Mostly Printed CNC Tube Notcher

I wish I was terrible with engineering stuff like this! Unbelievable project!

2 Likes

Hi Gato,

Thanks for the questions! It’s going to be a while before it runs in anger, I just became obsessed with this mechanism and had to build it. :thinking:

Because I haven’t sorted the workflow out yet, I can’t really give you crisp answers. These two steppers (rotation and feed) will basically just substitute for the X and Y steppers. I intend to unwrap each tube, flattening it in Fusion 360 to create simple line perimeters. The SKR will think it is cutting a 2D shape, but instead of scrolling X and Y it will actually be rotating and feeding the tube. I will either have to go back into the Marlin firmware and calibrate steps/mm or warp/scale my 2D design before creating the cut instructions. I went through and customized Marlin for my 1.3 already, to allow me to use all of the heating circuits to control SSRs, and remove the temperature safeties, etc. so I am hoping that this won’t be a big deal.

I have to admit that it probably won’t be well synchronized with the LR, but I intended to make a plasma torch mount for the 611 plate, and I was thinking that I would run the torch over to index in front of the feeding apparatus. Truth is, when I power off and switch over from the LR steppers to the tube feeder steppers, the 611 plate will lose its holding power, so I may need to make this notcher thing into a standalone tool. Thanks, food for thought!

There is (currently) no tensioning to hold the tubes in the three gripper wheels. My intention was to add a single wind of some kind of friction tape to each annular gripper contact, until it held the tube tightly. There should be very little drag that the feeder has to fight. I may also have to look at the acceleration and jerk in Marlin if mechanical tape wraps alone (compression) can’t suppress slippage. I have a feeling that this thing won’t tolerate too much beefing of the tube grip before it just won’t run at all. These aren’t massive steppers after all.

This is probably premature to show anyone, I was just pumped up. :stuck_out_tongue_winking_eye: If I redesign this thing it will have some improvements to allow easier assembly for sure.

Cheers!
Brian

2 Likes

Thanks Dan, but I really don’t math well or have any normal engineering skills. I figure that an ME would look at what I was trying to do and say “oh yeah, you just need to use a hyperboloid sinus grind on your reluctor teeth” or something. My fusion models are awful, all my sketches are riddled with horrible geometry shrapnel, etc.

I also Fusioned and printerated what I think is a cool controller box, designed to clamp to the (eventual) CNC unistrut rail. It holds an SKR V1.3, the TFT28, an 80 mm fan, and has some other silly features. All the cabling will go through the unistrut clamp and be hidden in the strut. Eventually. :slight_smile:

8 Likes

But what about this car?

1 Like

Hi Tony,

I know, I am asking me the same thing. :sweat_smile: I have to finish the shop and tools first, then I can start on the car. Two seat, mid engine, manual trans. I kind of want to go for a 1960’s vintage GT vibe, with shades of Jag XJ13, Lola T70, Porsche 917, etc. That is a looong ways down the road. :roll_eyes: I want to design and build it unique, from scratch, so when someone asks me if it’s a kit car or replica, No It Is Not. :slight_smile:

Cheers!
Brian

2 Likes

That’s great! Make sure to create an off topic thread, because there are a few gear heads here (present company included) who would enjoy that, and a bunch of other people who would just dig a project that complicated.

Man, a car from scratch is NOT for the faint of heart! I’m looking at something like the Midlana for my next build because I’m not an engineer by any stretch of the imagination, but I’m not so sure about the styling on it. At least by then, you should have this tube notcher figured out!

I’m in the middle of a Factory Five build right now, but mostly stalled at wiring. Decided to rebuild the whole damned harness and then I thought I’d put together a cnc to help me with one-off fiberglass forms and the next thing you know I’m cutting all kinds of stuff for the wife and friends and now we have to move so we can build a house with a bigger shop, lol. But I promise, my thread is coming soon because the weather is finally under 110 out here and the house is under contract (we’re in a rental).

Make sure to take really good notes on the design and build. You might have a book in your head. The guy who designed the Midlana (Kurt) sells both of his for $100 each. I’m not sure how many he sells, but I get the impression at his forum that at least 100 guys have one, and he probably gets half that (based on what I read at the on-demand publisher’s website).

4 Likes

I’m IN! :grin:

1 Like

Show a video of this running please

Hi Doreen,

Sorry, I don’t have one yet. :slightly_frowning_face: It’s going to be a while. It has more internal friction than a tectonic fault line, so I will be redrawing and upgrading the stepper mounts and motors, and revising how the stepper shaft joins the printed drive shaft(s). The redesign will allow me to use a NEMA 23 (I have a 425 oz-in unit on the way) to overcome the internal friction. I also need to figure out how to use the step-enable-direction pins from a BTT SKR V1.4 to run a DM542T external stepper driver for the beefy steppers. It seems like it should be NBD but I always think that right before I fail spectacularly. I hate to be honest, but it might be 6 months or a year before I can post a video of running this for the first time.

If it works as intended I may go back to the drawing board and redesign it with a gripper carrier that would allow swapping of the grippers (to accommodate different sized tubes) without having to take apart the whole thing. Cheers! Brian

1 Like

This is an incredible machine and design. I am sure this is going to be something you already know. But whenever I make pla stuff that fits together, wax works really well to smooth it out (even on stuff like herringbone gears) and it loosens up a lot when the stiffest peaks wear out, which is pretty fast. If it doesn’t turn at all, then you can exactly wear it in. But if you can move it, it may get loose pretty quick.

2 Likes

Hi Jeff,

Thanks for the feedback. :grinning: I did also suspect that it would smooth out after I had “run it in” and it really has helped to keep spinning the green and orange gears by hand. Before assembly I had quickly flamed the parts with a torch to get rid of threads, then I chased all the gear teeth with folded sandpaper, and I think I knocked down a lot of the nubs that would cause early stiction. Comparing now to when I first assembled it, it has really improved. Honestly even early on it rotated fairly easily. It was always (and still is) the tube advance that was stiff. I’m not using wax, but I will look into that (as in paraffin or something) next time. It does work just like the simulation, so take that, 2020! :crazy_face:

Cheers! Brian

1 Like

DAM.

That is one hell of a project. Can’t wait to see it in action. I just found this thread and I was scrolling thinking…please make it, please make it…boom there it is. For your first shot, it looks amazing. So complex and it all works well. I never get that lucky on my first couple tries. The Primo core is V19 in my file system and there were many attempts that did not get numbered…and it is not nearly that complex.

2 Likes

Hi Ryan,

Thanks for the kind words from the master, but this took me so long (seriously, months of time), I don’t think it is really a legit “first shot.” In Fusion it says it is V91 :slight_smile:

And let me thank you for organizing this website/forum/knowledge collection and sharing your CNC designs (for free, no less!). After finding a link to your CNC router design (Low Rider, because I needed full 4x8 sheet ability) I was over the moon thinking about the possibilities. I mean, seriously, the LR has dominated my free-time thoughts for a year and a half.

Best wishes!
Brian

5 Likes