As my first project, I'm trying to design a holder for glass vials, for a scientific application. The photo below shows the latest design iteration, and also shows the problem with it:

vial holder

As you can see, one of the clips that's meant to hold the vial in place has broken off. This happened after inserting the vial once or twice.

I think I understand the reason for this. It's because I'm printing in the orientation shown below, in order to avoid the need for supports. (This will be more important later, when I scale it up to an array of many holders.) This means that it's relying on the strength in the z dimension, which is much weaker than in the other two directions, because it relies on the cohesion between the layers.

enter image description here

My question is whether I can do anything about this by changing the design of the clips, without abandoning my hope of support-free printing. I don't care at all what the design looks like, except that as much of the vial as possible needs to be visible from the front. I'm printing in ABS.

I have tried varying the thickness of the clips. Thinner ones don't break as easily, but they are a bit floppy and aren't very good at keeping the vial vertical. The ones shown are the thickest I've tried - I'm afraid that if they're too thick they won't bend at all. (The vial is supposed to be inserted from the front rather than the top.)

Edit just as an update, here's what it looks like using John Biddle's suggestion, which works perfectly:

enter image description here

  • $\begingroup$ Try printing the original design vertically. This is about print layers which tend to segregate. $\endgroup$ – fukanchik Mar 22 '16 at 19:03
  • $\begingroup$ Nice T-slots :) $\endgroup$ – Trojan Mar 26 '17 at 0:19

It looks like those clips are thin and need to bend pretty far to let the vial out. Try to make the clips thicker, but with a smaller clip to retain the vial so that it doesn't have to bend as much.

This is what I'm thinking, in beautiful MS-PAINT form:
enter image description here

  • 6
    $\begingroup$ +1 I tried suggesting this in my answer, but my MS-PAINT skills are lacking. $\endgroup$ – tbm0115 Mar 16 '16 at 20:46
  • $\begingroup$ Man, I was going to suggest this, finished my MS Paint example, only to scroll down and see yours :( $\endgroup$ – 8bitwide Mar 16 '16 at 23:37
  • 1
    $\begingroup$ Thanks, this is a really good idea. If I don't change the printing orientation then having the clips sides be vertical also means each pair of layers has more surface area in contact with each other, so it should be stronger. $\endgroup$ – Nathaniel Mar 17 '16 at 1:06
  • 2
    $\begingroup$ This works perfectly, and makes a nice satisfying "click" when the vial is inserted. Thanks to all the other answers too, they were all very useful. $\endgroup$ – Nathaniel Mar 21 '16 at 13:56

I agree with the previous assessments -- printing vertically for that part of the clip will definitely alleviate the layer adhesion being your weak point.

You might also consider splitting that clip as a separate modular piece (which will clip/socket/bolt into the stand). Depending on implementation, this would give you the ability to

  • Print that part vertically without supports
  • Swap the part if it breaks, without having to print the whole assembly again
  • Preattach the clip to vials, which may be good or bad depending on your usage
  • $\begingroup$ Printing the clip as a separate module also saves a ton of printing time and material required. $\endgroup$ – Trojan Mar 26 '17 at 0:13

Considering that you wish to print the clip in the vertical orientation but without supports, I would suggest that you create a suitable taper of the clip from the bottom. The taper would provide necessary support-free support but may reduce the amount of bottle access one has to remove it from the clips.

As such, you could also reduce the amount of the straight vertical segment of the clip at the front, which would allow the support-free support to start at a higher level from the bottom.

If the bottle bottom encirclement is raised, you'd receive additional retention. You could then reduce more the upper encirclement and taper for easier release.

I too prefer to design models that require no support, but have run into the same circumstance, where the weaker axis takes a disproportionate load.


Come on. Wider clips will just survive a bit longer. The real cause is the orientation (plane) of printing.

Continues filament layer will always be more durable than few layers sticked together.

So the question is why not to change printing plane? You can use better design which utilizes support without using "support material" enter image description here


The Short

  • Just print vertically
  • Reduce the clip size to bring them closer to the centerline
  • Try a vapor treatment or epoxy to increase strength

The Long

Ultimately, printing the object vertically (so the clips are printed horizontally) would be the strongest solution.

Alternatively, you could adjust your model to bring down the size of the clips to reduce the strain. Consider making more of a cup or cone on the bottom for the vial to slide into center. Then reduce the length of the clips to bring the closer to where the center of the vial will be. Make sure, however, that your clips still curve slightly around the center (and maybe a mm or two beyond). From a design point, the bottom clip is probably unnecessary.

Another option might also be to use a small vapor treatment on the ABS. Depending on how deep the alcohol penetrates, it could diminish some of the issues with the part cracking along the grain. This is just a fairly labor intensive process compared to some of the others posted here.

Side note, those are great looking prints! I like how smooth ABS can look. Its hard to tell, but it looks like your part has a bit of warping on the bottom. Definitely look into some of the warping questions here on Stack.

  • 1
    $\begingroup$ Thanks, this is all useful. It doesn't look warped in real life! I think that's an issue with the camera's optics. $\endgroup$ – Nathaniel Mar 17 '16 at 1:50
  • 2
    $\begingroup$ No problem, glad to see my other suggestions for clip design helped too. $\endgroup$ – tbm0115 Mar 17 '16 at 17:22

Printing the model vertically should certainly be considered, as that will alleviate your issue by a whole lot. If altering the design by adding a taper to your model isn't what you are looking for, placing a manual, removable support structure at the outer edge og the overhang could also be an option. This way, you could print the overhang as a bridge, making it much easier to print vertically.

Also, have you considered using another material than ABS? There are a large variety of filament types available that have better semi-flexible properties than ABS, without sacrificing overall strength. For instance, PETG and Taulman's Alloy 910 filaments are both very strong, and surprisingly flexible.

  • 2
    $\begingroup$ Thanks. Probably I will print it vertically in the end. The reason not to is that it will be an array of these stacked on top of each other, so when it's finished it will be quite tall and the bottom part will become an overhang. But maybe I'll just print each layer separately and stack them. I'm limited to materials available for the Zortrax M200 and the eventual design needs to be white, so I think ABS is the only option. $\endgroup$ – Nathaniel Mar 16 '16 at 11:55

If you don't want your clip to break, you'll have to engineer in some form of flexibility in a strong part of your design. For example, right now it looks like the bases of the clips are sitting rather solidly on the underlying surface such that it can't move. I'd suggest that instead it might be better to have two clips attach separately to the bottom using a "post", and have each post attached to the solid part of the base with a piece of plastic whose horizontal cross-section is a somewhat-rounded accordion-style double zig-zag.

From the top: Top view

Having multiple "pleats" will mean that each degree that the clips flex outward will only require each pleat to flex a fraction of a degree. You'd have to play around to find out what number of pleats would give a suitable degree of stiffness and flexibility, but if the printer renders the design by drawing the zig zags, stresses should be concentrated in the directions where the material is strongest.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.