Simplify3D producing terrible ABS prints on Mini

so, been chasing an issue here for a little while as ive been dialing in my new lulzbot mini. it as determined a week or so back that my ABS had become contaminated and thats why my prints looked like crap. dried it out, did some testing tonight…and the Simplify3D results…are pretty awful. CURA, which was also making the prints look awful prior to drying out the filament, seems to be now printing pretty nicely.

printed this test cube, that i made in fusion360 so i know its 20x20x10

.50mm nozzle
Auto Extrusion width
.20 layer height
100 percent infill
black ABS @240, bed @ 100



The scars on the top make me think there’s some over extrusion. Slowing the speed some might help with the somewhat messy corners. Outside walls and corners can also be made to look off by over extrusion.

Note that the flow % settings for Cura and Simplify3D don’t seem to match exactly. So you need to tune the esteps or just accept that there are differences. I have different settings files for both that I use.

Somewhere on the forum here there’s a Simplify3D file that has improved settings for various materials. That might be a good place to start. If you post your factory file, some of us can load it up and try to help out as well. I can post my fff file tonight, but honestly, that other one is probably the better place to start. I haven’t printed a lot of ABS, but I do get good prints with it.

Had to go find the post I was talking about…

I don’t see a way to link individual posts, but for me it’s the first post on the second page by scottw. He has a nice customized S3D profile there that should help you get good settings to start from. The HIPS profile I posted might also work decently on ABS, though I would turn the fan down some for ABS.

I fought that top scar for a while and finally figured it out. In S3d open Edit Process Settings and go to Advanced tab. Under Ooze Control Behavior, check the box at “Only Retract when crossing open spaces”.

I’d reduce the Extrusion Multiplier (flow) a few % until it looks better. Also change Extrusion Width from Auto to Manual. I set mine a 0.50 mm. That seems to make more sense since your nozzle size is 0.50 mm. And lastly, avoid using 100% infill because it limits where “extra” plastic can go if it does over extrude a bit. 90-95 should probably be the max infill you need. I usually run 50-70% for high strength items.

Here is a copy of the S3D settings for different filament courtesy of ScottW. Theses are excellent places to start.
S3d Lulzbot v2 Universal Profile.pdf (100 KB)

youre saying that if i (example) print something in CURA at say .97 on the extrusion multiplier and then use everything the same as far as settings goes in S3D, the extrusion multiplier set at .97 as well…the results will vary?

this is the factory file that i used.
20x20x10_Calibration_Cube.txt (3.63 KB)
i assume you have to change the file extension to a .txt so it will upload on the site then you can change it once you download it(?)

i left that unchecked because thats what the PDF that i have, the one you posted, said to do. unless i read it wrong. however, i will check it off as you say and do a print tomorrow hopefully and see if that helps.

ok, i have read different views on this. personally i think that setting it to .50, the size of the nozzle, is the right thing to do in my brain. BUT as i said, a lot of the threads ive come across online have said things along the lines of “keep that extrusion width on auto because youre not smarter then the software”. to me, id rather it be AT a set value that i know and then if im designing a part i can model it accordingly, knowing that ill need XXX thickness for a wall persay because im going to need XXX shells if im printing at .5mm extrusion width.

with the 100% infill, and again this is just what ive gathered from reading online so ANY and ALL help is appreciated here, that in order to fine tune the Esteps you want to print at 100% infill because that will allow you to see how everything is going. as apposed to just waiting for the top layers to print and checking those out.

for Gcode start/end sequences in S3D do you guys use the stock S3D supplied script? or do you just cut and paste from CURA?

I use the stock scripts from cura with hard coded temps. I also remove the move to cooling position gcode from the ending script. Removing that bit solves a connection problem with octoprint.

The main difference I see in your S3D Factory file, vs. default Cura settings, is cooling fan. You have no cooling fan in the S3D profile, whereas the default Cura profile for ABS has 40-60% fan after first layer. With black ABS, no fan, a fairly small object (quick layers), and fill set at 100% as in your factory file, the bad results you are getting may be related to the part retaining too much heat.

Try importing the FFF profile linked below into S3D. Then edit your process settings, select the imported profile, and choose Material:ABS and Quality:Medium. Give that a try with the default 20% fill. If that works out, try it again with your 100% fill and see how it looks.!Aj5_Sqpk1E8cvGxCK1m2Bjli1Oxq!Aj5_Sqpk1E8cvG2cTNvm0d3VgDIY

I will hopefully attempt this tonight. Thank you! I was under the impression that you didn’t want the cooling fan to run while printing ABS?

Like so many things with 3D printing, I think the answer is “it all depends.”

I’m still learning, but it seems with larger parts I get better layer bonding and less tendency to warp with the fan OFF. However, with small-ish parts, between the heat from the bed and layers being laid down quickly, the heat can build up and cause deformation – so for smaller parts, I usually go with the 40-60% fan for layer 2+ (which I got from the default ABS profile that ships with Cura Lulzbot Edition). And when I say “40-60%”, I use the options in Cura/S3D that start the fan at 40%, but slow the print speed as low as 20% to try and achieve at least 15-20 seconds per layer, and increase the fan speed to 60% if still less than 15-20 seconds.

Another option often recommended (instead of using fan) is to print multiples of smaller parts, so that one can cool a bit while the other(s) are being printed. That’s perhaps a better option for quality, but can be wasteful and time-consuming if you really only want (1) part.

When viewing the thread, right click on the title line of the post to which you want to link (or if you are on a Mac, control-click). Then copy the address, paste it into your message, and use the URL button to make it clickable.

I don’t print at 100% infill, since there is no place for any extra filament to go if you are over-extruding a bit (though I guess I can see trying it if you are playing with calibration, as you mentioned).

Here’s a link to a good procedure for fine-tuning your e-steps if you want to take it beyond the basic “extrude 100mm and measure the distance it moved”: Triffid Hunter’s Calibration Guide

This has been another issue for me. Different schools of thought for how to go about what I want to do. Other then the free air extrusion method, which then Ive read you’re supposed to go ahead and print a single walled square to dial in the extrusion. But others have said that the single walled square doesn’t really tell you much about how it will extrude while infilling and such.

Extruding into the air and then measuring the extrusion. You should get .5mm right? The size of the nozzle.

What does the collective use for extrusion width settings? Manual or auto? I’m still not sure what would best suit my needs

I’ve never measured the diameter of the extrusion when extruding into the air. It will not necessarily come out the same as the nozzle diameter. It is somewhat dependent on temperature, extrusion speed, and the amount of weight of filament hanging off the nozzle. So varying your e-steps to try to get your exact nozzle width when extruding in air won’t really get you anything as far as indicating e-steps calibration.

I figured there would be too many variables. I wasn’t doing that to calibrate the esteps, but did measure it a few times after extruding into the air and was just curious is all, because it was a bit over .5mm.

It’s a common thing when running something through a die to reduce it’s size. I used to work in a company that draws steel wire down by passing it through a “funnel-shaped” carbide die. After the funnel, there was a bit of straight-sided cylinder (called the “bearing”) built into the exit end of the die. The length of the bearing was important in controlling the size of the wire: with little or no bearing the wire would come out larger than the size of the die. Same thing happens with filament.