Still struggling with ABS

[simon 1]

Hi all, I'm slowly coming up with solutions to printing with ABS but here's one thing I really can't solve. I've mentioned it before in other posts but its driving me crazy and I'm starting to think that the UM2 won't do this. Logically it doesn't seem right that I can't find the right combination of setting, but I've played around with print head temperatures, infill speed, flowrate etc. can can't get it right. The step Size is 0.1mm.

 

This is just a simple 20 x 20 x 5mm box with 20% infill. It's obviously just a test print, but until I can get this right, there's no point in trying more complicated stuff, it's just a waste of plastic.

 

It all prints fine until I get to the top layer which is dense infill. When printing the first layer of dense infill, the strands break before bridging the gap between one rib of sparse infill and the next, so it has nothing to anchor it down, The short broken strand then sticks up. As the head passes across the surface on the next layer it's like passing across a ploughed field. It eventually smooth's out to what you can see in the photo.

 

I've tried temperatures from 230 -265 and the result is identical.

 

If anyone can see what I'm doing wrong, or can tell me what setting to use, please let me know.

 

I'm going to try 0.2mm steps as well. Annoying thing is that my Afinia can do this so easily (0.2mm steps). I believe the Ultimaker to be a more capable machine, but I'm really struggling.

 

[gr5 2,172]

I've never printed with ABS but I've seen this exact pattern on PLA and the fix was more fan. One person had no fan. Another had a fan hooked up backwards so it was sucking instead of blowing. It doesn't sound plausible but I'm telling you I've seen this several times.

But in case that doesn't work...

Don't mess with infill speed - it should be the same as print speed - put 0 in there.

Try printing very very slow - 20mm/sec for the top layers. So when it gets there you can just lower the "feedrate %" to the correct value to get 20mm/sec. Maybe even 10mm/sec. This will help get a consistent flow.

.2mm seems like it might help also to get a thicker strand that won't break.

Also consider looking at your nozzle with a magnifier glass to make sure it isn't damaged (no holes, nicks, grooves).

 

[nick-foley 5]

Just use thicker top/bottom values? It looks like those are just way too thin. Set it to 1mm at least.

 

[simon 1]

Thanks gr5. Fan is the only thing I haven't tried. For ABS I was thinking no fan was the best option.

There's no way in Cura I can just get it to automatically print the top layers slower is there, or do I have to use TweakAtZ and work out where it needs to slow down?

 

[simon 1]

Thanks Nick, I will try that also. I do think that the underlying problem is the broken strand issue though. Although I'll try a thicker layer, I feel this is treating the symptom rather than find a cure. I might have to settle for that though.

 

[gr5 2,172]

There's no way in Cura I can just get it to automatically print the top layers slower is there, or do I have to use TweakAtZ and work out where it needs to slow down?

 

Correct.

Again - I have no idea why fan makes such a difference. There is more here (post #10):

http://umforum.ultimaker.com/index.php?/topic/1872-some-calibration-photographs/?p=17300

 

[simon 1]

Gr5, just wanted to let you know that sorted it for me. I actually used the default UM2 ABS material setting. Thanks for your help!

 

[simon 1]

Hi gr5, I figured out last night why the fans make such a difference. Wasn't really thinking about it at the time, it just came to me. If you're drawing out a strand of plastic in open air, the strand will tend to extend from the portion of plastic which offers the least resistance. If the entire strand is more of less the same temperature then the section of plastic which has the smallest cross section will be the weakest. It will keep on reducing in diameter until it snaps, which is what it was doing.

However if you cool the plastic with a fan, the smaller cross section has a much greater surface area to volume ratio. It is governed by the square law. This means that the thinner part cools very much more quickly than the thicker part and starts to stiffen up more quickly than the thickest section. The result is the thinner cooler part of the strand is no longer the weakest, as it strengthens when it cools. Thus the thicker more voluminous section of the strand which is still hotter will start to thin out.

In reality, this is a continuous process rather than having distinct stages, but I believe this explains why the fans allow plastic to extend evenly and bridge gaps.

[gr5 2,172]

This is a great theory - not sure if it is correct. Here is another:

When making a solid layer over sparse infill, each pass is right next to the previous pass. The fans let the previous pass cool slightly more before the head comes back and lays down the "stripe/string" right next to it. Having the filament cooler makes it stronger and less likely to break.

Both theories make sense to me. Maybe it's a little of both?

Illuminarti has a 3rd theory I believe.

 

[gr5 2,172]

By the way - I think you will find these pictures informative as the topic "bridging" is similar to the current thread:

http://umforum.ultimaker.com/index.php?/topic/3406-experiments-on-bridging