Reducing protruding "seam" at perimeter shell join

[problemchild]

I'm trying to build a circular PLA cylinder about 14mm dia and 13 cm long. (See http://www.thingiverse.com/thing:34778 file: shaft.stl .) Since it is a working part in a moving model, I've been trying to use 5 perimeter shells for strength. In all builds, the long axis of the shaft is vertical. At the point where the perimeter shells meet, or where layer starts if you prefer, there is an anchor point that creates a bump or blob on the outer surface. I've tried a few things, none of which seems to solve the problem. The current profile is attached.

With randomized layer starts, the whole surface is covered with the little blobs. With non-randomized starts, they line up and form a ridge, which looks easier to file off. Other attempts are: coasting with 1 and 2 mm gaps; reducing the extrusion factor, now down to 0.75; outside-in and inside-out perimeter building; overlap removal (30%); eliminate all infill; reducing the number of shells. The next thing to try is reducing the lift from 0.2 to 0.0.

A strange thing is that in many of these trials, the first few layers are almost perfect, but as the build grows, the "seam" grows and grows. The seam protrudes on the inside as well as outside. Also, as the build grows, a vertical bump forms above the surface of the layer at the join.

Any suggestions will be appreciated.

[Karl_Williams]

You could try making your retraction vertical lift equal to your layer height of .25 and lower the retraction distance to something like .50.

[wbailey]

I'm working on reducing or eliminating these blobs too through trial and error. I'll post if I have any luck. I'm using an M2 by the way, I may make a post about it on the makergear google group if I don't see any results.

[problemchild]

Welcome to everyone who wants to contribute. I just purchased a camera that I hope will produce pictures to post. But capturing the results may require more photo skill than I have. I tried Karl's suggestions, but they produced some very strange results which I'll try to photograph and post shortly. But it has spurred me to try some other approaches. One approach was to use Slic3r. That has yielded some excellent results. but has other shortcomings. I'll try to post images and profiles next week.

I'm thinking that this needs a methodical process. For example, start with very thin walled cylinders, then add layers and infill while keeping the seams flat. Randomly distributed starts don't really help, they just distribute the little bumps around the object. The height of the seams can be measured with digital calipers to quantify the effect.

Keep us posted, keep notes. Thanks to everyone who has the persistence to nail this little issue.

[wbailey]

My problem actually seems to be different, as I'm getting bumps in the same place on every print even with random starting point turned off. I think I'll start a new thread.

[problemchild]

Randomized starting point distributes the bumps in a different place on each layer. Turned off, the bumps are at the same place in each, but stacked vertically layer upon layer. Is that what you mean by 'in the same place', or do you mean in the same layer? I don't know if it is also random from build to build. In computers, a pseudo-random number generator can generate the same sequence of numbers from one run to another, but the numbers will be distributed randomly within the sequence. I depends how it is programmed.

[col55]

did any body find the answer to this would be helpful to have posted the fix I'm having the same problem and tried everything, any help would be great thanks

[andrewk72]

This is always going to be a tough challenge for 3D printers. Getting this perfect definitely takes some work and experimentation. And once you get it right, changing filaments could mess it up again! S3D already has most of the tools you need, so you really just have to start experimenting to find what combination of settings works best for your printer/filament. They have a good walkthrough on their quality guide support page
https://www.simplify3d.com/support/prin ... s-and-zits

[dcullen]

Watching this thread as I am having the same issues. I'm trying to printed a part that is a cone of decreasing diameter. I can get better surface finish with the Craftworks Slicer (change direction of rotation) but I need S3D to vary the number of layers to support changes in diameter. It is my one frustration with S3d. I'm left with the choice of a seam or randomly distributed blobs.

[mechy nick]

Hi All,
This is my first post here, and I'm fairly new to the 3D printing process but a producer of things as a hobby and a job for 25 years.

Just like to add my two pence worth to this discussion as I've been banging my head against this same problem since starting to use S3D.. (Cura doesn't do it and so it must be possible to eliminate right?!)

I think I've finally cracked it just now and that is - TURN OFF COASTING ! ... I just tried that and the pre-vis of the print is flawless. - No blips in the outer shells at all....

I'm about to try a print with this setting and will let you know if it works.

Regards,
Nick