Unusable results with ABS

[asb 7]

Hi,

I got my UM2+ about a week ago; opposed to my former UM2 it worked out of the box, e.g. I did not have to dismantle the feeder, the build plate was not misaligned. Nice! Printed a couple of things with the supplied "Ultimaker PLA Silver Metallic", no problems encountered whatsoever. Print quality is very similar if not identical to the UM2. All parts printed in PLA are solid and usable. No failures, no misprints. I was happy for almost a week.

Then I switched the material to "Form Futura EasyFil ABS White", 2,85 mm, which my vendor especially recommends for the UM. I changed to the default settings for ABS and applied glue to the print bed (never needed that for PLA, but I followed the recommendations). The printed base did not stick to the glue provided in the UM box. In the next attempt, parts of the ABS stuck to the prinbed, others stuck to the nozzle.

Tried to encapsulate the open front of the UM2+, next attempt resulted in no object printed at all, the UM2+ filled with a unstructured spiderweb of thin ABS fibers. Everything I tried resulted in totally unusable pieces of plastic, mostly not even resembling the object to be printed. The best result I could get was an 8-hour-print of the InMoov face. It cracked horizontally in seven areas during printing, and the left side split open for about 10 mm after 4 hours of printing. Is this to be expected? Am I doing something wrong?

1) As I said, I'm working with the default settings for ABS. Should these settings provide a reliable base, or are they totally wrong?

2) Which settings need to be changed? Where? In another thread it was mentioned that for printing ABS, "Fans should be at 0% for ABS (off)." My fans were on. Are there settings/printer profiles which include those recommendations?

3) Are issues like this to be expected with printing ABS on a unmodified UM2+, or is the material ("Form Futura EasyFil ABS White") not good, e.g. not compatible or otherwise not recommended for use with the UM2+?

I am really puzzled. No problems with PLA and zero usable results with ABS. Is the UM2+ really designed to support both types of material equally?

Thanks, Sandro

[asb 7]

Here are some photos for illustration:

[gr5 2,167]

Welcome to ABS. It takes maybe 100 prints to get really really good at PLA. You probably forgot this - we tend to forget the bad and remember the good. Well it also takes quite a few prints to get good at ABS.

You mention several problems including: spaghetti, not sticking to glue, parts not sticking, horizontal cracks. It would help people posting answers to stick with just one or two problems per post. So don't be surprised if after my post you only get answers to one of your issues.

The sticking is the easiest to fix. make sure bed is at 100C. Minimum. It makes it easier to get there if you cover the front and/or top of the machine but not mandatory to get ABS to stick like hell. Probably you just need to squish the ABS into the glass harder. The leveling procedure only gets you close - then you want to ram it in a little harder.

I'm going to paste some information here about sticking - the issues are identical to PLA except ABS has MUCH more shrinkage force than PLA and a higher glass temp but otherwise same advice.

1) Make sure the glass is clean if you haven't cleaned it for a few weeks. You want a very thin coat of PVA glue which is found in hairspray, glue stick, wood glue. If you use glue stick or wood glue you need to dilute it with water - about 5 to 10 parts water to 1 part glue. So for example if you use glue stick, apply only to the outer edge of your model outline then add a tablespoon of water and spread with a tissue such that you thin it so much you can't see it anymore. wood glue is better. hairspray doesn't need to be diluted. When it dries it should be invisible. This glue works well for most plastics.

2) Heat the bed. This helps the plastic fill in completely (no air pockets) so you have better contact with the glass. For ABS 90C is enough.

3) heat the bed (didn't I already say that?). Keeping the bottom layers above the glass temp of the material makes it so the bottom layers can flex a bit (very very tiny amount) and relieve the tension/stress. For ABS you want 110C (100C is good enough).

4) rounded corners - having square corners puts all the lifting force on a tiny spot. Rounding the corner spreads the force out more. This is optional if you use brim.

5) Brim - this is the most important of all. Turn on the brim feature in cura and do 10 passes of brim. This is awesome. (I notice you did this already)

6) Squish - make sure the bottom layer is squishing onto the glass with no gaps in the brim. The first trace going down should be flat like a pancake, not rounded like string. don't run the leveling procedure if it is off, just turn the 3 screws the same amount while it is printing the skirt or brim. Counter clockwise from below gets the bed closer to the nozzle. Don't panic, take a breath, think about which way to move the glass, think about how the screw works, then twist. This may take 30 seconds but it's worth it to not rush it. You can always restart the print.

