weird bug at .19 and .39

I've seen this in a couple models, but it only happens at .19 and .39 layer height. This has been printed on other printers fine, so it isn't a problem with the model.

The surface is stepped during printing too, so it isn't just in the visual rendering. And the model is sitting flat.

This is the model loaded.

Screenshot2014-12-12080505_zps2ff49d43.j

Here it is sliced at 0.14. No problems.

Screenshot2014-12-12080526_zpsdce62447.j

Here sliced at 0.19 you can see the stepped surface. I get the same results at 0.39.

Screenshot2014-12-12080549_zpse1dc9894.j

Regards

Jamie

That’s weird… It might help to provide the STL so they can test it.

Good call John... Come on Jamie.... ;)

Check your angle. Maybe the part is not 100% flat/horizontal. When you have a very slight angle on a flat surface you sometimes see these single steps. And the step can shift its position depending on the resolution setting.

Check your angle. Maybe the part is not 100% flat/horizontal. When you have a very slight angle on a flat surface you sometimes see these single steps. And the step can shift its position depending on the resolution setting.

Nope, as mentioned above, the part is perfectly flat. I can double check by looking from under the virtual bed. The whole area is dark, meaning the part is sitting flat.

Plus that doesn't explain why at higher resolutions the step is not present. If it was there it would show in every layer height setting.

I've attached the file. You will need to rotate it 90 degrees to sit flat.

So anyone able to  confirm this ?

I tried on 9.1 - 9.5.. Did the same on every version..

Yea same problem for me. I opened file in solidworks and converted it and back to stl again and then it is ok in z-suite too.

Yea same problem for me. I opened file in solidworks and converted it and back to stl again and then it is ok in z-suite too.

Thanks Trhuster. That is strange as the file was originally created in Inventor, and I've never seen a problem like this before. I've also seen it a number of times with customers parts. But only at .19 and .39 ( Not that I print with .39, just noticed it).

It also works fine on the UP/Afinia.

Perhaps someone from Zortrax can chime in ?

Slicer is ok, some files from internet need to be fixed. You should check broken files with repair tool designed to detect and repair errors in meshes.

Fixed versus Unfixed hinge:

editedntfab.png

I used: https://netfabb.azurewebsites.net/

Strange thing...

If you open it in Netfabb is absolutly clean, shows zero problem.

Then you perform the repair routine over this "clean" file.

Put both on Z-Suite and you have same results as Sam showed.

Moral: never trust, always repair your file before printing.

Well thanks guys, it is very odd. And I'm still puzzled as to why it doesn't show the error at the other resolutions! ?

the world of bit and bytes is a mystery... :D

It happens because you can print multiples of layer thickness only. Half or third of a layer cannot be printed for example.

For instance part is 10mm high and resolution is set to 0.39mm. Part contains 25 layers. 10mm = 25 * 0.39mm + 0.25mm.

Slicer has to do something with that remainder after division which in this case is 0.25mm. If remainder is higher than 0.195mm aka (0.39mm / 2) new layer is created. If it is lower than 0.195mm aka (0.39mm / 2) nothing happens. In this case new layer is created.

Easy to understand but hard to explain. :unsure:

It happens because you can print multiples of layer thickness only. Half or third of a layer cannot be printed for example.

For instance part is 10mm high and resolution is set to 0.39mm. Part contains 25 layers. 10mm = 25 * 0.39mm + 0.25mm.

Slicer has to do something with that remainder after division which in this case is 0.25mm. If remainder is higher than 0.195mm aka (0.39mm / 2) new layer is created. If it is lower than 0.195mm aka (0.39mm / 2) nothing happens. In this case new layer is created.

Easy to understand but hard to explain. :unsure:

Hey Sam, makes perfect sense. Thanks.

I think you’re wrong.

Is the problem is how the slicer interprets the relation between layer thickness and height what netfab corrected?

This was part of the problem but not all.

I have similar problems, but for me solution is... rotate part :). Rotation depends from 3d model. For example for some parts I have good results with -270 in x-axis and with others +90 in x-axis.
 
This part:
load_zpsd19a9b56.png
 
When I want print this in horizontal position then glitch shows for +90 in x-axis:
glitch_zps766393ac.png
 
But it's not appear when i give -270 in x-axis:
NoGlitch_zps6545fe81.png
 
It's still same part place and direction :D It looks like it's bug in slicer.

Rafal then tell me why 360 degree rotation of part in other direction remove this model "bug"?

Rafal then tell me why 360 degree rotation of part in other direction remove this model "bug"?

If you doing rotate in Z-Suite then every triangle coordinate is recalculated to new position and rounded to not so crazy floating long values as in original - I didn’t looked at slicing engine since year but so far as I remember floating numbers problem and flat triangles can happen and it is mostly bad job of software which saved .stl

YosemiteSam showed example of part re-saved with Netfabb and woking correctly afer this - nothing more to say.

Best Regards

Ahhh of course I didn't take into consideration that a flat surface can be badly written in STL :P My bad ;)

Anyway quick fix for this is … rotate the part :smiley: