What does my calibration cube tell you?

[Jimustanguitar]

I just got my Rostock Max up and running this week, and without any tweaks and tunes, I decided to pint a calibration cube out of ABS to get a baseline of its performance and accuracy.

http://imgur.com/a/40YzJ

What things do you see from this that I should note while I'm tuning everything in this week?

Observations of mine:
The top of the cube feels rough, there are a few holes where you can see into the infill spaces.
The bridging at the top of the holes didn't work well.
About every tenth layer you can see a line on all 4 sides where there's a gap or something going on with the spacing and alignment.
On the towers, the extruder left a small burr at the end of each pass where it lifted to go to the next section.


Thanks for the tips and tricks, I'm looking forward to printing this again with better results!

[cassetti]

Actually looks pretty good for a first print of the calibration cube. Looks like you might have 1 or 2 sticky delta arms (read the delta arm blues thread). Also check for excess side to side play. No arm should be able to freely move side to side.

Once you've checked your delta arms, try the single-wall calibration cube tests (get a 20x20x20mm cube (or bigger), set to 1 perimeter, and 0% infill, with 0 top layers, and measure the wall thickness (preferably with a micrometer) ) - I had issues with that calibration cube until I was able to print a proper single wall cube.

[Jimustanguitar]

Actually looks pretty good for a first print of the calibration cube. Looks like you might have 1 or 2 sticky delta arms (read the delta arm blues thread). Also check for excess side to side play. No arm should be able to freely move side to side.

Arms are good and tight. Brian from trick lasers, and John from seemecnc both commented that I did a better job than most on the arms and joints. I did have a loose bearing carrier thugh. I snugged that one up shortly after this print. Forgot about that, maybe that was the culprit.

When all 3 axes move to create a straight line, could a sticky point in an arm affect the whole layer on all 4 sides? (or all 16 sides of the 4 towers in the same layer?) I would expect it to create a wave in every layer, perpendicular to what I'm experiencing.

try the single-wall calibration cube tests (get a 20x20x20mm cube (or bigger), set to 1 perimeter, and 0% infill, with 0 top layers, and measure the wall thickness (preferably with a micrometer) ) - I had issues with that calibration cube until I was able to print a proper single wall cube.

I've got a 1-2" micrometer and a 2-3" one... Shopping trip tomorrow!



I know it's the last step to worry about, but what would help the bridging between towers?

[cassetti]

Yeah, a loose skate could do that. I had to tighten one up myself before. It could be the culprit for the shift. Though as mentioned, I had a major shift issue about 150mm up one of my prints (After a few layers, shifted back to original position), after I shimmed my loose delta arms, the problem disappeared.

In answer to your bridging problem, dunno, Haven't spent too much time playing with that. First, make sure you have the proper extrusion multiplier for your prints (using the single wall calibration). After that, if you're still having sagging issues with your bridges, try playing with your bridge flow rates under the advanced settings in Slic3r.

[Jimustanguitar]

I did the .5mm thin wall test to figure out a pretty decent extrusion multiplier for Slic3r. Here's the result of attempt number 2:

http://imgur.com/a/m8xK7

[cassetti]

Looking real good!

What extrusion multiplier did you end up using? When you say .5mm, don't you mean .55? Since rule of thumb is to go 10% larger than nozzle diameter, your single wall tests should be targeting 0.55mm, not 0.50 (someone please correct me if I'm wrong).

Did you enable retraction for your print? Or was it off? If it was, try with a 5mm retraction
Another thought - did you test your hotend with a thermocoupler to determine the true temperature of your hotend? You might be running a bit too hot?
Also, you can play with bridge flow rates - set to a lower number to use less plastic for the initial bridging.

About the time I reached the point you're at, I started playing around with larger models to start looking for other major defects in printing quality.

[Polygonhell]

Looking real good!

What extrusion multiplier did you end up using? When you say .5mm, don't you mean .55? Since rule of thumb is to go 10% larger than nozzle diameter, your single wall tests should be targeting 0.55mm, not 0.50 (someone please correct me if I'm wrong).

Did you enable retraction for your print? Or was it off? If it was, try with a 5mm retraction
Another thought - did you test your hotend with a thermocoupler to determine the true temperature of your hotend? You might be running a bit too hot?
Also, you can play with bridge flow rates - set to a lower number to use less plastic for the initial bridging.

