Unusable results with ABS

[gr5 2,170]

Unlike PLA, ABS will have bad layer bonding if you print it too cool and/or with too much fan and/or without an enclosure. The part will look great and seem strong until you stress test it and it will break along the layers and you will realize that it is about 10X weaker than it should be. 10X weaker than for example PLA. Because of this reason alone it's often wise to print hotter than 230C. But if you have an enclosure, or if you have 0% fan then 230C is probably fine.

[asb 7]

For a couple of days, ABS worked quite well with the modified settings. I just have to remember to check settings after switching the UM2+ off; "Fan: off" is saved permanently, Nozzle and print bed temperature is not saved permanently to the profile.

These InMoov parts were printed in a couple of days. Quality is good. No misprints in almost a week.

But no luck with thing:172505 from Thingyverse. This is a base plate that doesn't fit flat on the UM2+ print bed (with brim and taping the edges):

I just could not get it printed upside. During the 21 hours print time, something always cracked. E.g. the edges:

Or parts of the support structures:

Once I tried just to continue printing. After about 18 hours, the whole piece cracked from the lower support structures:

Printing this thing is either for experts only, or requires a bigger print bed.

The workaround is to make two pieces of it and print them laying on the print bed. This works like a charm with 8 hours print time for each part.

Other parts of this thing printed fine as well, e.g. these mechanical eye lids.

Then the nozzle suddenly clogged for no obvious reason, which is just scary. I did not even turn the printer off of changed the filament, the nozzle just stopped spitting out material during a print. Applying the "atomic method" seven times has no (zero!) effect. Trying to melt or burn the ABS did not work well (and I am unsure how much heat I can apply before the nozzle deforms). Then I tried to dissolve the ABS in acetone. This partly works (I can remove most of the dissolved ABS from the inside of the nozzle with a  wooden tooth pick), but the nozzle still does not print. One step ahead and two steps back. Everything is delayed again until I can figure this out.

I guess the nozzles have a limited lifespan, in my case about one month. Is this to be expected? Another crucial information missing in the documentation?

The replacement parts I found on the net range between ~€15-18 plus ~€6 shipping for an "original" nozzle and ~€2.50 for a set of five no-name noozzles, which is quite a huge span. If I need €15-18 just for spare parts each month, that would be rather bad.

Are there any recommendations for replacements and where to get them in Europe? At https://ultimaker.com/en/products/spare-parts, they only list a 0.4 mm nozzle that is "only compatible with the Ultimaker Original." Where do I get spare parts for the UM2+ that work well?

Thanks!

Edited March 18, 2016 by Guest

[ultiarjan 1,223]

Nozzles have a much longer lifespan than a month, they can last years I think when you just use them for abs/pla.

only when using abrasive materials (metal fill, glow-in-the-dark) they don't last that long.

Printing that base plate standing up will never produce a good result, crazy overhangs, no base volume.

But I don't see the issue, fit's fine on an UM2, even with brim ....

Edited March 18, 2016 by Guest

[asb 7]

Nozzles have a much longer lifespan than a month, they can last years I think when you just use them for abs/pla.

 

So what is the trick to make them last longer than a month when using them with PLA and ABS?

 

But I don't see the issue, fit's fine on an UM2, even with brim ....

 

There are two issues; first, my build plate does not heat evenly. When I set it to 100°C, the outer areas (~5 cm from the edges) are way cooler. I have a surface temp of 90°C there max. Second, I need about 3 cm extra for thick tape. Without tape, the brim and the printed object rise on the edges from the build plate (visible in DSC02233.JPG, right side; yes, I am using glue).

This happened even on one side of the two-part print with ABS, a nozzle temp of 260°C, print bed heated to 100°C, fans off, glue, and tape; cleaned and leveld build plate directly before the two prints.

Here is an inner edge of the baseplate as is is supposed to be:

Here is an outer edge of the baseplate that is distorted/warped:

I can tell if it is still useable if I manage to fix the other issues with the printer and then manage to print the remaining parts.

Here is the two part print that came out almost perfect, but seems to have somehow ruined my 0.4 mm nozzle:

Now I tried to switch to the 0.6 mm nozzle from the Olsson block included in the UM2+. Regenerated the files in Cura and reconfigured the printer setting, including leveling the build plate once more and applying fresh glue. Now all I get is this:

100% failure rate. Even with a factory fresh nozzle I am back to square one and the initial question ("unusable results with ABS").

Edited March 18, 2016 by Guest

[SandervG 1,521]

Did you also ever try an ABS slurry?

(it is mentioned here )

[asb 7]

Did you also ever try an ABS slurry?

I have read about it, but never tried it. The German distributor recommends a 1:2 mix of acetone and ABS, which I'd consider a very last resort if everything else does not help. I'm not particulary happy to add even more hazardous fumes into my office if it can be avoided. So far, the glue worked most of the time.