If you do all this you will then ask me "how the hell do I get my part off the glass?". Well first let it cool completely. Or even put it in the freezer. Then use a sharp putty knife under a corner and it should pop off.

[gr5 2,167]

Now to address split layers and something you didn't mention.

The split layers are an ABS only problem. They are caused mostly because you don't have good layer adhesion (never a problem with PLA). The glass temp of ABS is about 100C so you need to get the next layer down well above that - maybe 140C when the layer above touches it. This means you don't really want the ABS to cool below about 90C. The simplest fix is to reduce fan. You really need very little fan with ABS. I would think layer thickness matters but I'm not sure (thicker layers have more thermal mass to melt below but if you print extra slow and thein you have the nozzle almost heating layer below).

More importantly: less fan and consider enclosing printer and print hotter.

Now to the second issue - how hot? Well unlike PLA, ABS will cook to a solid gummy gunk in just a minute or 2 in the nozzle at a mere 255C. But if you print at only 240C you probably won't get good layer adhesion (unless you enclose the machine which helps but isn't mandatory). So if you *do* print at 255C or hotter than print fast - leaving the nozzle at 260C for a few minutes without extruding is disaster.

PLA also cooks to solid gunk but it takes more like 10 or 50 minutes. Not 2 minutes.

[Nightram 0]

So print fast and hot?

45mm/s?

Do you have a g-code Marvin or something you can share

[gr5 2,167]

I'm not an ABS expert - I have only printed maybe 30 abs parts. If there are absolutly zero overhangs (all sides vertical or within 10 degrees of vertical) then print at 245C and zero fan.

If I have overhangs I usually set the fan to 30% in the ABS profile which multiplies 0.3 X the gcode fan speed. In other words it never goes higher than 30%. Oh and I'll make sure to print 35mm/sec minimum and 250C minimum (sometimes I increase to 255C after the first layer is done).

If you enclose your printer you should be fine at 240C to 250C with 30% fan.

My settings are not to be taken as necessarily that great - I just don't have enough experience with ABS - most of my expertise is in reading what other's posted. But I have to say I have had my share of frustration with ABS and learned a few things (clogged nozzle due to printing too slow at 255C, layers not bonding due to fan at 50% and nozzle at 245C, and plenty of successful prints also).

Also nozzle temperatures vary by +/- 5C quite commonly from printer to printer mostly due to how well the temp sensor fits. If it is loose it reads low and your nozzle is 5C hotter than nominal. If it is stuck in there well or if it is all the way in it reads hot and your nozzle is 5C cooler than nominal. So my 255C might be the same as your 245C or 265C.

[asb 7]

Thanks gr5!

I think I won't touch ABS anytime soon anymore. Because the UM2+ is supposed to support ABS, I thought I could make it work out of the box by changing the filament and changing the settings to ABS. I would expect having to do experiments when printing with exotic (not factory supported) filaments like wood or metal, but not with a supported material like ABS. I do not have time to make 100 test prints with ABS. At the moment I just want to get things done.

So I'm back to the PLA filament supplied with the UM2+. Here are some reference images how the first objects looked like before I tried ABS:

I'm quite happy with these results. Some sanding, and I can work with them. That's the quality I expected to get with a €2300 printer.

However, after removing the ABS filament spool, cleaning the nozzle with the "atomic method", and leveling the build plate, I can not get clean prints with PLA anymore. It appears as if trying ABS once had totally misaligned the whole printer.

That's what I get now:

Some of the objects printed are no total failures; with a lot of filing I can fix some of them. But I'm not happy with the quality anymore.

Other observations: I have to level the build plate after every print job that ran for 6 hrs. or longer. That means: The UM2+ requires manual maintenance at least once daily, usually preceeding every print. From what I have noticved, the probability to get bad prints with PLA increases the more narrow the nozzle is to the print bed. It touches overhanging material and rips the printed structures apart.

Also print quality seems to get worse from day to day.

