Help printing ABS

[mopar]

I'm at work right now so I can't upload the stl of the part I'm trying to print or a pic (which I'll try to add later), but I do happen to have the gcode with me (had to zip it because of size) on a flash drive so I've uploaded it. I coated the glass with Elmers Xtreme glue stick before heating the bed. The bed temp was at 110º and the nozzle was at 250º at the start (I manually dropped it to 230º roughly an hour into the 6hr print). My printer is a Prusa i3v from Makerfarm with an e3d v6 hot end and Gregs hinged extruder. Used Slic3r 1.2.9 to create the gcode. The printer is in my basement which is usually around 15º. I made a crude cardboard box to help hold some of the heat in. I bought the white filament from Makerfarm when I bought the printer.

The part is roughly 105mm in dia by 33mm tall (looks like a 6 spoke car rim). About half way thru the part started curling up off my glass (about 1.5mm all the way around) which isn't my only problem with this item, but probably the worst.Another problem I have is I seem to have a lot of bubbling (have same issue with PLA as well). I don't know if it's from moisture or over extrusion. Knowing this was going to be a long time printing and use a lot of material I checked the calibration of the extruder before printing. It was about 6mm short at 100mm so I adjusted the steps and tested again. Came out right at 100mm (maybe 1mm more) so I left it at that. Everything started out good, then some time into it I noticed that I could hear the hot end drag across the part and I could see bumps in the print. I went in and adjusted the steps for the extruder half way between the new settings and what they were originally while it was printing. It seemed to help some, but an hour or so later I could here the same problem so I dropped the temp on hot end to 220º thinking maybe I was having oozing problems. Continued to let it print and still noticed the problems so I dropped the steps to just under the original value, dropped the feedrate over ride to 95% and let it finish the print as it was time for bed (roughly half an hour left at this point so I felt safe letting it finish. Because of the design I had to add support material ended up being pulled off glass in a few places but seemed to stay put for the most part (although the print ended up separating from the support sometime during printing). I also have layer separation in several spots.

I know without seeing the actual print it's hard to give suggestions (I'll try to get one uploaded shortly) but anything to start with would be helpful. Seeing I have the same bubbling effect with PLA at the original extruder steps I'm at a loss on what to do. If I remember correctly I had the bubbling problem when I printed the 24mm test cubes when I 1st got printer a month ago with both PLA and ABS.

Sorry for the long post, but wanted to make sure I tried to get everything.

P.S. On another part I tried printing with ABS I tried using hairspray, blue painters tape and plain Elmers glue stick and they all pulled off the glass after printing to a height of about 6mm. The Xtreme was the only thing I found that would hold the print down. I have some 3M spray adhesive I could try but was worried part would never come off.

[mopar]

Here are a couple pics of the part. The side with 6 wide spokes was down on glass. As you may be able to see there are gaps in the fill (besides the holes) on the spokes. There is also separation of some of the borders in pic#2 that hopefully you can see. The spokes are 9mm wide and 3mm wide on side with cross in center of rim. The roughness of the part isn't a concern as I can sand, file, acetone that clean. Just noticed that none of the pics show the separation of layers. I only printed in with the wide side of spokes down because I figured the supports would print better this way than if I flipped the part and it had to print support all the way up to the step where the spoke goes from 3mm wide to 9mm wide. None of the pics show the bubbling that well, but you can see it slightly in the 2nd pic.

Thanks.

[3dex ltd]

Sorry to hear you're having difficulties. ABS is a difficult plastic to print with but rewarding when you have it mastered.

So the warping issue should be fairly easy to deal with and can be done by treating your print bed. Many people benefit from sticking painters tape or masking tape to the print bed and then printing directly onto this tape. This gives great adhesion. Personally, we are a fan of the glue stick method. We smear our print bed with a layer of paper glue (called Pritt Stick) which works really well for securing ABS to the bed. However, if these two methods don't work then I have heard of people mixing ABS and Acetone together and then spreading the mixture on the bed. The result is that the print sticks to the bed incredibly well (can be a pain to lift off the bed)

The other issues you have are probably solvable by tuning your printer. You might have to play around with the settings to find the best. The layer separation is probably due to the fact that heat is escaping from your printer. ABS benefits from being printed in a hot air temperature. To make this happen a lot of people enclose their printers in a box so that the air within the box remains hot. You might also have success if you can fix a fan to blow hot air over the print for its duration.


I hope some of these help!

[djprinter]

Getting more heat into you're build chamber will help a ton with the warping. I cut a hole in my enclosure and attached a hairdryer which I keep on low, increases chamber temps up to about 125 degrees and I get zero warping.

[mopar]

When I tried the painters tape and hairspray I didn't have a chamber around the printer to keep the heat in so that could be why they failed. When I 1st tried the Elmers Xtreme glue stick (tried regular glue stick which failed, but I didn't try to contain any heat either) I took a little portable heater I have and had it blow directly on the ABS which is why it probably worked. Haven't tried the ABS slurry yet, don't want to deal with the mess and difficulty removing the part.

The consensus is it looks like I need a better way to contain the heat than some cardboard boxes tapped together I guess. Hard part is dealing with the filament holder as it sits on top of printer and makes it difficult to cover completely. Gonna look for a better holder and talk to the local sheet metal shop that my work deals with and see if they can make up a thin box with a small window (can glue some lexan to it so it's sealed) so I can see inside while it's going. Ribon cable for lcd is quite long so maybe I can have it so I can put it outside the chamber so I can adjust things on the fly.

