Problem with Extrusions and 3D Printing

[Anonymous]

Hello,

I have been having an issue with Fusion 360 and 3D Printing my designs. I am certain that the issue I'm about to discuss is of my own making but I need some assistance in discovering what I'm doing or thinking that's causing my problem. If I have posted this in the wrong location, please forgive me and move it to wherever it is the most appropriate.

I don't know how to state my problem succinctly but it has to do with extrusions and the way they turn out on my 3D printer. I also don't know if it happens all the time with every extrusion or with just some extrusions, but when it does occur, the parts that get 3D printed are very weak along the boundaries of the extrusion. I have had this problem a number of times and I'm confused as to what is happening to cause the problem.

Example 1:

I will begin with a simple example to start off the post but this isn't the example I will be describing in more detail later on. I apologize that I don't have pictures of this one to show but hopefully the description will suffice, and perhaps more particularly when viewed in conjunction with the next example that I will provide several pictures.

I recently made a simple box shell in fusion 360. The top was open but that's mostly an irrelevant detail I think. On the back of the box, from the original sketch, I made four "standoffs" using extrusions from circles in order be able to have it mount to a plate that already has some bolts (bolt heads) in it that I wanted to clear. The "thickness" of the standoffs were 4 mm, so pretty thick. The thickness of the back plate was 2 mm, so also fairly thick. The box is for a model and is attached with screws to the aluminum plate. The plate was threaded for the screws. 

Everything came off of the 3D printer seemingly in perfect shape. The box was well-formed, the thickness of the floor was right at 2mm, the holes for the screws were right at 5 mm and the standoff around the outside was right at 4mm. All as designed within a small variance. When I went to mount the box to the plate all of the holes lined up perfectly between the box and the plate. There was no twisting or bending involved in any way. But as I was tightening down the screws, they ripped the standoffs right off of the box as if they were nothing.

In examining the box where the standoffs had been, there was hardly any damage. It was as if they were hanging on by a thread to begin with and not integral to the structure of the box, which they should have been by 2 mm's worth of thickness along the structure of the box itself, and 4 mm's in diameter around the screw hole. And yet they tore off like they were simply paper.

Example 2:

Fusion 360 Part for examination: Deflector Plate

For my second example, this is a part I have been working on for a while and have gone through several iterations of it with each getting successively weaker in the areas between the extrusions. Which is pretty much the same problem I had with the box model in the previous example. The most recent version I printed out was so fragile it broke apart very shortly after I pulled it off of the 3D printer, just by handling it.

As you can see in the pictures below, rather than being solid models, they just fall apart at the extrusions.

In my sketches, all I'm doing is creating the various circles and whatnot and then highlighting what I need to extrude and entering the thickness. When I look at the part in Fusion 360 I don't see any problems with it. It is only when I download it and slice it up (and I've used several different slicers) and then print it out that I have problems.

I have other examples of other models that I've made that have this problem. When I examine the part it always comes back to problems around the extrusions. I do not have this problem with other parts that I download and print. Only the things that I design myself.

Can anybody help me understand what I'm doing wrong??

Thanks,

John

Here are some pictures which illustrate the problem:

Example of Finished PartAnother Example of Finished PartHow 3D Printer is Creating PartAn Iteration Showing Extrusion ProblemAnother View of Extrusion ProblemA Different IterationDifferent Iteration With Extrusion ProblemDifferent Iteration Another View of Extrusion ProblemExtrusions Seen in Simplify3DBottom Layers in Simplify3D

[mavigogun]

---

@Anonymous wrote:

 But my goal is to tweak and tune the process, up and down the line (and most particularly at the hot end) to get things to work the best they can and hopefully keep my designs pristine.

---

While I reckon that should, generally, be the first impulse, accounting for dimensional changes in the fabrication process is a common aspect of design in many, many modes of production- from casting to ceramics, all sorts of machining and engineering.

Can you qualify the nature of deviation from design geometry?   By what metrology?

[Anonymous]

Yes, as I mentioned above, there are generally two types of deviations that I routinely encounter which are discrepancies in both the outside and inside features and which are manifested in different ways, as measured with digital calipers versus the original design document.

My thinking is that they are both affected similarly by whatever small x, y, z miscalibration / misalignments exist, and the inner features are further affected by extruder issues such as E-steps setting, heat (oozing) and the like.

The first issue is simply a matter of mechanical alignment, meaning accuracy (precision) and repeatability. I know that my printer is off a little bit. According to my calculations the variation is somewhere in the vicinity of 0.8 to 1.0 percent (-ish) of the original specified design-- at least with respect to the scale of my test objects. So for a real example, if I print a 20mm cube, I typically see 20mm per side +/- 0.4-ish mm (say, 19.96 to 20.4 on a side), which I think is pretty good but not perfect.

There is a slight different in "error" between X and Y, but not a lot and both are in that approximate range. For larger items it's no big deal. For smaller items, since I think the error is more or less fixed, it becomes a larger issue as the desired size gets smaller. Some additional adjustment might make it a little better. My longer term plan is to build another 3D printer using some better parts-- linear guide rails and gearing down the motors 2 or 3 to 1, to be specific-- and see if that helps with better precision.

As for the second issue, oozing is typically a symptom of too much heat so just turning down the heat is a simple way to address that issue at the hot end. E-steps is a matter of comparing the filament input versus the filament output for a known amount of filament, also something which is essentially dealt with at the hot end (versus the document). It can also be affected a little by minor variations in filament diameter, but I haven't generally found that to be too much of an issue versus just working to tune the basic E-steps calibration in the first place.

As far as I can tell, from my own experimentation and observations, extruder issues seem to manifest themselves most as semi-occlusions of inside features such as holes, rectangles, etc-- anywhere the hot end makes "revolutions" around some center-- "revolutions" not always being "round" in this sense, but rather going around and around some central point, which is a fairly common thing in a 3D printer. The inside "curves" (tracks) tend to "ooze" over somewhat while the outside curves tend to run a little lean. The goal being to modulate it as well as possible to achieve the best balance between the two.

It would be neat if you could design in "tolerances" and such in Fusion 360-- maybe you can, for all I know, and I'm just too much of a noob to know it. And of course you can always create a tolerance artificially-- I like to use a named parameter for this type of thing-- and apply it however you want to the various features. It makes the drawing more complicated, as in more stuff to keep up with and more places to forget to do it. But by making it a parameter you can play with it and adjust it to get a good setting without foobaring the overall design too much.

Here is a link to a photo album with some parts that I've been playing around with for calibration. Known dimensions and diameters put together in various ways to work out tolerances, "fudge factors" and discrepancies: