Showing posts with label Z-Axis Wobble. Show all posts
Showing posts with label Z-Axis Wobble. Show all posts

Thursday, May 31, 2012

The Stereo Microscope is My Friend

If there is a tool that I count on more than my stereo microscopes, then I sure can't think of one.  I find a low power microscope will tell me things that I just cannot validate any other way.

And, the performance and precision of a 3D printer is no exception.  So, for the first time, I was able to examine the prints from the cube under 10x to 40x and analyze the characteristics of printed surfaces.

My favorite test shape is an extruded triangle.  It has straight lines, flat surfaces and sharp edges.  It's perfect for analysis.  Most of you know that I already had a RepRap printer before purchasing the Cube.  And, I put its output under the same microscope a while back.  That's how I found out that that one of the rods that controls the up and down motion of the print bed on my RapMan was slightly bent.

Here's what it looked like at various powers from 10x to probably 40x.  This is an edge of a triangle.  Notice the regular pattern of hills and valleys.

Here is probably about 10x...

RepRap - Triangle Edge at about 10x

This is probably at about 20x...

RepRap - Triangle Edge at about 20x

This is most likely around 40x...

RepRap - Triangle Edge at about 40x

Now, in reality, to the naked eye, it is not as ragged as it appears under the microscope.  Remember, the layers are .25mm and that is VERY small.  You can feel it rather than see it.  And, I know that some RepRaps might not show this particular issue.  But, it is so common that it has a name.  It's called Z-Axis wobble.

So, now let's put the Cube print under the scope.

First, the same kind of triangle edge with the grain going the same way...

Cube - Vertical Triangle, Edge at 10x

Look at that smoothness!  That's remarkably flat.  So, what about other grain orientations in an extruded triangle?  The is the edge of a triangle with the grain in a slightly different orientation.  The first triangle was vertical and this one was lying on its side.

Cube - Horizontal Triangle, Edge at 10x

But, the really cool view is at the apex of the edge looking down on the extrusion.

Cubify - Vertical Triangle, Apex of Edge - 10x

What about analyzing other features?  Like a hole without support...

Cube - Hole in Wall, No Support - 10x

The bottom of the hole is on the left in the above picture.  So, now let's look at the same hole.  But, this time printed with support.

Cube - Hole in Wall with Support - 10x

This is very thin support.  But, I haven't tried to remove it.  So, I don't know how clean the hole ends up after the support is removed.

But, all of you know by now that I love the behaviour of Pentagon shaped holes.  They NEVER seem to need support!  And, look how clean they are at every apex!

Cube - Pentagonal Hole, No Support - 10x

Cube - Pentagonal Hole, No Support - 10x

And, let's begin to wrap it up with some up close and personal shots of the smoothness of the edge of a sphere.

Cube - Circumference of a Sphere - 10x

Look how smoothly that sphere's arc is.  That's beautiful.

And we will end our wrap up with a look at the edge of a thin-walled vertical column.

Cube - Wall of extruded Column - 10x

I don't know about you.  But, I think these images clearly demonstrate the capabilities of Cube to deliver precision prints  This is EXTREMELY important if the things you design require tight tolerances.  I plan several microscope to camera interfaces and it is amazing how precisely the center of the camera's lens must be aligned with the center of the microscope's lens.  There is no margin for inaccuracy.  I'm convinced that the Cube will finally allow me to achieve this goal consistently due to the straightness of the walls it prints.

I hope this is helpful and not confusing.  I really do this for myself because it allows me to work WITH what I have.  And, as it turns out, I have a lot with the Cube.  :)

But, there is a reason why I share it with you.  I want YOU to know that my enthusiasm for this little printer is based on cold hard facts and not just emotions.  People can be enthusiastic on no basis at all.  Admittedly, the initial enthusiasm that gave rise to the name of this blog was a gut level response based on years of dreaming what my perfect consumer 3D printer should be.  But, then it was deepened by my driving from the DC area to Rock Hill, SC to see one for myself.  I didn't have my microscope.  But, I could see that the objected printed out right in front of me was remarkably smooth and precise.  But, now I have the objective tools to validate or negate my gut feelings and, as you can see, my instincts seem right on target.  

Let's compare the green Cube output with the white RepRap output one more time...

RepRap (White) vs. Cube (Green)

The scope doesn't lie.  A picture really is worth a thousand words.

Wednesday, March 28, 2012

The Cube's Astonishing Accuracy

As a former science teacher, I have an abiding interest in microscopy.  So, I have many of them around my workspace.  Among my favorites are low power stereo microscopes because few things are as useful when trying to analyze the performance of mechanical things.

I pulled out a 10x-40x zoom stereo microscope to try to analyze a phenomenon called Z-Axis wobble common the RepRap style 3D printers.  The wobble is not bad in my particular RapMan.  But, it is something that I want to try to eliminate entirely if I can.

Here is what Z-Axis wobble looks like in a printed piece.

Z-Axis Wobble Evidence

Notice that the edge is scalloped.  This is because the bed that held the piece wobbled slightly as the screws that raise and lower the bed turned.  One or more of the threaded rods has a slight bend in it.

Now, compare the edge on the above image with this edge under the same microscope at 40x.  Notice that it is perfectly straight with no scalloped pattern.  This piece was printed on a Cube.  At 40x the depth of field is poor.  So, just focus on the edge.

No Evidence of Z-Axis Wobble

It is evidence that the Cube's platform is very, very stable and that the alignment is virtually perfect.  My impression, when I compare the two Cube pieces that I have with the pieces I print with the RapMan is that the Cube is noticeably finer right out of the box.  The RapMan can print this fine; but, only after I solve the threaded rod wobble issues.

This next image, with the microscope set to 10x, demonstrates even more starkly just how accurately and uniformly the Cube prints.

Very Smooth edges from the Cube

To me, the above image is remarkable.  Just look at the walls and the uniform slope that was delivered by the Cube's print control system.  The layers are even and the edges straight.  I doubt that any extrusion printer gets much better than this.  (The indentations are the brick design patterns.)

You can be sure that once I get my hands on one, I will be this particular about analyzing the output that I am able to achieve.  And, I have the perfect STL for testing.  In the meantime, I will be working with the RapMan printer in an attempt to bring it to this level of print accuracy and smoothness.  I am sure it can be achieved.  But, it will mean some serious thinking and work.

The Cube, however, should come right out of the box printing just as we see here.  And, I am REALLY looking forward to having that experience.

I'm truly looking forward to the arrival of the Cube.