft sticks bolted above and below the platform… with an oiled cotton-wad trapped in the plastic cavity between…
My earlier wooden BuildYourCNC machine ignored all that, however, and just ran and ran and ran. A 24V motor running on fixed 12V for about 6000 RPM and a machine max feedrate of 24 in/min (609 mm/min); i.e. all very relaxed and stress-free. I’m pretty sure this is where the 6000 RPM cutter target speed originated as well, certainly for me…
You should actually target the 8000 RPM (no load, free-running)… once the needle engages the foam and starts cutting you’ll drop a couple of thousand RPMs. I always shot for 6000 RPM (under load, while cutting). And I used a “rule of 10” for the ratio of feedrate to cutter RPMs – 600 mm/min for 6000 RPM, 800 mm/min for 8000 RPM, etc – and get 10 perforations per millimeter of travel… 10-15 perforations per millimeter gives nice cuts in DT foam board (paper on).
I read that. It is the primary reason for the copper tube and the brass plate that it is soldered to. That makes a large, conductive radiating surface, that also is pretty good at holding oil.
The nylon gears should not pose a problem directly from the motor. I am slightly worried about the hole where the needle passes through the gear though. Thst needle will be spinning in the hole quite fast.
I figure that if it opens up, what I will do is embed more of the copper tube through the gear to act as a bearing surface.
I wanted to gear down the speed in order to maximize the available torque from the small motor. I was originally going to go to a 4:1 ratio, but after measuring the free spinning RPM, I decided on 2:1. Even then, I wonder if I may now be under speed. Well, 8000 RPM is the goal at full power. if that works out, then it’s all good. If I have to cut a bit slower, then so be it. Gotta be better than the mess that the router makes!
Though I will say that the black foam board cuts very nicely with the laser. It’s $2 per sheet while the white is $1.25 per sheet. Maybe still worth it for the laser cuts… Now that I have built this though, got to at least test it.
We’ve actually had designs that used fans to help conduct heat away if that is a need… similar to the part cooling fan on 3d printers.
I didn’t realize your gears are nylon… or how the needle was attached to the gear. I would expect it to wallow out over time… however, as you said, using copper tube to bush it out would probably be a good first step.
Routers do make a mess. That always seemed to shock folks… that the needle cutter really doesn’t generate a lot of debris. The rapier through flesh and soft tissue/organs analogy, though yucky, seems pretty accurate.
Now that I’m a “laser guy”, I wish black foamboard had been readily available when my late buddy and I were building and flying RC planes in my pasture. But it wasn’t. Necessity is indeed the mother of invention and we did what we could with what we had at the time. And having gone this far, please do see it through… at least far enough to satisfy your curiosity. IMO that’s the fun and value of it all.
The story of 1000 unfinished projects in my basement/garage. My ADHD tendancies get me started on a project, and usually when I get the problem of “how does this work?” solved, I often lose interest when it’s just a bunch of work to finish it. Most often, I get a “proof of concept” build done that is functional, but maybe not so pretty, usually fully intending to refine it to a nicer model, which often never happens.
Sometimes I try to skip the proof of concept model, and the results vary. It is different when I do,so.ething for someone else. There I can usually get something nice finished.
The CNC actually helps a lot, since the difference in the “proof of concept” and “something nice” involves a lower difference in the amoj t of work, and figuring out the problem of designing something nicer holds my attention longer. The 3D printer helped that way, too.
As it sits, the needle cutter works. I will want to get my tachometer working so I can do some performance metrics on it, but since Dollarama just raised the price on the white foam board, but not on the black (now $1.50 and $2.00 respectively) the price advantage of the white over the black stuff has been reduced, and I can cut the black stuff with the NEJE pretty fast, no hold down problem, not even masking tape necessary…
I will see how it works. I have some FliteTest plans ready to import into Estlcam.
at least it isn’t too short!
