Feature request; Inner wall inset or flow

[Adamada 0]

Hello,

 

I'm currently battling a problem where my inner walls do not connect with each other or the outer wall - especially around smaller circular holes. 
I've searched around a lot online and talked to people in different groups, this seems to be a common problem with no real solution, other than increasing flow % (I have experimented a lot with different settings to little success) - which can have adverse effects in other areas of the print. 

"Outer wall inset" fixes the problem fantastically, but affects the dimensions of the printed piece; for a modeled 1/8" hole, the "outer wall inset" will squish the outer walls closer into the print and end up with a hole larger than 1/8".
(Reading the description of "outer wall inset" I'm using it under different parameters than designed for; I have a .6mm nozzle with a .72mm line width rather than a smaller than nozzle diameter line width. Regardless, the outcome is favorable, just not the part deformation.)

I would like a way to squish the inner walls into the outer wall, rather than vice versa. Perhaps a "inner wall inset" if you will. This way the dimensional accuracy will be kept while achieving the same effect.
This could potentially also be solved by allowing a separate control of inner wall flow; if I could bump my inner walls to 110% or something and overextrude for them to better connect with each other and the outer wall, that'd be fantastic. 

I think this feature would be very helpful. 
(If there is a better place to post this feature request, I'd appreciate somebody pointing me in the right direction)
 

[gr5 2,167]

When walls aren't attaching it's always caused by underextrusion.  This is one of the places that people notice it first (when it's only like 10 to 20% underextruding).

 

Slow down your print.  When you print a .72 line width through a .6 nozzle it takes MUCH more pressure to squeeze that filament into the small gaps.  Assuming 20% underextrusion, try slowing it down by 40% (print at 60% speed).  You can control the speed on most printers out there in the TUNE menu if you have a variant of Marlin firmware on your printer.  That way you can tweak the speed live while it prints and see if it looks better or worse with feedback in a few seconds rather than waiting for an entire print to finish.  Don't experiment on the bottom 2 layers which are affected more by leveling position and tend to run at lower speeds anyway.

 

If it all possible, print at .6 line width.  That way you might not have to slow down the print.

 

There are lots of causes for underextrusion - hundreds - here's a random one: A single CF filament print long ago wears down the feeder spline and now the feeder slips too much as the splines aren't sharp.  The causes are different for each printer so contact an expert who knows your printer very well.  Other causes could be bad teflon, overheating stepper driver, print too cold (bad sensor), printing too fast, printing layer heights too thick, and I swear there's dozen's more.  My list for the UM2 and UM3 is about 20 things to check.

[burtoogle 513]

A new feature of the 4.2 (beta) release is that you can set the flows for features individually. So using that you could crank up the inner wall flow...

 

[Adamada 0]

gr5; I've got my e-steps dialed in, and get no under extrusion elsewhere in my prints other than these inner walls. I'm printing inner walls at 30 mm/s, I'll drop it lower and see if that helps. 
For my line width, I read a few places that 1.2x nozzle diameter is recommended so I gave it a try. I was printing at .6mm line width with the same issue occuring. 

burtoogle; Fantastic! Thanks a bunch for letting me know. I'll download and play with it. 

I still think an "inner wall inset" would be a nice option to have as well. There may be scenarios where that may be preferred over inner wall flow - though none come to mind right now. 😁

Thank you both for your quick replies, very much appreciated!

[gr5 2,167]

On 7/20/2019 at 4:10 PM, Adamada said:

I read a few places that 1.2x nozzle diameter is recommended

Instead of increasing flow and making the feeder push even harder (potentially turning the filament into powder) try 1X or even 0.8X nozzle diameter - just to see what happens.