Odd results from Benchy Torture Test

[arseniow 0]

These past few weeks I have been tweaking and modifying a locally sold 3D printer. I have managed to remove the z ribbing but random z artifacts still persist. What I find particularly odd is that the print becomes really smooth when it prints the chimney part of the Benchy torture-test (you can get it herehttp://www.3dbenchy.com/about/).

Here is how it looks like. Notice the chimney being REALLY SMOOTH

Thinking that this could have been due to some weird accident, I printed it again. Both had smooth chimneys but everything else was rough.

I had a thought that z artifacts become a lot more pronounced when bigger objects are printed. I tested this be printing a hollow cylindrical column, a big irregular shape, and a medium sized cylindrical column.

Findings: The z artifacts on the small one was practically gone while the big one had deep z artifacts.

I am wondering why?? Has this happened to you? It is clear that the printer has the possibility of printing really well. The only hypothesis I have why the other parts are ugly is that this maybe a software issue on Cura 15.04.

Any thoughts about how to solve or what is causing this will be much appreciated.

Thank you in advance! Thank you for reading

[gr5 1,070]

Well if you have inconsistent resistance in one or both axis or if you have loose belts that could explain why it's better on the chimney. for example if you have quite a bit of backlash then I would expect the patterns I see in your 3rd (last) photo because cylinders don't care if the head moves clockwise or counter clockwise but when you get to a corner like the larger piece above it makes a huge difference and cura is seemingly random on which way it will go (clockwise versus counter).

So with power off hold a stepper still and move the head around to see how much play you have. The play can be in belts being loose or any part of the printer being loose enough to allow the nozzle to wiggle around without the stepper moving.

The second common cause of backlash would be very high mechanical resistance.

This is just one theory but the one that fits all your pictures the best: backlash.

Other theories would have to do with temperature stability - you could monitor the temperature with some kind of controller on the front maybe or through USB with pronterface (don't know what kind of printer you have). But I don't know why that would be different for bency's chimney.

also the Z stepper could be bad where it isn't moving the same distance very time - this is extremely common problem and can be anywhere in the Z stage but again I don't see how that would explain benchy's chimney.

also extruder inconsistencies or even filament diameter inconsistent but only my backlash theory explains everything as far as I can tell.

[gr5 1,070]

I see that both benchy's have lines in exactly the same spots. backlash explains that also.

You may alternatively have missed steps on some sharp corners. It could be that your jerk and/or acceleration settings are too high for your printer. Does your printer use Marlin? Tell me more.