Taz1 fine tuning

I have a Taz1 here at work that I’ve been trying to get dialed-in. This machine has been unknowingly abused and it’s taken some time to get it back to a usable state. (We also have a Taz5 and it’s printing beautifully)

I’m fighting what I believe is a retraction issue. Here’s a screengrab of how the nozzle traverses the part:

Here’s an overview of the part as it was completed:

And a couple of close-ups of the defects:

What I believe is causing these artifacts is a little bit of melted filament being wiped onto the
inside wall when the nozzle passes over. I’ve tried retraction settings ranging from .5 to 2.0. The above
was printed with retraction set at 1.0 If I add more retraction I get voids in the print immediately following the retraction, any less retraction and the artifacts are larger.

I’m printing Lulzbot ABS at 235 nozzle and 87 bed, .1mm layer height and “Fine” defaults from Simplify3D

All comments are welcome as I could be completely misdiagnosing the issue altogether. :blush: (Looking closely at
the photos it appears it’s under-extruding a tad, I’ll check the extrusion calibration)

Sorry for the un-centered photos, click on them for a larger view. I couldn’t figure out how to scale/center them…

Thanks in advance!

Thats an unusual one. It definitly looks like a bit of underextruding, but that shouldn’t be causing the extra plastic loops. From the print image I can see it’s entering the new layers right around those 4 points, so it probably is some sort of nozzle ooze related issue, which would make sense that retraction affects it. Thats a fairly large amount of material error right at that spot not to show up as effects anywhere else in the print though.

Check the extruder out really well. Is the small gear on the extruder motor locked down and tight on the motor shaft? Is the large gear excessivly worn and or able to move on the hobbed bolt without actually rotating the hobbed bolt? A Taz 1 had unsealed 608zz bearings in the extruder itself, check the idler arm bearing and the two bearings on the hobbed bolt and make sure they are still turning correctly and smoothly. It wouldn’t be unusuall at all for any of those bearings to be putting excess resistance on the hobbed bolt.

Maybe check the hotend too. make sure the peek isolator is not melted or out of true in the bore, that everything is alligned and fillament isn’t binding at the junction between the extruder and the hot end. You may also want to run some extruder calibrations and see if your e-steps is off at all. (http://www.instructables.com/id/How-to-calibrate-the-Extruder-on-your-3d-Printer/)

You might also try upgrading the Slic3r version you are using. if you are using the stock taz 1 software, that would have came with an older version that had some circle issues.

It also looks like your belts might be slightly loose. It could just be the STL file, but the round area of the inside of what I am assuming is some sort of fan bracket looks slightly flat in segments, which can be an indication of a slightly loose belt.

On the extruder calibration you might need to run it slower or with a higher temperature. Your extrusion might be off but it should be consistently off. If its varying there might be something odd.

Ok, that’s good to know. I’ve had inconsistent checks on a T1, T4 and T5. I’ll try slowing it down and see if I can gain some consistency. The prints on my T4 and T5 are really nice so I didn’t think it was a problem. Perhaps because it generally runs at a slower rate… I’m using 235C and 100mm/min…

A quick update on this issue. I found that one of the extruder hobbled bolt bearings had a slight flat spot in it so I replaced them both. I also noticed that the large gear was slightly rubbing on the extruder mount housing and fixed that issue as well. For a different project, I’d also set this printer up with PLA so I decided to try to print the above with the PLA just to see how it’d work. I sliced it with S3D and let it print. The print came out “ok” but still looked pretty junky. I then DL’d Lulzbot’s Cura PLA profile, sliced and started printing. There’s a HUGE difference in how these prints were sliced! The retraction distance is huge and apparently some Z retraction is also being used as well as many, many more retractions. It’s going to take some time to dissect the differences in the two slices. In the meantime, the print was looking stellar after just a couple layers. More updates later…