The photo above is from a 9 hour print as I found it this morning in the printer. I somehow assumed that it would be possible to run the UM2+ unattended. This is a total failure, wasting a lot of material and energy. So do not let your UM2+ run unattended. Get a respirator and ear protection and babysit your printer. Really?

Currently, the UM2+ does not produce anything but gibberish anymore. Here are some other attempts from today:

Somewhere during printing, the structures crack. The PLA seems not to stick to the neighboring layer anymore.

All the source STL files are from the InMoov project, so they were printed a hundred times across the globe and I consider it safe to assume the the STL files are as good and reasonable as they can be.

I absolutely do not understand why a device received less than two weeks ago would produce that amount of failues with default settings for PLA and exactly the material supplied with the printer. I just don't get it.

Edited March 2, 2016 by Guest

[gr5 2,167]

It's probably going to take you 100 PLA prints also before you can get reliable prints. Owning a 3d printer is like owning a lathe - there's several things to learn. The main problem you seem to be having is related to leveling.

Do not use the leveling procedure more than once. You already did it. You are done with that procedure. Now you need to fine tune it while your printer is printing the bottom layer. There's two different leveilng heights I print at. 95% of my prints are printed such that the bottom layer get squished down a bit such that when the part comes off you can see a very tiny rim around the part where it touched the glass. Even if you don't use the brim feature.

You haven't leveled to that particular recommended height so you are getting parts that aren't sticking well to the glass and hence the "spaghetti". I only have to level about once every hundred prints (actually less often than that) so once you get the proper level height you can stop. So start off by turning the 3 leveling screws counter clockwise as seen from below - turn each screw the same amount - try 1/3 to 1/2 a full turn. Then start your print - watch the begining of the print - if it's not squishing the filament flat like a pancake then it won't stick well so adjust the 3 screws while it's printing! Hopefully it is already "level" just not at the right height which means you want to turn all 3 screws the same amount but you can compensate a low corner at this time also.

Ideally you do this at the start of your next 10 prints until you are an expert and by then you won't be moving it anyway.

For me when I print something relatively different (smaller, larger, mechanical, articstic, water tight, hollow, whatever) then there are new things to learn and new challenges.

For example you can reduce stringing by lowering the temperature to say 190C but the lower you go in temperature the slower you have to print. That should probably be a different topic for you on this forum - in otherwords feel free to start a separate topic about each of your issues - one for strings. This topic for parts not sticking. Here is the full advice to get parts to stick to the bed well.

If you stick with this printer a while longer eventually you will stop being frustrated with it and will love it.

=======

lifting corners, curling corners, part sticking to glass

1) Make sure the glass is clean if you haven't cleaned it for a few weeks. You want a very thin coat of PVA glue which is found in hairspray, glue stick, wood glue. If you use glue stick or wood glue you need to dilute it with water - about 5 to 10 parts water to 1 part glue. So for example if you use glue stick, apply only to the outer edge of your model outline then add a tablespoon of water and spread with a tissue such that you thin it so much you can't see it anymore. wood glue is better. hairspray doesn't need to be diluted. When it dries it should be invisible. This glue works well for most plastics.

2) Heat the bed. This helps the plastic fill in completely (no air pockets) so you have better contact with the glass. For PLA any temp above 40C is safe. I often print at 60C bed.

3) heat the bed (didn't I already say that?). Keeping the bottom layers above the glass temp of the material makes it so the bottom layers can flex a bit (very very tiny amount) and relieve the tension/stress. For PLA 60C is better than 50C. 70C is even better but then you get other "warping" like issues at the corners where they move inward but if you are desperate it's worth it. For ABS you want 110C (100C is good enough).

4) rounded corners - having square corners puts all the lifting force on a tiny spot. Rounding the corner spreads the force out more. This is optional if you use brim.

5) Brim - this is the most important of all. Turn on the brim feature in cura and do 10 passes of brim. This is awesome.

6) Squish - make sure the bottom layer is squishing onto the glass with no gaps in the brim. The first trace going down should be flat like a pancake, not rounded like string. don't run the leveling procedure if it is off, just turn the 3 screws the same amount while it is printing the skirt or brim. Counter clockwise from below gets the bed closer to the nozzle. Don't panic, take a breath, think about which way to move the glass, think about how the screw works, then twist. This may take 30 seconds but it's worth it to not rush it. You can always restart the print.

