CNC And 3D Printing

The obligatory Thingverse.

Vim’s G-Code highlighting: set syntax=ngc

Wikipedia G-Code and RepRap G-Code

Sensible tips for calibration of the step motors for good dimensional accuracy. From the same source, how to calibrate extrusion thickness.

G0 - Rapid position, fires axes independently to get each axis' position as quickly as possible.

G1 - Interpolate, slows an axis down if needed to create a straight line.

G21 ;metric values
G90 ;absolute positioning

Model: Monoprice 113860 Maker Select v2

Marking: IIIP

Work Envelope: X=200mm Y=200mm Z=150mm? (Z=130mm safe w/ spool on side)

Filament: Standard 1.75mm - up to 1kg spool

Tip: .4mm, I measured the extruded filament at .011" (.28mm)

Based On: Josef Prusa (sometimes used as closest options, e.g. "Prusa i3" in Cura)

Putting it on a rigid flat surface helped more than I would have liked. In other words its intrinsic structure is not quite enough to keep it aligned.

Don’t move it once it’s set up.

Move the z limit switch up high until your table can’t reach it even with all the springs uncompressed. Move it just low enough to start compressing the springs. The idea is that you do not want the springs all tightly compressed. Setup is harder that way. OTOH, with the springs lightly compressed, the leveling nuts back off during operation and just fall off! True story!

I used an indicator which was probably a mistake. Try to tram in a square 100mmX100mm in the middle. If the outside corners are off, oh well; they’re not usually as important anyway. If the middle is good, it’s also a good starting point for getting the corners ok too.

Watch interference with those binder clips holding down the glass. That’s a bad system that needs to be improved upon.

It is entirely likely that once the bed is heated, the bed will be a different alignment.

We had serious problems with the first layer sticking. The first thing to be clear about is that the part program seems to run a code that goes to the limit switches and uses them as zeros. This means that even if you set a zero right, they can be unset when you run the part. You need to have the limit switches physically placed so that Z homing from OctoPrint will put the nozzle perfectly over the table at the correct distance. What is that distance? A .12mm/.0047" thick piece of paper seems about right. It’s helpful to note that a reasonable Z layer step is .15mm/.006" and being much higher off the table when starting is unlikely to be successful.

Our Z limit switch had a slot for adjustment but only in discrete locations. I found this annoying and opened it up to a proper fully adjustable slot with coarse sand paper. But it can but solved even without that improvement. Just remember that you’ll have to move the table to match too if you use a new position for the limit switch.

WARNING - The "Stop Print" will only stop after it finishes buffered commands which can be substantial.

This video covers a lot of mods that can be done for this printer. Lots of good links in the video description.

These diagonal braces are exactly what I was thinking of designing because the Y to Z perpendicularity is shite without some kind of help. Not only that but the lack of triangulated structure seems to exacerbate vibrations.

I need a spool holder. Perhaps something like this.

It might make sense to remove one of the four Y axis bearings and move the lone one to the middle. Actually this guy says that mod did not wear well. But the thicker baseplate makes sense. Ordered!

The bed springs are madness and should be replaced by a second nut. Level with nuts between the plates; lock with nuts underneath. I would make easy to turn lobes calibrated to freedom units so my indicators are happy.