About the time I reached the point you're at, I started playing around with larger models to start looking for other major defects in printing quality.

Extrusion multiplier should be close to 1m it's used to tweak flowrate and adjust for filament variation.
Nozzle diameter you should set to your Nozzle diameter 0.5 in most cases, it's only used to compute extrusion width is you don't manually set it.
Extrusion Width I would set to Nozzle Diameter + or - 10% and at least 1.5 x the layer Height, so I use 0.55 by default, but that's just a rule of thumb.

[Jimustanguitar]

My mention of .5mm was the dimension in the .stl file itself, not the Slic3R setting or my measured dimension. Sorry for the confusion... That is correct, right? (the .5mm in the file, turns into .55mm in reality, and also should be my "nozzle diameter" setting in Slic3r)

I printed two thin wall cubes and measured them with a mic to calculate an "extrusion multiplier" of .83 I didn't turn off retraction or do any of the other tweaks yet, but I didn't measure the side with the starts and stops in it because of this. Now that I've printed a few thin wall cubes and gotten consistent measurements at .55mm with that multiplier, should I just leave that as a Slic3r setting or should I do something with it in the firmware? Is this the right number to change by that coefficient in the firmware? http://i.imgur.com/AWwzBV0.jpg

Did the PID autotune last night, I want to do it again today so that I have 6 values to average before I apply the setting. I have not measured its accuracy with another device yet. I'm sure those will both help with my bridging. What bridge flow ratio should I try? It's at 1 right now, should I go down by .05 increments?

The bridging worked ok where the hotend traveled straight from one tower to the next, but most of the time it was trying to do a diagonal zig-zag that just floated in the air like a normal fill layer. That's a big part of why the bridging on this cube isn't so hot. I'm sure there's a checkbox for this that I'm missing somewhere

[Polygonhell]

My mention of .5mm was the dimension in the .stl file itself, not the Slic3R setting or my measured dimension. Sorry for the confusion... That is correct, right? (the .5mm in the file, turns into .55mm in reality, and also should be my "nozzle diameter" setting in Slic3r)

I printed two thin wall cubes and measured them with a mic to calculate an "extrusion multiplier" of .83 I didn't turn off retraction or do any of the other tweaks yet, but I didn't measure the side with the starts and stops in it because of this. Now that I've printed a few thin wall cubes and gotten consistent measurements at .55mm with that multiplier, should I just leave that as a Slic3r setting or should I do something with it in the firmware? Is this the right number to change by that coefficient in the firmware? http://i.imgur.com/AWwzBV0.jpg

Did the PID autotune last night, I want to do it again today so that I have 6 values to average before I apply the setting. I have not measured its accuracy with another device yet. I'm sure those will both help with my bridging. What bridge flow ratio should I try? It's at 1 right now, should I go down by .05 increments?

The bridging worked ok where the hotend traveled straight from one tower to the next, but most of the time it was trying to do a diagonal zig-zag that just floated in the air like a normal fill layer. That's a big part of why the bridging on this cube isn't so hot. I'm sure there's a checkbox for this that I'm missing somewhere

To reiterate hopefully with less typos.
Set the nozzle size to your actual nozzle size 0.5mm.
In Slic3r if you don't explicitly set the extrusion width in the advanced dialog then it computes it for you from the nozzle size, it will be MUCH larget than the nozzle size and you need yo look at the comments at the top of the generated Gcode to know what extrusion width to expect.
If you do explicitly set the extrusion width then the nozzle size is ignored. I recommend setting the extrusion width for a 0.5mm Nozzle to 0.55mm.

Print the single wall cube, the wall width should match the extrusion width (either the one you set manually or the one from the comments), if it doesn't adjust the extrusion multiplier, rinse and repeat.

[cassetti]

Interesting Polygonhell - I thought the nozzle size should be 0.55 - hmm, looks like I need to go back and do some tweaks and test the results!

@Jimustanguitar - FYI, Don't know about everyone else on here, but when I print these calibration cubes - I use a solid 20x20x20mm model. I then adjust my slic3r settings for 0% infill, 4 bottom layers (0 top layers), and 1 perimeter to create the single wall effect when printed