Though at the moment I neither have a useable 0.4 mm nozzle nor can I get the factory fresh 0.6 mm nozzle to work, so it seems I'm doing even the basics wrong. Everything appears so totally erratic :-(

[gr5 2,170]

Try 110C.  Make sure it is at least 100C on the sides - like you said it is cooler on the sides.  

One way to get the glass up to temp faster is to put a towel on the glass by the way.

To get to 110C you will need to enclose your printer.  I know - you said something earlier about "but UM said it works with ABS so I shouldn't have to modify my printer".  Well putting a box over it which takes a few seconds to do isn't much of a modification but - well - printing in ABS *and* printing to the edges of the glass probably *will* need modification if you consider putting a box over it a modification.  Also cover the front with saran wrap.  You want the air around 40C to 50C.  Much hotter than 60C and it gets bad for the steppers.  But it's really hard to get the air hotter than 50C anyway so no big deal.

I'd like to see a photo of where your brim meets your print when you've only printed 2 or 3 layers - I want to see what the brim looks like - does it have gaps?  Is it squished - from the photos above it looks fine but it's hard to tell from the angle.

Ideally print this flat - printing this part vertical is definitely the hard way to do it.  If you ahve to print it veritcal then tilt it diagonally so it is away from the edges of the glass (in other words one end is near the rear left corner and the other end near the front right corner - that's what I mean by "tilt").

If your part hasn't literally ripped chunks of glass off the bed (that happens to everyone eventually) then you can do more to get your part to stick.

Edited March 18, 2016 by Guest

[gr5 2,170]

Now regarding the nozzle - I don't know but soaking it in acetone overnight (8 hours) should help a lot.

Did you check your feeder to make sure you didn't grind the filament badly there? I'm thinking you got the cause and effect backwards. it can easily happen either way:

1) too many retractions causes filament to grind causes filament to stop moving causes ABS to bake into a permanent plug

2) dust causes nozzle clog which causes filament to grind to dust which causes ABS to bake into a permanent plug

1 - is more common.

I've been able to burn out PLA - it helps to also use a wire and to do some test feeding over the stove. I haven't had to burn ABS out yet - toothpick was always good enough in the past so I'm not certain what the next step is other than soak in acetone for 8 hours.

[peggyb 170]

first layer with thicker nozzles is more difficult and needs some practice to get it right. Try slowing the first layer down and the leveling of your bed and the 'first layer height' might need adjustment. While the brim prints, gently push/pull the bed by hand to see what the effect is: better or worse for up or down, and dial the screws accordingly. Always watch he first layer and grab the loose stuff with tweezers so it doesn't melt to your nozzle, was that the case in your picture?

[asb 7]

I think I'm giving up. The 0.6 mm nozzle just does not work at all; if glue is on the print bed, it works like soap or grease (zero adhesion). The more I heat the print bed or the more I enclose the build space, the worse it gets. The print only sticks without any glue on the print bed after the print bed was cleaned with ethanol (not just tap water). It's quite the opposite of what everybody is telling me. I do not understand why this is so absolutely not working.

The 0.4 mm nozzle I can not get clean. When pushing filament manually into it with the "atomic method", some plastic drips out of the nozzle, so it's not totally clogged, but nothing gets through in normal operation.

Wasted one day trying to clean the "old" 0.4 mm nozzle (without success), wasted another day trying to get the new 0.6 to work (without success). This thing is a total productivity killer for me and I have no ideas left where to start debugging gazillions of issues.

Btw, if changes to the ABS profile are saved of not is erratic. Sometimes the 260°C nozzle temp setting is restored, sometimes it is not not after switching the UM2+ off. The German distributor says that "most of the profiles" can not modified (whatever that is supposed to means) and that modifying the setting is not documented intentionally. Amazing.

Also, if the fan is set to 5%, it does not turn, the setting is identical to 0%. I don't think that this causes any of the issues, but I just don't understand it.

[gr5 2,170]

There's a new bug introduced recently that makes editing the profiles buggy. It's fixed in source code but not released to the general public yet. There are work arounds involving saving to materials.txt doing certain edits and then loading it back in. The work around seems to work. The other work around of course is to just set it every time.

asb - I know what you mean - it's hard for us to tell exactly what's going on without making sure your filament is feeding the exact amount requested when printing through your nozzles to make sure it isn't underextruding and also if I were there I would be playing with the leveling as it prints the bottom layer. But it could be a bunch of things - most of which can be fixed quickly. Like getting another .6mm nozzle.

When you use the .6mm nozzle you *do* set the nozzle size to .6, right? And shell width is a multiple of .6, right? If you skip either of those 2 steps you get a disaster. Fortunately Cura hints that there is a problem with bright yellow background for shell width if you do that part wrong.

[asb 7]

When you use the .6mm nozzle you *do* set the nozzle size to .6, right?  And shell width is a multiple of .6, right?

