My custom Cartesian

wilriker

Eh, I needed to fetch a beer from the storage. So I hopped in to the "workshop"
Anet steppers are 37,8mm long and Wantai real length is 46,8mm.

Thanks for that! Will measure on the weekend of I can fit them. Biggest issue might be Y axis as my heat bed holder has only a certain width cutout for the motor.

And the beer is imperial IPA with 10% alcohol. And it is brilliant.

I am all but a beer type but happy that you have something you enjoy.

OBELIKS

If it's aluminum bed holder, just chop it.

wilriker

@obeliks It is aluminium. Though, I have some concerns of messing with the statics of it if I chop it. But first I need to find out if it will be an issue after all.

wilriker

OK, I looked into the dxf file that is used to cut the bed holder and it has a clearance of 58mm for the Y motor. 46.8mm motor + 6mm damper leaves 5.2mm for the thickness of the motor holder (which I am not absolutely sure but I think is 3mm) and some clearance to compensate any amount of mounting skew... this will be tight... And as you can see on the image of the seller below there is not much room for removing further material without coming really close to some existing holes

heat bed holder

OBELIKS

Damper? I think those will not be needed.
As I said these are a lot quieter.
You should measure on the printer itself. Any other theoretical calculations will be off, by my experience.

wilriker

@obeliks For the current motors the dampers make a huge difference especially for the Y motor. I feel a little hesitant to assume that the "long" silent motors are even more silent without the damper - and they would at least have to be equal in noise to the current setup. But still it is tempting. And 5dB are nearly 4-times louder/quieter (depending on reference).
I never measured the actual difference when adding the damper, though, except for listening at them.

And of course I will have to also measure on the printer itself to be sure about the cutout.

OBELIKS

@wilriker A question regarding the dampers. Are they stiff enough not to introduce backlash?

wilriker

@obeliks Actually I don't even know that. I guess so because they feel very stiff. But of course it introduces another elastic element besides the belt into the system. So it will at least add a very minimal amount of backlash.

Now that I think of it my ghosting problems with infill visible on the outside of a Benchy might be backlash issues...

OBELIKS

Do you have a picture of that Benchy?

wilriker

@obeliks This is the best one I could find (and not a good one to be honest):
0_1535116706809_1130815247278436466352977980638233011.jpg

Given this is actually printed on the original acrylic frame but sadly the aluminium frame did not improve in this very issue in any way.

And this is not just visible because the filament is a tiny amount transparent but also you can feel it.

EDIT: There already was belt tensioner in use and I used glass-fibre reinforced belts from the start - I never used the steel-reinforced belts that came with the A8.

OBELIKS

This is not really ghosting, what value do you have for the perimeter/infill overlap?

wilriker

@obeliks I knew that this would come up. It's at 10%. And I am using 2 perimeters with 0.4mm each so a total wall thickness of 0.8mm.

OBELIKS

Haha
IDK, usually these kind of issues are down to slicer, temperature, filament errors. But you probably checked those before.
Just one more, is this PETG? I have seen something like that with it.

wilriker

@obeliks No it is PLA actually. But I see the same with PETG. That is one thing I have yet to get rid of - but maybe I need to try harder instead of all the time tinkering on the printer.

Regarding the filament I checked its width and AFAIR it had a little over-diameter but not enough to adjust anything. And temperature was determined by temperature tower.

OBELIKS

You know what. Send me sliced file, I'll try it here. We will be able to definitely rule out slicers if it is all OK on my front.

wilriker

@obeliks Uploaded it to my Dropbox 3DBenchy_75mm.gcode.

Of course you have to adjust start and end codes as for me this simply calls two files in /sys via M98 and temperatures are set at the beginning of the file as well.

OBELIKS

Good news, I have the same result. And I think I know what is the problem. Try disabling Print Infill Before Walls. When it was printing walls I could hear it dragging against the infill and I can feel it. With Slic3r that prints walls before the infill by default I never had this issue.

wilriker

@obeliks First of all thanks for your test print!

I have to check my settings what the actual value of this switch is tomorrow. I know so much as that I am using the default of Cura.

But interestingly I always thought that printing infill before walls was exactly to prevent the issue I am seeing. But my brain had so much input in the field of 3DP since I (only) started end of March in this adventure it would not be unlikely for me to mix something up here.

wilriker

OK, I checked the the default settings of Cura. As you said it will by default print infill before walls. In the setting's description it even says this can lead to infill sometimes showing through. I assumed "sometimes" would not mean "nearly always". Anyway, it also says that enabling it will improve overhangs significantly.

But then I also found a setting "Connect infill lines". From the description of it it might alleviate the issue so I get best of both worlds at the price of some more filament because this basically adds another perimeter on the inside.

I will try both after I switch filament next time because currently this is a borrowed spool that I can use for a specific purpose but I don't want to stretch the generosity by printing an army of Benchies.

OBELIKS

If you need more help from me just say. I can print more tests. Just prepare them for PETG.