Getting running in ABS as soon as I get my M2

[Mind]

Hi All,

I expect to get my M2 during the 2nd week of February. I plan to go on vacation in the 3rd week of Feb and need to take some items I have already designed (I am a ME). It is a strobe arm to connect an underwater camera to a strobe. Fairly simple. So I just need to get them printed during that first week with no major screwups or downtime. I am hoping you guys can help me with that!

Requirements are:
- decent tolerances (around 0.1mm is what I expect to get)
- ABS because it is going underwater!

I plan to use Simplify3D (I still need to purchase it).

Plan:
- purchase Simplify3D several days before the printer arrives. I understand the software can take a few days to get your registration code to fully enable it.
- I already have two rolls of ABS in 1.75mm black from reliable suppliers.
- get S3D settings/etc from this forum.
- scale models up by 2% to account for ABS shrinkage
- level the bed

Questions:
- is this tolerance expectation ok ?
- will I really need a heated cabinet for ABS ?
- what else can I do to ensure success ?

Thanks in advance!

[jsc]

Do you have your model already created? Did you design it or was it a thingiverse model? Designing for 3D printing has some constraints that are not immediately obvious; if you could provide the model, we can tell you if there's anything you are likely to have trouble with.

ABS is a tricky material to get started with. Success depends on the size and shape of the part, and your ambient temperature, and a whole host of other factors. For large blocky parts, your success rate will be markedly lower. I would lay in some PLA as an easy-mode backup plan.

By tolerance, not sure what you're referring to. The M2 has very good positioning accuracy due to its solid build. The MakerGear web site says something about a minimum positioning accuracy of 0.01mm X/Y, but that may be a theoretical limit. If you mean mating tolerances, I've found it a good practice to design in a 0.1 to 0.2mm clearance for mating parts, or have a set of files handy. Here's a thread with some other suggestions on settings for accuracy: viewtopic.php?f=2&t=275&start=0

Check out the "M2 tips" thread in the tips forum for some other common getting started issues.

[benavery]

ABS suffers from shrinkage as its printed, so I have found I have to normally re-adjust my models to account for it. I have heard some people mention just applying a constant scaling factor, but that never worked well for me.
I would expect you'll get within 0.1mm tolerance in consistency across prints, but I'm not sure you'll get +/-0.1mm from what your 3d model specifies, if printing in ABS.

[Mind]

I did the design myself. Reason's for accuracy required (over multiple components, NOT all these are on the attachment!):

1) hole through which a bolted fastener is slotted (this can be drilled out to make it bigger if it is too small)
2) one part will mate with a ABS pipe and capture both the outer and inner wall of the pipe and be glued to the pipe. This needs to form a pressure-tight seal as this is a buoyancy arm.
3) holes are for thread inserts, so I am not worried.
4) hex holes for captured bolt head and nut

Worries:
- (2) might be too tight to insert the pipe
- (4) might be too loose to hold the hex head securely

I could just print the part, then change the dimensions up/down until it fits snug, but I would prefer not to do that

Other items that spring to mind.... I am not sure how thin I can make the wall around the end of the tube. I have gone for 4mm, but it is complete overkill.

Attached is a zip file containing the end cap that needs to go over the ABS tube. The other end has a M6 thread insert.

[jsc]

Your part seems like it would print fine. Two potential problem areas, especially if you're trying to make it airtight, is the curve at the top of the innermost wall, and possibly the outer wall curve at the bed, although that one's probably fine. As a general rule, you should try to limit unsupported overhangs to 45 degrees. For the inside wall curve, my gcode preview shows that it is steep enough so that the very last perimeter is almost completely unsupported and would drop in. You would have to print with supports, which is not the end of the world but they can be a pain to dig out.

Unsupported 90 degree surfaces (parallel to the build plate) are okay as they will be bridged, as long as they have something to bridge to.

It is best to have thin walls be an exact integer multiple of your extrusion width (I select 0.40mm manually) so that you don't get gaps.

As far as ABS-ability, this object is pretty good. It's small, round, and not completely solid. Don't know about making it airtight, though, you'll have to try it and see. Giving it a brushing of acetone would help in that regard to close up any pinholes.

I would get rid of the fillet (convex curve) along the bottom rim, though, or convert it to a 45 degree chamfer.

As far as fitting it to pipes, you will be lucky if it fits the first time. There are too many variables: ABS shrinkage, the fact that 3D printed circles are faceted, you probably won't have your extrusion completely dialed in.... I'd plan on one or two test prints. I find it helpful to use simplified test models to make fitting easier, just some thin walled cylinders connected at the top or bottom for example. But I've also found that when doing test fits to real world objects, you can get a press fit on a thin wall test print, and completely fail with a larger more solid print due to the flex.

[Mind]

Hmm... ok, so the porosity and micro-channels for leaks issue is something I had not considered. How much of an issue is this with 100% fill ?

The end caps are intended to fit on the end of a ABS tube to form both an arm and a floatation device for the camera assembly to try alleviate some of the negative buoyancy from the joints (which have stainless bolts in them) and the camera housing and base. If this floatation arm just fills with water, there is going to be a problem

[Mind]

So you have an idea what the end result will hopefully look like:

[jsc]

Nice. Are you planning on printing those tubes? It would be far easier to use sections of PVC pipe if you can.

You shouldn't need 100% infill, I've printed watertight models with low infill and two perimeters. But that's just sitting in a container at sea level. If you're planning on diving with this, I'd run some tests.

[Mind]

Yes, the tubes can be had at Home Depot $4 will be enough for two complete arms with a 5 and 8 inch float arm each.

Design requirements for the assembly are static external pressure of 4-5bar max (4bar = 40m water depth, which is the recreational diving limit). The pipe has a burst pressure of 24bar. I expect crush pressure to be slightly higher.

I could do some non-linear FEA on the assembly, but ... it's a pain to do correctly (extract the lowest eigenmode, use it to perturb the structure, then do a RIKS analysis on that). The design is complete overkill, so why bother?

CAVEAT: assuming the printed part is actually solid/watertight.....

[insta]

Acetone vapor baths will be your friend here, as will low layer heights and overextrusion.