You mean in Cura? Yes.

Here are some more attempts in printing with the UM2+.

0.6 mm nozzle, increased print bed temperature to 110°C, enclosed the device as good as I could:

Increasing temperature definitely does not help.

Removed enclosure, decreased print bed temperature to 100°C:

100% failure rate. No idea where to look for solutions.

Didn't touch the UM2+ over the weekend. Then switched it on again with identical settings. Suddenly it printed more or less fine. The prints with the 0.6 mm nozzle look mostly crappy, but printing a couple of pieces completed. This is absolutely erratic. There is no pattern I could recognize; same device, same settings, same filament, same files. One day nothing works, another day it works well. The only differences I could think off: Outside it's raining and a couple of degrees warmer than two days ago. Is the wheater and the humidity outside responsible for 100% failure rate? I hope not.

ABS filament is now empty. Wasted about 70% of the material without learning anything except: Print success depends on luck.

Switched nozzle again. Now trying 0.25 mm nozzle and another PLA filament.

Everything begins at square one. Absolutely nothing works. It does not stick on a clean print bed, and it does not stick on glued print bed. That's how the "results" look:

Zero adhesion to the print bed. The nozzle appears to scratch the printed material off the print bed which then sticks to the nozzle. Supplied glue is empty. Trying another one called "uhu" (water soluable glue for paper). Same results, nothing sticks to the print bed.

Strangely, the fans do not turn. If I remember correctly, they did turn with the PLA profile. With the "Tune" menu, I set fans to "100%" (too bad I do not have a reference for the factory settings). After choosing "Resume print", the fans go off again. WTF.

Again, factory fresh nozzle, PLA filament, 100% failure rate.

Switching off the UM2+ and let it cool down. Sometimes this seemed to help. Switching on again. Now it doesn't even start anymore. That's great. This thing is about a month old and already dead. Quite a disappointment for a device that is considered easy to use and reliable.

Edited March 22, 2016 by Guest

[gr5 2,170]

Man you are having so much trouble!

Well I feel like I'm talking to a brick wall. In the first photo with the brown stuff your leveling is off again. What did you do? Did you level it again? Or is the new nozzle shorter?

Do you see how in the first photo the brim has gaps? That's not good - why didn't you stop the print right there? The first layer is the most important by far. If the first layer isn't going to stick why bother doing the second layer? This was what I talked about in my first posting in this thread - you need to squish the bottom layer into the glass. You've made lots of long posts. I've made lots of long posts. Still not doing the "squish" thing on every print. I see some of the other's look fine though.

So when you start a print have your hand on the 3 screws - be thinking about which way the bed moves when you turn CW or CCW. Be ready to fix that brim before it finishes the brim. Now it's so easy I don't even think about it. Like I don't have to think about which way to turn the steering wheel when driving a car.

The photo with the glue - too much glue - take a wet napkin and remove most of it - the layer of glue should be invisible once it dries.

And really it's too bad about the ABS. PLA is hard enough as it is, ABS just has a couple more things making it extra tricky. There's a new "coffee" filament from protopasta that is PLA but after you print with it you bake the parts for a short time and it changes color and then is heat resistant just like ABS. So if you really need your parts to withstand boiling water that might be a consideration. There may be down sides - this is not a popular material yet. Another bonus - it smells great when you print with it.

About the fan - this thing is very simple - if you looked at the gcodes and read about them you'd hopefully understand by now. There is a gcode that sets the fan speed - the gcodes are just read in order - very simple - no if statements - no loops - just read in order like reading a book. If a gcode sets the fan to 10% it does it right then and there even if you set it to 100% a few seconds ago on the control panel. It just goes to whatever speed was sent most recently.

If the fan speed is < about 10 or 20% it doesn't turn at all because that's not enough power to get most of the fans to turn.

[asb 7]

The photo with the glue - too much glue - take a wet napkin and remove most of it - the layer of glue should be invisible once it dries.

Thanks again for your hints. Though this is just not what I am experiencing. The ABS filament I had been using did either stick well on a completely cleaned glass plate (see above, ethanol, not water), or on thick layers of (mostly dried) glue. Glue and wet napkin - glue becomes like grease. Zero adhesion on a thin film of glue. Zero adhesion if glue was diluted with water.

Again it's different with the "ICE PLA Wondrous White 2.85 mm 750 g" which sticks on nothing, neither on a clean print bed nor on any thick or thin layer of glue.

 

Do you see how in the first photo the brim has gaps? That's not good […]

…and that's my point. It's erratic. Without changing print bed levelling, it looks like this (with more heat), or has no gaps at all (pictures below, with less heat).

I need procedures that I can reproduce. After wasting over 2 kg of filament I still have no clue why the UM2+ reacts differently everytime I switch it on. It's mindless trial and error.

Anyway, since it can not even be turned on anymore, someone else has to bother with it now. Let's see what the vendor says.