Need help with ABS bridging / MAKE bridging test

Help! I can’t bridge. Because I was having trouble bridging in other models, I thought I would try to print the Make Magazine bridging test (Thingiverse link). In the magazine, the TAZ 4 scored an impressive 5 out of 5 on this test.

My print looks like this:

… and from the side:

I can see the problem as it happens. When the extruder goes to bridge, the filament doesn’t stick to the source point. So it just squirts spaghetti into thin air. It does generally ‘stick the landing’, and after a few layers, it manages to print on top of the spaghetti pile.

Any ideas would be greatly appreciated!

Here are all the details…

I’m using slic3r 1.2.1 with fine_ABS_no-support_pt35nzl_pt14layer-6s.ini configuration file, with the single modification of “Bridges fan speed: 100%” (at the suggestion of Ray in Aleph support).

slic3r slices without complaints and produces this gcode file (dropbox link).

I’m using Lulzbot ABS black 3mm filament. I start the hotend/bed at 235/85c. The profile knocks it down to 220/80c after the first few layers. I have good first-layer adhesion.

Maybe Make used PLA in their tests?

Yes, they definitely did. Is it not possible to bridge with ABS?

We used Ultimachine orange PLA as a control variable
(the team agreed that it was a solid,
widely available choice, representative of
what would commonly run through desktop
machines), some exceptions had to be
made (noted in our print-quality summary)
for machines that refused to function or
jammed without proprietary filament.

To bridge ABS with high resolution results, you do need a cooling fan for that one specific process. for every other piece of ABS printing, a cooling fan usually causes more trouble than it is worth. But it does help bridging.

Yes. It’s in the g-code I linked and I can see the fan turn on for each bridging move… still fails as described. The bridge doesn’t stick to the ‘jumping-off’ point.

You can tweak your bridging settings, by increasing the bridging speed, or dropping the bridging temperature.

This may be related to the slic3r version. It sounds more like a slic3r bug, if it isn’t even doing something that would work (e.g. just trying to print in the air without connecting to the post).

That’s really interesting. I wish I could use Cura, but I have a separate problem (forum link) with that.

I’ll go through the slic3r G-code and see if I can spot anything. I’ll also try slicing with Cura on other computers to see if I can eliminate slic3r.

The farthest Ive gotten ABS to bridge was 1cm so far, with a clean first layer.

When the bridge starts and you say it doesn’t stick to the start point, is your fan on at that moment?

What temps are you running?

I finally got Cura working (details in the other thread I started). I made a print from Cura’s gcode (standard ABS fine profile). The fan was on because of the minimum layer-time cooling-logic and warped the print. Turning the fan off hurt the bridging. Overall, Cura’s bridging result was better than slic3r. Comparing the two gcode files in a path visualizer, slic3r was moving non-optimally and retracting more often, notably just before a bridging attempt. Cura’s code would make the bridge in one continuous feed with the perimeter of the layer. I’m pretty inexperienced and I will have to look at retraction/ooze-control before I know how important that aspect of the tool path is in this case.

It would be nice if Cura had a ‘fan on for bridging’ option.

Thanks to everyone who responded.

Can you post your cura INI file for us? I had switched to cura after getting it to finally install and can’t bridge anymore… Danke and props on getting bridges to work!