Thanks for the info.

Would a wooden frame wrapped in poly work to contain the heat if I can seal it up pretty good? Not that I'm any better at carpentry than I am at 3D printing, but the deck I built last year is still standing so I have some hope. lol

[mopar]

OH yeah, would it be better if I flipped the part 180 and put the 3mm width of the spokes on the glass bed? The outer ring is 5mm wide on the 3mm side compared to the 3mm width of the ring on the 9mm wide spoke side.

Even though I need more steps per mm (than default) to get actual extrusion length correct is it possible that I would need to run it with less steps to stop the bubbling?

It only seams to bubble on infill. and flat layers which I assume would indicate over extrusion? I typically try to do the perimeters 1st to get the print better dimensionally so if I'm over extruding on perimeters I'm making the area requiring infill smaller which in turn because I'm over extruding the printer puts in more filament than the void con hold forcing the filament up?

Pretty sure I had this same problem on smaller parts with the ABS with the default steps so I can't imagine it would be good to reduce the steps further. I should probably measure the actual filament size as I'm sure that can cause some problems as well. Typically what should I look for in size variance before I should compensate for filament size?

Thanks again.

[LambdaFF]

You shouldn't change the steps during the print. It either is the correct step or it isn't. The reason why your print head bumps into the print during the print isn't step calibration. It is most probably due to the warping but it could also be due to other factors : improper levelling, restrained Z axis movement, lack of consistency on filament diameter ..

By changing the steps during the print you are muddying the issue and you make it harder to find the root cause. You could also try printing he first layers real slow.

The bubling can be due to humidity. How humid is your area, has this rolled been opened long ? Try to store it with desiccant at least. I don't pre-heat my filaments but others could give you tips on that. The humidity may contribute to uneven flow and cause local overextrusion that will make the head bump into the print.

Having the spokes on the bed would definitely help. That's a good point.

[mopar]

Did some test printing on a 24mm cube last night. I did both with the same settings with exception to the extrusion multiplier. In the 1st pic the part on the left the multiplier was set at 1, the part on the right was set at .95 (I had the steps set the same for both at what I calculated they should be). Dimensionally they are the same on X & Y axis at 23.8 (which is .2 to small but is another problem to fix later). However you can see the bubbles on the sides of the cubes. The bubbling is in different areas on individual sides (except one side has none) but both cubes match each other (which I assume is either the start or ending points of the outer perimeter since they are very similar). The height of both cubes are the same at 24.05 so my Z is repeating and accurate. Don't think it's over extrusion like I did before, but something else.

The cube I did when I 1st got the printer was sliced with slic3r 0.9.9 with Makerfarms supplied config. file for ABS. It did not have the bubbling on the outside but the dimensions on the X & Y were the same at 23.8mm. The new cubes were sliced with slic3r 1.2.9 with same config. files as 0.9.9. Not sure in 0.9.9 but I've heard that the default extrusion width slic3r uses for a .4mm nozzle is .68 in it's calculations (when the value is left a zero), so wouldn't that cause the dimensions on X & Y be off roughly .3mm? Assuming that, I'll go home tonight and put a value of .4 in slic3r create new gcode and print another cube. Does the layer height play a hand in extrusion width as well?

Since the bubbling is in the same spot (or very near) on both cubes I don't believe it's a moisture problem (have no idea what the humidity is honestly). My understanding is that moisture problems usually result in voids where the water was vaporized and blow out thru the filament. I check and set bed level any time the bed moves when I try to take a print off.

I checked the filament diameter last night and it's actually 1.69/1.70.

Not sure if you can see it in the 2nd pic or not (this pic is of top of cubes) but the cube on the left (1 in multiplier) seems to have higher ridges than the one on the right which was printed with a multiplier of .95. Almost like the printer was trying to extrude more filament than there was room for and the extra got pushed up. Could this possibly be linked to the extrusion width of .68 slic3r is using as well? Don't recall seeing that problem on 0.9.9 sliced cube. I would think that the extrusion width would be the horizontal step over slic3r uses when creating the gcode for top, bottom and perimeter paths, but maybe not?

I know I have a lot of questions, but I'm just trying to understand how each settings effect things. I don't like changing the wrong setting to fix something that it wasn't intended to control.

Thanks.

[printbus]

EDIT: I agree with your thinking that the bumps are not moisture related. They just aren't random enough.

Take a look at the gcode file for the cube in a viewer like gcode.ws. See if you can find a correlation between what the gcode file is instructing the printer to do and those bumps on the walls. They're likely something like the spots where the printer momentarily stops XY movement to raise up to the next layer.

There seems to be a trend of people using MakerFarm's old config files for Slic3r 0.9.9 with newer versions of Slic3r. Note that slic3r has added new configuration settings along the way. I would imagine Slic3r defaults to some value for the configuration parameters missing in those old files, but is that something you really want to rely on?

[tpete61]

Try lowering your bed temperature. I use Aquanet hairspray at 95C and have better luck than the higher bed temperature.
I definitely warp less with a lower temperature. Also try printing with a brim of 2 or 3 layers that you can trim of with a hobby knife, this sometimes works well. I've printed items that are 10" X10" on my bed and successful more often than not. The shape of the part also seems to be a factor in the success of a part not warping.

On a side note I never change the nozzle temperature during an ABS print it tends to effect the surface quality of the part when I did that in the past.