Honeycomb infill weirdness

[CalifDan]

Jim,

Just for experimentation, I would try setting it higher to see if that makes a difference. You would not have to print the complete part, just up to where it starts doing infill and see what happens. It sounds to me like you extruded filament is not sticking to the lower level. At 20%, there is not a lot to stick to? I mentioned temperature also because with a 20% infill, there is not much heat retention. Certainly less than with solid infill.

[JimG]

I will also try a higher temp and report back, but to date I have run from 230C to 250C. It might be slightly better at 250C, but you would not notice unless you were watching pretty closely during the print process. The final product, after 2 more layers are placed, looks about the same at 250C as at 230C.

I am curious, though, why the shorter runs of thread in the corners, or where interrupted halfway across by a hole, wouldn't experience the same effects if it is a too-low temperature issue? I don't mean to sound like I am challenging you, just trying to understand!

Jim

[CalifDan]

Jim,

No worries. My thinking is that if the temperature prevents adhesion to the next lower level, then a longer run might exhibit more distortion than a shorter run would as it cools. You had indicated that the shorter runs seemed ok, but not the longer.

Also, if the issue is adhesion but not a temperature issue, then it might be a question of having enough material to adhere to. With a 20% fill, the lower level does not provide a lot of surface area for adherence. That is why I suggested you might try a higher fill rate to see if that helped. Again, if it is adhesion, I would think that longer runs would exhibit more distortion while cooling than shorter runs.

Having said all of that, I only have worked with PLA and do not have any experience with ABS. Hope it helps.

[printbus]

... I have been thinking about trying the rectilinear pattern to see if the behavior is any different.

Personally, I prefer rectilinear since it is easier/quicker for the printer to do, but I'll point out that with rectilinear, you don't have full coverage on every layer. Since layers alternate direction, the only place where you are ensured a full stack of filament is where the layers cross.

[AbuMaia]

One test I'd suggest is rotating the infill 45 degrees in Slic3r, so that the "long diagonals" are no longer diagonal. Probably won't make a difference, but it helps to eliminate as many variables as possible.

[jdolan]

I experience this issue and have not yet been able to resolve at 20% infill. I only print in PLA and the issue only occurs with larger prints. Maybe 3" wide and up and only wide open flat surfaces. I've tried honeycomb and rectilinear. Only makes a mess in the middle with long diagonal strands, but is fine when it shortens up in the corners. The strands are adhering just fine to the infill but they bow up like a sail in the middle of the actual honey comb. Very weird. Going to higher infill % solves the problem but uses more material and more time of course.

[JimG]

I have had a chance to try a few of the suggestions I have received. Not discounting the others, but just have not had time yet to try them. Here are the results so far:

1. Rectilinear infill pattern (45 degrees) at 20%, 250F --> significant improvement vs the honeycomb at 20%
2. Change to 40% infill, increase temperature to 260F --> a little more improvement vs #1 above.
3. 40% infill, 250F, increase the flow rate by 10% --> slightly worsened vs #2. Too much plastic extruded.

Next I will try reducing the flow rate by 10%, 40% fill, and 250F. But based on what I have seen so far, I think the real keys are changing to the rectilinear fill pattern and increasing the infill from 20% to 40%. Since the result worsened when I increased the flow rate, I want to see if reducing the flow rate below my standard esteps makes an improvement.

I will also have a look at the suggested 90 degree infill pattern to shorten the thread lengths.

Jim

[CalifDan]

Jim,

It appears that you are got pretty good results from changing the pattern and some additional from changing the infill percentage. Sometimes, when I have a problem, I do a big change and if that solves the problem, bracket back to get to the tipping point. Why not try 80% infill at 250 or 260. If that solves it, you know it is infill related. Then bracket back to 60 or 50 and try to find the tipping point.

As I stated before, I only use PLA, but I do not generally go below 50% infill.

[JimG]

Next I will try reducing the flow rate by 10%, 40% fill, and 250F.

This sample is probably the best so far, but only by a slim amount over the #2 trial.

Next up is a change in infill to 70% and use a rectangular, rather than diagonal, infill pattern.

Jim