Posts: 7
Joined: Fri Jun 17, 2016 5:38 pm

Outer perimeter seam optimisation

Sat Dec 29, 2018 7:51 pm

Hi All,

I'm running into an issue which seems impossible to fix. The fact that there is an option for fixed, optimised or random start points is very convenient, but there is something that doesn't make any sense. When I select 'optimise', the end of the inner perimeter touches the start of the outer perimeter (= very short travel move). But when I select 'random' (to make sure there is no line shaped seam), the end of the inner perimeter is on a different place than the start of the outer perimeter. It looks like the start of the inner as well as the outer perimeters are also random, but relative to each other, which is something I can't imagine any useful application for..



This end to start travel move creates a tiny gap at the start of the outer perimeter. Which I think is caused by the filament pressure that drops when the printhead travels from the end of the inner perimeter to the start of the outer perimeter.
These are the gaps I mean.

I have tried A LOT of settings, values and combinations to try and solve this: wipes, retract wipes, coasts, inside out and outside in, (un)checking "only retract when crossing open spaces", extra restart primes, slower outer perimeter, linear advance. Nothing works. Funny though: an old version of cura (15.04.06) creates an almost invisible seam with no adjustable settings. So this is what I still use for certain picky customers.

-> The solution is to have the end of the inner perimeter touching the start of the outer perimeter, also when 'random' is selected.
-> Also another possible solution is an adjustable X millimeter overlap between the start and end of the outer perimeter, closing the visible hole. You could see this as a wipe-like action, but instead this overlap function needs extrusion, where a wipe doesn't extrude.

User avatar
Posts: 1876
Joined: Tue Mar 10, 2015 3:09 pm

Re: Outer perimeter seam optimisation

Sun Dec 30, 2018 6:34 am

Do I detect a misunderstanding as to the terminology used in the three options for seams??

It difficult to hide the joint between the start and the end of a layer....we all know that! And the three options are provided to allow some flexibility in how the starts/ends of each layer are placed in terms of how & where the filament is extruded...and not the expected visual result of the print! I'll explain each....

Fixed...the starts/ends are aligned to be as close as possible to a defined XY point. This will deliberately cause a visible seam...but because you have control over where the seam is, you can position it where it will not be seen.

Random...the start/ends locations are randomised so there is no "seam" generated. Visually, this is the best compromise for where the full perimeter is viewable. There will, however still be possible flaws visible at layer level, but no "seam".

Optimised....the slicer is left to decide where each start/stop is placed so to reduce the wasteful un-productive nozzle travel time. In other words, do not try to hide anything...just get the print done! You can guarantee a "seam" with this setting and have no control over where it is!

As you have suggested, there are a raft of settings that affect the start and end of a perimeter for the single layer...mainly the Retraction, Coast & Wipe settings...and you need to understand the mechanics of each setting before you can get them working in your favour.
Doug Kightley
Volunteer at the National Tramway Museum
Railway modeller and webmaster at

Posts: 7
Joined: Fri Jun 17, 2016 5:38 pm

Re: Outer perimeter seam optimisation

Sun Dec 30, 2018 9:58 am

I know, random is the option I select to prevent a seam, but instead get speckles all over the print, which is totally understandable. But there is one flaw.

The problem with 'random' compared to 'optimised' and 'fixed', is that random creates a different start point for the inner perimeter(s) as well as the outer perimeter, independent from each other. This often creates a (longer than needed) travel move after the end of the inner perimeter when it heads to the start of the outer perimeter. This travel move is totally unnecessary.

The start of the inner perimeter should be randomised, BUT after the inner perimeter finishes, it should travel to the closest possible location to start the outer perimeter.

I have tested this by creating an individual process for the first 30 layers, and manually set the starting point for these layers. This makes the starting points act like "fixed", but on a different location for each of the layers on my test cilinder. This actually works a lot better and creates a much more clean "speckle" pattern, without the gaps I've shown. Of course this method is way too inconvenient to implement in every situation.

Posts: 6
Joined: Wed Feb 10, 2016 11:39 am

Re: Outer perimeter seam optimisation

Fri Jan 18, 2019 11:46 am

I totally agree with you Doug. This behavior has resurfaced with the latest release. I thought they had fixed that.

I too think this extra movement between each perimeter start (inner ones to outer ones) is ridiculous. It used to have the same start point for each perimeter in the layer, then they broke it, then they fixed it, and now it's broken again.

Posts: 37
Joined: Sun Apr 20, 2014 1:04 pm

Re: Outer perimeter seam optimisation

Mon Feb 11, 2019 1:26 am

Glad it's not just me.
I've spent two days playing with settings looking for a sweet spot...ended up with a seam every time.

Why oh why?

Posts: 420
Joined: Wed Jun 20, 2018 11:45 am

Re: Outer perimeter seam optimisation

Fri Feb 15, 2019 8:59 am

Hi guys! Thanks for reporting this issue. We're aware of the change in functionality and it was not intended. I'd look for a fix in our next update.
"A bird does not sing because it has an answer. It sings because it has a song."

Return to “Troubleshooting and Bug Reports”