If you do all this you will then ask me "how the hell do I get my part off the glass?". Well first let it cool completely. Or even put it in the freezer. Then use a sharp putty knife under a corner and it should pop off.

[asb 7]

Hi gr5,

I have read your tips the first time you posted them, so thanks again.

As mentioned above, the more narrow the nozzle is to the printed structure, the more probable is that the nozzle tears the structure apart during the print. See the last two photos from my post above how that looks like. The surface the UM2+ prints now is not flat, it's uneven because there is clutter everywhere. When you sit near the printer, you can hear the nozzle hit these bumps, and I believe that hitting those numerous bumps is causing the print bed to become unaligned after every print.

Here are two detail shots that better illustrate what I mean:

That stuff of the nose and below the eye hits the nozzle when it passes by.

Here is another solid bump on the bottom right. I don't understand why the UM2+ does not print flat surfaces anymore. It did that just finde in the first prints.

I have tried to adjust the 2 (!) leveling screws during printing, it does not work for me. I do not even see exactly how far the nozzle is away from the print bed when the nozzle is moving. My eyes are not fast or not sharp enough for that. It only works for me when the nozzle is not moving and I can put the calibration card between nozzle and the print bed. Then I can try to fine tune if the calibration card can freely move or if it clings a bit to the nozzle (calibration with help of haptics, not visually).

I do not have 3 leveling screws. I have two on the UM2+, one on the left and one on the right of the print bed. The third point for leveling is not operated with a screw.

I'm not sure I understand what you mean with 2) and 3). My UM2+ heats the print bed before it starts printing. Do you mean the preset temperature is too high or too low?

Also I do not understand why the first set of prints before trying to use ABS were fine and after trying to use ABS everything is borked.

Please remember, I am not interested in optimized, super perfect fine art prints. I just want quick, useable results.

And no, I never had issues to get prints of the glass. As soon as they cool down to room temperature, they can be easily lifted from the glass with a cutter knife or a spatula.

Edited March 2, 2016 by Guest

[gr5 2,167]

>The third point for leveling is not operated with a screw.

Oh. Really? That's... that's horrible. How do you adjust the leveling during a print? did they add something in the tune menu? I think the tinkerMarlin lets you adjust Z after the print starts.

Adjusting leveling screws only helps the first layer. Not other layers. It allows you to have a tradeoff between a perfect bottom layer dimensionally (like a gear that is perfect) and having the part stick well. 95% of the time "sticking well" is the more desired property for the bottom layer.

Some of your photos showed parts not sticking to the glass (the parts which had more spaghetti than the mask). That's why I brought up that issue. That is fixed by my comments above about getting parts to stick better.

new topic...

The head commonly hits the part and makes a loud noise and can be the equivalent of hitting the print hard enough with your fist to move the printer a few inches -- in other words quite high forces! The fix is to get the part to stick extra well to the bed. (again raise the bed more than the leveling procedure tells you to - for example don't use any paper also use glue). This banging usually is related to overhangs like on the mask but can be related to globs of filament being dragged around and stuck down into a layer - then it hits the blob for a few layers. This is easier to fix than the overhang issue as it is usually an error in the model. Overhang issues are best fixed with more fan (for pla anyway). Anyway the banging is rarely a problem as long as the part is stuck very very well to the bed.

I think it's a coincidence when you stopped printing ABS. The mask is a different print than you printed in the past and it's not great - the eyes have an air-ring all the way around them so the center of the eyes is printed in mid-air. This is not good. Not a good cad model. Ideally you should remove the eyes and print them later? I'm not sure what the goal is to have floating eyes there. Anyway this is problematic. It makes me think there may be other problems with this mask.

Ideally you should look at your models in layer view in Cura and after a little experience you sould have seen that the eyes were gong to be a serious problem before you started the print.

[ultiarjan 1,223]

I do not have 3 leveling screws. I have two on the UM2+, one on the left and one on the right of the print bed. The third point for leveling is not operated with a screw.

 

I find this pretty hard to believe, just does not make much sense.... can you show a picture of the underside of your printbed to see what it's like? Only thin I can imagine is that it would prevent over tightening of the back screw and hitting the heated bed connector...

