almost a mendel replicator


Hi There!

A friend of mine gave me a pane. This is a part of kiteboards.
Usually it has two symmetric ones in the middle and four wing shaped ones, two at each side.

My first attempts to copy the shape using AutoCAD were a mess.
After 2 hours creating splines I gave up and started a web research.
And what I found was a David 3D-Laserscanner.

Unsure, what kind of software I would need, I found one - of course not cheap - complete kit,
containing the camera, the laser and most important: the software!
I ordered it and it was delivered 5 days later.

It needed only one day to understand what is important to merge the scans to a
3D model and after that, I only had to make a plain bottom by open scad - finished!
The software fortunately exports the 3D model as *.stl file which is compartible to open scad.

Creating the 3D model was the first quest.
The major quest was to replicate it with a suitable quality.
Printing a shrinked version of the pane with the normal PLA settings ended like this:

You can see big bulges - it looks ugly!
Usually nobody prints that small parts - but I wanted to.
If little parts work it is no problem to print bigger parts.
But the other way around you soon reach limits.
The miniature pane was at the limit.
Reducing the flow rate for getting rid of that bulges was no good idea:

Playing with dimension (for extruder retraction) made no difference.
The bulges disappeared, but now the threads were incomplete.
Turning oozebane or splodge or combinations did not work satisfying.

The most important thing when printing "clean" - without bulges - is a continuous flow of filament.
So I had to think smaller. My intention was to print thinner than the common 0.4 mm layer thickness.
When printing thinner you need a reduced flow which means less bulges.
First I tried it with 0.35 mm. Not much thinner and not the ultimate goal, but for the
beginning it was worth trying.

While testing the printer settings with higher feed rate, the "z driven pulley" broke.
It was made of PLA without filling. The high speed was not good...

I cut off a half of the y axis and replaced the broken one.For testing the next printer setups I decided to print "z driven pulley"s as test objects.

This is a very ugly one:

And then - surprisingly - I found the right setup by reducing the layer thickness to 0.27 mm and playing around with speed, flow, the first layer, unpause and many many other:

After that I tried the same setup with the miniature pane:

The right picture shows the miniature pane beside a 8 mm nut - just to show the size. It is 44 mm long.

Then I tried again to play a little with dimension, oozebane and other funny features of skeinforge,
but it was the same effect as the last time:

The bulges are smaller, but the outer shape seems to be too harsh.
So I decided to life with little bulges with the advantage of a nearly perfect "clean" shape:

The next try was the big brother of the miniature pane. The maximum z feedrate was another adjustment I did to prevent from bulges when lifting z axis.
It was set a little to high, so it didn't lift up every time. But it was the last adjustment to do to print a more detailed pane - and here you can see the final result:

Here the original versus the replicated pane:

And this is, what I needed to tune my Reprap mendel:

Next aim: thinner than 0.2 mm...

greetings from the lake of constance