1 Topic by nfhill

  • nfhill
  • Member
  • Offline
  • Registered: 2013-03-10
  • Posts: 26

Topic: Extruded Filiments Not Bonding Horizontally

I've always had a problem with extruded filliments not bonding well horizontally.  For example, I use a lot of 1.5mm thick vertical walls in my designs.  The printer builds these by laying down a perimeter filliment for the outside surfaces and then a fill pass inbetween.  The result is usually three separate vertical walls one filliment wide with no horizontal adheason.  The strength is very bad.  The fill in wider wall thicknesses is also often bad:the fill often doesn't touch the perimenter shells.  Thus my prints generally have a lot of voids.

Today I created a special test design with varying wall thicknesses to see if I could find some parameter that could be changed to correct the problem.  The object consists of walls of 0.5mm, 1mm, 1.5mm, 2mm, 2.5mm, and 3mm thick with a 0.6mm thick bottom layer.

With default settings, the 0.5mm wall prints as on filliment pass with thickness 0.36mm to 0.4mm; the 1mm wall prints as two wall shells with a fill pass inbetween that bonds the two shells together well (the way it shoud be); the 1.5mm wall prints as two wall shells with a fill pass inbetween, but the fill doesn't bond to the shells leaving a very weak wall; the 2mm wall prints as 2 perimenter shells each side with a fill pass inbetween and again the fill doesn't bond to the perimeter shells and the shells don't bond to each other; this continues the same for the thicker walls: the shells don't bond to each other and the fill doesn't bond to the ajacent shells.

I decided to try increasing the Extrusion Mulitplier.  The stock value is 0.79.  I started by increasing this to 0.9.  There was a noticable improvement in the horizontal bonding, but still not as good as it should be.  I tried values of 0.95, 1.0, 1.1, 1.2, and 1.5.  With each increase the horizontal bonding improved.  The values of 1.2 and 1.5 resulted in very rough surface in the fill box bottom.  A value of 1.1 seems to produce good horizontal bonding even though it's not 100% bonded.  However, the layers are squishing noticably causing vetical openings to be significantly reduced.  For all multiplier values, the single filliment wall is ~0.4mm wide.

So my question: are there other parameters that can/should be tuned to fix the horizontal bonding problem other than the extrusion multiplier?  Can I get solid vertical walls without reducing the size of vertical openings?

Post's attachments

Wall Test Print Box X.stl 19.03 kb, 12 downloads since 2013-05-15 

You don't have the permssions to download the attachments of this post.

2 Reply by Rincewind

  • Rincewind
  • Member
  • Offline
  • Registered: 2012-10-16
  • Posts: 334

Re: Extruded Filiments Not Bonding Horizontally

Can you post a picture? If your infill is not reaching your walls, it might be a backlash problem. Try adjusting the belt tension and/or a firmware compensation.

3 Reply by nfhill (edited by nfhill 2013-05-18 05:09:19)

  • nfhill
  • Member
  • Offline
  • Registered: 2013-03-10
  • Posts: 26

Re: Extruded Filiments Not Bonding Horizontally

OK, a couple days of tuning.  Here is a picture of the test print with the poor horizontal bonding.  You really can't see the problem.  But, if you run a fingernail across the top of the walls, the layers, except the 1mm wall, easily separate like pages in a book. 

It has always seemed to me that the tool paths that Slic3r has been creating are too widely spaced for the width of the extruded filliment.  After 23 prints of this test box with varying parameter values, I've concluded that the primary problem has been that the factory default for Slic3r was set to 0.42mm for the extrusion width (print settings/advanced) and my extrusion width has been running from 0.34mm to 0.37mm.  By changing this parameter to 0.36mm and the extrusion multiplier to 0.9 my horizontal adheasion of printed walls is much improved.

I also adjusted the tension in the two y axis belts.  They were definitely not equal.

Unless someone has more suggestions, I'll run with these new settings for awhile and see how the prints turn out.

Nick

Post's attachments

P1010029.JPG
P1010029.JPG 101.37 kb, 1 downloads since 2013-05-18 

You don't have the permssions to download the attachments of this post.

4 Reply by Shameless

  • Shameless
  • Member
  • Offline
  • Registered: 2013-01-11
  • Posts: 24

Re: Extruded Filiments Not Bonding Horizontally

Have you already tried this? http://solidoodletips.wordpress.com/201 … flow-rate/ It would help give you a specific extrusion multiplier setting, rather than trial and error. I don't know how low you can get the wall thickness either but 0.34mm sounds way too small, given the nozzle is 0.35mm.

5 Reply by Rincewind

  • Rincewind
  • Member
  • Offline
  • Registered: 2012-10-16
  • Posts: 334

Re: Extruded Filiments Not Bonding Horizontally

Did you measure your filament diameter accurately? it can go as low as 1.65 and it can make a difference

6 Reply by nfhill

  • nfhill
  • Member
  • Offline
  • Registered: 2013-03-10
  • Posts: 26

Re: Extruded Filiments Not Bonding Horizontally

I thought I already posted a reply yesterday, but I guess not.

It's not a problem with the filament diameter.  I measure it often.  I thinks it's just that I never calibrated the flow rate.  I didn't know that it was a 'must do'.  I contacted Solidoodle support shortly after receiving my S2 about this problem, but 'John' didn't mention the need to calibrate the flow rate.  In fact, he didn't provide any help at all.

I don't know who's idea it was to put a muliplication factor of 0.79 in the default configuration files, but he deserves some major head slaps!

My tests were really a trial and error method of calibrating the flow rate.  After doing all that, I stumbled across the instructions for calibrating the flow rate.  I've decided to settle on a multiplier of 1.  This gives a extruded filament of about 0.42mm and with this set in Slic3r configuration it seems to work well.

I still have some backlash problems in Y, but I'll need to print and install better brackets on the Y belt idler pulleys to get that out.  The stock ones flex quite a bit so there is no way to get equal movement in Y across the X axis.  Fortunately, another Solidoodler has designed and posted stl's for better ones.

Thanks for all of the responses.  Much better support here than from Solidoodle!

Nick