The CoreXY is awesome and I am trying to figure out if I should attempt this with a belt or with the chain. The belt would offer greater precision and control (I think) but then I wonder if I need to upgrade to a Nema 34 motor based on the size of this and the potential forces/load over the distance. Thoughts?
one thing to note, they show the two belts/chains crossing, if you have one
above the other, they won’t hit each other (they will just connect a different
heights to the sled)
look on youtube for the ant pcb mill, it’s a good example of this (showing
hooking the sprockets up at different heights for the two belts)
If anyone wants the parts that I am going to upgrade past, please let me know. They are solid nylon (MJF) and super solid. I have the parts that TomD designed for 1.5" square tubing and the top brackets that I designed. Someone might find value in using them even if they dont work for my crazy project. Happy to ship them if you pick up the cost of the shipping. The material is amazingly dense and taps really well. I have it set up for 608 bearings and M8 bolts.
i think the cost for the 3m ball screw is $180. Rack and Pinion would be about $100 USD plus the cost of the gear and steppers.
why do you think that a belt would be more precise? (the only reason I can think
of is the cogging effect of a small sprocket but that’s a pretty minor effect)
belts will strech more than chain, and not handle anywhere close to the same
True on the load perspective, but they are constant where as with a chain there is always something awkward with the last linkage attachment and geometry. With fiber reinforced belts the stretch would be minimal. But good points that you bring up. Obviously the more solid we can make the mechanism the better which is why I went to ball screws or rack and pinion
for the last linkage, get a bolt and cut the top down so it’s 1/8" thick, drill
a 3/32 hole through it and use a master link to connect the link to the bolt.
This gives you a very solid connection and a tension adjustment point i one.
For the stepper motors, I am trying to understand how much torque I will need for this. Will a 270oz Nema 23 3A stepper motor be enough for this?
I just found this thread, mind blown, good stuff all! Bookmarked!
Jeez, I guess I have been away for quite awhile from this thread. I just skimmed through it but it looks like my serious reading list has just gotten another entry for the next couple of days.
I am making rudder for my sailing boat. In Fusion360 I’ve design NACA hydrofoil profile, cut in half (like two half’s of wing) and generated nc file consisting of four operations:
Bore and plane went quite well - Maslow behaved normal but when I loaded Adaptive (a lot of long cuts on different depth) and Parallel (finishing passes along profile) everything almost stoped. GroundControl reacted on clicks within several seconds. Sending commands does nothing.
My question is:
Is there limit in number of gcode lines which Maslow can process?
Is there some exotic Gcode commands which can make GC to halt?
Should I chop code into separate pieces and execute one after another?
System do not crush - seems like it slowed down to almost halt making cutting impossible.
here is my G code:
Any suggestions?pletwa1.nc (411.7 KB)
If this is the issue, my guess is that your gcode is made up of too many small segments and the limiting factor is the time it takes to send each segment over the serial connection to the machine.
Turning on buffering of gcode in the settings may help speed things up.
Thank you @bar - I will try it tomorrow
Thank yoou @bar very much - I have firmware and GC version 1.13. Cannot find in settings option for buffering. Where should I look for it?
Darn! My version 1.13 do not have this option! I double checked it. The last position is “Z-axis safe travel”
Should I upgrade to newest versions of firmware and GC? Will I loose calibration settings?
Thank you bar - I really apreciate your help
Ahhh gotcha! Yeah it was a later addition to solve exactly this problem.
You can update safely without loosing calibration. If you are worried about it make a backup copy of the .ini file where the calibration is stored.
Here’s where that file is for all the different OS:
Hi Tom, could you resend the technical drawings for the frame? I also want to use this setup for my Maslow.