Z-height of single layer wide box too high ... how to fix ?

[Mind]

I have a 20x20x5mm high thin wall box (sides only). I print it with 1 layer thickness in ABS with a 2% expansion/contraction allowance in S3D.

My S3D settings are:

Extruder:
=======
Extrusion width: 0.55mm
Extrusion multiplier: 0.9 (I will try change this later to get perfect fill once I have X/Y/Z sizes correct.
Infill: 0%

Layer:
=====
Layer height: 0.1mm
Top solid layers: 0
Bottom solid layers: 0
Outside/Perimeter Shells: 1

S3D slices it with layers 0 to 52. That means 53 layers. I don't know why it uses 53 layers ?!? With a layer thickness of 0.1mm, it should use 50 layers !

Part final sizes are: 20.4 x 20.4 x 5.27mm .... so pretty darn close. If I change the extrusion multiplier or extrusion width, I can get it down to 20.0x X,Y

I tried adding a Z-axis offset of -0.1mm, but that made no difference.

I cannot figure out how to get the Z height correct! It always adds 2 extra layers, it seems! Any ideas how to fix this ?

[jimc]

a couple things that i can think of that will affect this. the bed gap or z-stop bolt adjustment. there will be a little variance here depending on where that is set. in s3d you also have your first layer height adjustment. if this is anything other than 100% that will adjust where the layers fall then the top layer will need to round up a little. what you have going on there is probably just a couple small things but they are totaling up in the end. this may or may not be the problem. its just what i can think of off the top of my head.

[Mind]

a couple things that i can think of that will affect this. the bed gap or z-stop bolt adjustment. there will be a little variance here depending on where that is set. in s3d you also have your first layer height adjustment. if this is anything other than 100% that will adjust where the layers fall then the top layer will need to round up a little. what you have going on there is probably just a couple small things but they are totaling up in the end. this may or may not be the problem. its just what i can think of off the top of my head.

The bed gap is 0.15mm when cold with feeler gauge as the light on the Z-stop is triggered. I measured it hot too (> 90deg bed temp) and there was no real difference in feel with the same gauge.

I had the first layer set to 90% (default in S3D), so I have changed it to 100% and the extrusion multiplier to 1.0 and now the Z-height of the finished part is 5.4 mm .... GRRR !!!!!

Then I remembered that I had added 2% to the scale of the part, so I set that back to 100% and did another print... same BS...

I guess I don't understand how S3D takes a 5mm high part and slices it with 0.1mm layers and gets 53 layers ... then prints a part that is 5.3mm high.... so I went looking in the G-Code file.

The GUI in S3D is saying layers 0 to 52 .... But the Gcode says:
; layer 1, Z = 0.1
...
; layer 51, Z = 5.1
... so, 50 layers and 5.0 mm of Z travel as expected!!

So now I am even more confused as to why my part is too high ......

[jsc]

Bed gap will affect the physical height of your part in reality, but not in S3D.

Maybe your model is ever so slightly off?

I just generated a 20x20x5 cube in OpenSCAD ("cube([20, 20, 5]);") and sliced it with 0.1mm layers. The last layer (before "layer end") says:

Code: Select all

; layer 50, Z = 5

which is correct (not 5.1, that would give you a 5.1mm cube).

[Mind]

Calibration file that I am using is attached.... It should be 5mm high.

Also attached, the gcode file I mentioned.

[jsc]

Same thing. Last layer shows:

Code: Select all

; layer 50, Z = 5
G1 X91.344 Y134.233 F18000
G1 Z5 F1200

When you rotate the object into printing position, then double click the model, does it show z=0 under minimum coordinates?

Attach your factory file (File->Save Factory File).

Also, as an unrelated note, your model is just the exterior faces only and is non-manifold. In this case, it doesn't seem to matter, but what you want is a fully solid object printed with the same print settings. That will get you a thin wall box just the same.

[Mind]

When you rotate the object into printing position, then double click the model, does it show z=0 under minimum coordinates?
Attach your factory file (File->Save Factory File).

Also, as an unrelated note, your model is just the exterior faces only and is non-manifold. In this case, it doesn't seem to matter, but what you want is a fully solid object printed with the same print settings. That will get you a thin wall box just the same.

Yes, it shows z=0 and object height = 5mm. Factory file attached. Note that I added a Z-Axis offset of -0.10 as the bed is 0.15mm off the nozzle tip when cold (using feeler gauge).

The geometry is non-manifold. I just want a single layer around the outside, so I can get extrusion width and multiplier correct (btw, what does everyone else get?).

Incidentally, what should the skirt look like ? A single smeared layer, or all 4 outlines as separate threads ?

[jimc]

1st layer should not be smeared or seperate threads. Really what you should have are single extrusions where the edges touch making one solid piece

[Mind]

1st layer should not be smeared or seperate threads. Really what you should have are single extrusions where the edges touch making one solid piece

OK.. I am getting separate threads at the moment. To me, this means that the extrusion width is thinner than S3D thinks it should be. But I have set it correctly, based on a single layer thickness box... so how do I 'fix' this ?

[ednisley]

To me, this means that the extrusion width is thinner than S3D thinks it should be. But I have set it correctly, based on a single layer thickness box...

The distance between the nozzle and platform is about twice as large as you expect.

Your settings tell the extruder to emit enough plastic to fill a 0.55 x 0.10 mm "rectangular" thread. If the platform is 0.2 mm from the nozzle, that same amount of plastic will produce a 0.27 x 0.20 mm "square" thread. The adjacent threads will be 0.55 mm on center, leaving a 0.23 mm gap on each side.

That's what makes printing with 0.10 mm thick layers so challenging: an initial Z error of +0.10 mm produces separate threads (with zero adhesion) and an error of -0.10 mm produces a smear (and clogs the nozzle).

Start with 0.20 mm layers, measure the actual thickness, and adjust the Z offset to get exactly 0.20 mm. Adjust the alignment until it's the same across the entire platform, then go back to 0.10 mm and see how much better it works...