Comment on Partially 3D printed CNC machine
philpo@feddit.org 3 weeks ago
Interesting concept,seems to mitigate some issues of the MPCNC. The Z Axis assembly looks a bit on the weak side, especially for heavier spindles. I see the most problems in CNC builds here,tbh - the Z axis is far more of a problem than the other ones as it tends to destabilize itself due to the lever effect it does suffer from.
BTW: In terms of Google: Why not put the file on something like Github/Launchpad,etc. so others can contribute,comment or fork it?
Thanks! Indeed the Z axis rests on only one MGN12H block. For stability it can be changed to two sliders per rail at the cost of work area. For circuit boards and wood it should be strong enough to handle the load.
As for github/launchpad, I never used those before for sharing anything. I will look into those.
The issue with the Z axis is not only the slider but also the threaded rod. Just a few thoughts/experiences on it based on my experience with the MPCNC and various different changes to Z axis group people made:
rigidity of the axis is paramount. When the router moves it does move toward the material. If there is not enough rigidity the material will push the router bit away variably and an angle is created which then exercises pressure on the router bit, creating enormous loads and usually kills it - which is bloody dangerous.
The other issue is the weight of the spindle. A makita RT0702C is around 2kg, but has its drawbacks. A water cooled “China spindle” weights around 5. That is, from my experience, too much for one threaded rod.
Don’t get me wrong, your design is great and imho you are onto something - it removes a lot of other issues the MPCNC has and might be the perfect middle ground between “MPCNC” and “full aluminium build”. I like it a lot for this.
Nice to hear! At the moment the spindle, if it has the right to be called that, is a 775 motor with a chuck attached to it. Which is good enough for circuit boards and wood. Perhaps aluminium is doable as well as long as the feed rate is kept slow enough.
Should milling steel ever be on the table I’ll make a larger machine and will use leadscrews on X and Y instead of belts and perhaps dual Z motors.
I’ve also designed the build around materials that were readily available such as the 18mm X rails. Apparently they’re quite obscure since they aren’t sold on typical Chinese webshops.
MysteriousSophon21@lemmy.world 2 weeks ago
Spot on about the Z-axis being the critical point of failure in most DIY CNCs - harmonic resonance at higher speeds will absolutley kill accuracy before anything else.