Effects of microstepping and segmentation on print qualty

[nebbian]

Bruceg,

Interesting prints. Thanks for doing this. Would you say that the golden ratio print is smoother?


btw what steps/mm are you running at the moment? From some of your earlier prints it looks a bit like you're running a steps/mm value of 100 (or 200). Is this correct? It's just that the lines seem to repeat with a period of 5 layers, instead of 1 or 2. If so then your golden ratio will be 0.129mm layer height.

[bruceg]

Yes, the golden ratio print is smoother in that there are longer sections in the xy plane without deviation and generally less artifacts in Z. Sorry my lamentable photography is not showing the results very well.

The machine is 1.8deg steppers, 20tooth pulleys, "2mm"/step belt and 16ustep (Repetier 0.92.2 + v1.3 Rambo) so is set to 80steps/mm (i am trying to save up for a Duet and 0.9 steppers).

One thing i have not checked since assembling the machine is for loose pulleys on the axis stepppers. Another potential problem may be the crumbling ATX style power supply which is down to 11.someV with the heaters off so perhaps that should get sorted before reading too much into my results.

However it might be interesting to muddy the waters a little further in the mean time by rotating the model around the z-axis to see if there are any clues re mechanical deficits.

[626Pilot]

Setting the layer height to a multiple of the golden ratio (once you figure out how much distance a full step takes) spreads the peaks out enough that they stop being visible.

Based on Bruce's last few prints, it looks like the Golden Ratio idea works pretty well. I see only straight lines, with little to no evidence of a Moire pattern. How did you come up with that idea?

[nebbian]

Based on Bruce's last few prints, it looks like the Golden Ratio idea works pretty well. I see only straight lines, with little to no evidence of a Moire pattern. How did you come up with that idea?

How does anyone come up with their ideas? In the shower, of course

It all started with me looking for the cause of some lines in my prints, and went down a big long rabbit hole involving segments, driver decay modes, firmware algorithms, and lots of test prints. I had a couple of puzzling results, and a couple of Ah-HA! moments. Some of this is not obvious, for example the effect that layer height has on the visibility of lines. It all clicks once you understand everything that's going on.

I'm going to write up my findings soon. First I need to take pictures of some of the 25-odd test cubes I've got lying around on my desk

[626Pilot]

Well, this is a significant find. I always thought Moire patterns were a necessary artifact of using Delta printers, and you could only eliminate them with some crazy high-res steppers. You've really got something here.

[nebbian]

More data from a user on facebook with an MKS SBase (32 bit board, 8825 drivers)

Left print is printed at 0.202 layer height, right print is at 0.161.

From this I can see:
1) That he's using 20 tooth pulleys, so his steps/mm value is 80 or 160
2) That his drivers are not in the best decay mode

For him to get better prints, he should print at one of the following layer heights:
0.081, 0.161, or 0.243 mm.

After doing some research, his drivers are in mixed decay mode by default.


I'm starting to really wonder why no drivers ever seem to be in the best mode by default. Why should we always have to mod them???

[bruceg]

drivers are in mixed decay mode

For those of us who might be neophytes such as myself and are trying to follow this, i found the information in the following links helpful to begin understanding stepper driver microstepping foibles and phenomena.

Hackaday How Accurate Is Microstepping Really?

and particularly for the 8825

DRV8825 missing microsteps

[DerStig]

Interesting. My steps per mm is actually set at 80.7. I calibrated that with a dial indicator (modified it until a Z axis move command equaled the change in reading on the dial).

How did you come up with the 0.1618 layer height?

I used to set my steps/mm according to what the caliper told me, but recently when looking at this issue I've discovered that you can get moire effects. Mine was set to 161 steps/mm, when the calculations say 160.

The 0.1618 layer height is calculated as follows:

80 usteps/mm = 5 steps/mm (at 1/16 microstepping)
= 1 step / 0.2mm

Golden ratio * step size = 1.618 * 0.2mm
= 0.3236

Halve this to get a reasonable layer height:
= 0.1618 layer height.


The magic is in the golden ratio, it spreads out the single step peaks so that they don't line up with subsequent layers.

As you have 80.7 usteps/mm, the layer height you should use is:

0.1604 mm.

THANK YOU!!!!!!!! I was asked this question a few weeks ago

[nebbian]

I've collated my findings in a new thread here:

http://forums.reprap.org/read.php?178,763881

I hope that it helps someone, or maybe you might find it interesting.