I come from a recent background in 3D printing and I’m just about done with the printed parts for the MPCNC. I hope to do very, very fine detail with it in materials like aluminum, PCB, and (hopefully) brass. I’m worried about the maximum resolution even after seeing some of the work done in the gallery section and the PCB threads. Here are my specific questions:
What spindle/router should I buy for very fine work? Is the Dewalt good or should I go for a Dremel or Proxon or something else?
Should I get .9 degree stepper motors?
I’m most comfortable with smoothie, so I was planning on going for an MKS SBASE v1.3 for my control board–to get that sweet 1/32 microstepping and beefier processor for high-resolution gcode. Is there any work towards getting the dual-endstop auto-squaring feature to work on smoothie? The SBASE has drivers for 5 steppers, so I can’t imagine it being much of a problem…
I’m worried about the surface finish on the standard conduit contributing to minute defects in the tool path. Is there a preferred provider for (polished?) stainless steel tubing?
Lastly, am I worrying too much? Am I expecting too much? Here’s a picture of the kind of the highest level of work I hope to do:
[attachment file=58626]
Can I get some pics of your finest detail work in here? Post them up!
Most people are actually going for giant parts, not so much for detail. I think it could be possible, use stainless and build it small to make it easy. Make sure your spindle and endmill have low runout.
To give you numbers I had to increase the arc resolution to .15mm per segment. Detail is possible and should be relatively easy, most just want giant aluminum parts (or at least they always ask about it)…
Thanks for the response! The plotting project does look very promising!
Any progress on making the dual endstops work on smoothie? I’m sure it has to be easy enough with how easy the config file is to work with for most projects. I just don’t wanna buy another smoothie and find out I can’t get it to work.
I looked more into stepper motors and sparkfun has some 68OZ/in, 0.9 degree motors. Obviously more torque is better but if anyone has experience with slightly lower torque motors I’d be happy to hear about that experience, too.
0.9 degree isn’t really needed, we currently have a theoretical resolution of 0.001mm or 0.0005mm depending on your step rate.
I have no experience with smoothie firmware coding or really even it’s capabilities. It took more than 700 lines of changes into just one of the marlin branches (we did tw0) to just add it to the two axis that didn’t currently have it. I am not up to a smoothie project at this time.
You are kind of chasing the wrong question or answer.
At great details it comes more down to physics and skill than electronics. Every decimal place of precision adds several zeros to the price tag. Carving at great detail (or all carving really) takes a lot of skill, as much as it does great hardware. That is a reason jewelers carve a wax ring and cast it instead of just carving out the ring, either way it still needs to be polished but the harder the material the harder the Gcode is to make.
Did some asking on the smoothie forums and found an easy way to do auto-squaring! Once mine is up and running I can write a guide. Here’s a link to the process:
Hey, while I’ve got your attention, will you sell me a bolt kit? Lol. I hate sourcing bolts at the local hardware place and sourcing them online will mean buying way more than I need. Seems like a good addition to the shop for those of us that want to use other control boards. Basically just all the bolts, screws, and nuts in the main MPCNC kit.
Well, I just placed an order that contains no bolts Q.Q If you feel like adding some bolts to my boltless order, I’ll throw some more dollars in the donate tab.
I am about to try 12 tooth, GT2 pulleys on one of my MPCNCs. I turn down the micro-stepping to half-steps to increase torque with my super cheap stepper motors, and hope the 12 tooth pulleys will both increase torque and increase precision.
Is there a reason the kits all include 16 tooth pulleys? Is there a good reason not to try smaller 12 tooth pulleys?
I don’t think this will do any noticeable increase, and you can introduce oscillations as it bounces when it locks in in larger steps. We already get the benefit of this when a current step locks into a half step or full step position. you lose the ability to lock into smaller positions. The steppers are not a weakness or a resolution issue. We don’t skip steps and the theoretical resolution is higher than most people can accurately measure. Your time is better spent perfecting your CAM, that is the low hanging fruit here. Better CAM = more accuracy.
Yes tighter arc = more wear on the belts, and less tooth engagement = much higher chance of skipping a tooth.
