How round should round be?

I've built a mendel, ramps 1.2, using the latest marlin, slic3r, and pronterface.

Everything is completely calibrated as per the wiki, and I get great, accurate prints...so long as there are no curves...see the attached for what I end up with when printing this shape: [www.thingiverse.com]

(I've also tried sprinter with the same results)

I have the X and Your belts tight (could they be too tight?) gears are screwed to the stepper shafts with no play. The carriages move smoothly with no obvious rough or sticky spots.

I already fixed a loose hot end that was flexing and made this much worse...

Any suggestions?

-RobA>

Assuming that is the orientation that the part came out of the printer it appears you have a bit of lash in the x axis. I had a similar problem when I started out and noticed visually that my x carriage would actual twist slightly before changing direction from moving left to moving to the right or vise versa. I had reasonable success with the lash plugin in Skeinforge (setting the X axis to .3mm) but I don't see a similar option in Slic3r. In the end I changed the location the belts attached to my extruder assembly to minimize the twisting motion. Not sure if this will help your specific problem but figured I would share my experience with similarly warped circles.

Thanks for your reply, joerlane.

I printed another circle piece (15mm diameter) on a square base so I could reference it better. I also noted the head travel was counterclockwise around the perimeter.

A bit of analysis gives me the attached.

To me, it looks like that when X axis reverses direction, the carriage stops moving for a bit, than catches back up again. My belt is tight, setscrew tight with no sprocket rotation on the shaft.

You backlash could be a good suggestion, I'll have to look closer. The other thought is belt stretch or belt tooth/sprocket slippage or mismatch, but they came from the same supplier....

How tight should the belts be? How movable are others' carriages with motors powered off?

Mysteriouser and mysteriouser....

-Rob A>

Edited 1 time(s). Last edit at 04/11/2012 03:35PM by RobA.

You should be able to pluck the belt and get a bass note.

The carriage doesn't move easily with the belt attached because the motor has quite a lot of detent torque when not powered. It gets a bit easier if you un plug it from the controller. It should be free when the belt is removed though.

---

[www.hydraraptor.blogspot.com]

Found the source of my problem...

The vertical plate on the x-carriage lower part (that the the x-carriage belt clamps bolt to) was broken, probably due to over-tightening. This was causing that part of the piece to both rub on the rear x rod causing drag, and the belt anchor to twist when the belt changed direction. You can see the break and the wear on this part here:


I cut this block off, and separated it into two bits, so I would have the spacer necessary to screw the belt clamps back in. I just looped one of the belt ends around the block and locked it with a zip-tie, and used a washer (so roll the orientation 90 degrees) with the belt looped and locked with a zip-tie on the other end. I then use a thicker zip-tie to connect the belt (washer) to the other belt clamp block, and tension to the right amount.


I think I'll stick with the zip-tie tensioning scheme as it makes it trivial to remove and reattach the belt for any reason, and give me tension adjustment as accurate as each click in the zip-tie.

So now the flats are gone from my round prints, and I have to to recalibrate again.

I think once I'm calibrated again I'll be printing up an open-X carriage to replace the sells design, as the sells carriage was a royal pain to perform this "maintenance" on.
-Rob A>