[asb 7]

The mask is a different print than you printed in the past and it's not great - the eyes have an air-ring all the way around them so the center of the eyes is printed in mid-air. This is not good. Not a good cad model.

As I said, it's not a mask but a InMoov body part and hundreds of people have printed it before without complaints. If UM2+ can not handle it, I just got the wrong 3D printer.

The difference between the prints before trying out ABS and after trying out ABS is that the prints before were clean, had exact bondaries and clean surfaces. All prints after are not clean, have bumpy surfaces and more or less distorted boundaries.

Currently I'm attempting to printing for the fifth time the JawV4 body part. This time I'm getting further than before because I increased the space between nozzle and print bed so that the calibration card did not touch the nozzle. The print is uneven and ugly as all other prints after trying out ABS, but maybe this time it will at least finish. In about four hours I'll know more.

When the nozzle hits residue material that isn't even supposed to be where it is, I can hardly believe that a proper fix would be to get the part to stick extra well to the bed. It's a 3D printer, not a hammer drill. The only real fix is to figure out why my UM2+ leaves this dirt everywhere. A temporary workaround is to try to cut or file the residue off during printing, e.g. with pliers or a Dremel. If I'm fast enough, the print continues. If not, the nozzle becomes misaligned, resulting in a bunch of spiderweb crap as shown in some photos above.

[asb 7]

The JawV4 body part finished, but cracked when removing the support material.

With some luck, easy parts like this finish as well:

The parts have to be flat; my UM2+ can not reliably build objects higer than 1.5-2 cm.

I made many attempts to "squish" the nozzle deeper into the printed filament, as recommended above. Maybe I do not understand correctly what I am supposed to do, but the failure rate with this approach is 100%. Here is a photo of what happens:

Starting with a print bed leveled as recommended in the documentation, the UM2+ starts to print the support structure on the base of the object. Then I reduce the distance between nozzle and object with the two screws below the print bed. A few moments later, the nozzle starts to scrap parts of the object off the print bed. The object then either sticks to the nozzle, or crinkles like paper, then cobweb-like structures occur.

Here is a detail of a crinkled object:

The earlier I adjust the two screws on the print bed, the sooner the print fails:

Here I tried to do that "squishing" thing when the support structures were printed. 100% failures and nothing to work with on this route.

I have partial success when doing this:

* Adjust the print bed more loose than the documentation suggests. The calibration card needs not to touch the nozzle.

* Start with a slightly warm print bed, but it must not be hot. When the printer was working before, turn it off and allow it to cool down for >15 minutes.

* Do not let the printer run unattended. Remove any excess material manually every 5-10 minutes. The nozzle must not hit excess material, if it does, failure is inevitable.

With this, I'm getting rather decent prints. This is a photo of a 16 hour print that I "supervised" for 4 hours. View from the front:

View from above:

Detail view:

Another detail view:

That stuff sticking on the lower left side of the print is what I mean by excess material. It doesn't hurt as long as it's on the side, but as soon as this residue is in the nozzle's way and is not manually removed, the print will fail with 100% certainty.

As long as I removed all excess material with tweezers, the object was printed cleanly. After doing this for four hours, I went to bed. Shortly after (about 2 mm of height later), the UM2+ went nuts again. Excess material builds bumps, these bumps refract the nozzle and the printer starts to spit out cobwebs.

These are InMoov skull parts which are rather complex. Cura estimates 12-18 hours print time for objects like TopBackskullV1.stl, TopskullLeftV3.stl, or TopskullRightV3.stl. I can not babysit the UM2+ for that amount of time as I need to print out about 30 obejcts as complex as this. If the UM2+ can not operate unattended, it is just not suitable for this printing task.

And here is an object I can not even remotely print on the UM2+. It's called SideHearV3.stl and is part of the inMoov jaw.

I spent a whole day trying to print this thing and made a dozend attempts. I'm not getting further than shown in the photo above.

I have no idea if this is to be expected with the UM2+ or if my device is somehow misaligned. Hoever, it's definitely not what I expect from a brand new €2300 third generation 3D printer.

Edited March 4, 2016 by Guest

[chrisw 70]

I'm at about 2k hours worth of printing on UM2s so far, spread over the last 18 months. GR5 and others here got me through some frustrating times early on. Sometimes I wanted to blame the printer and I even tried a couple other brands along the way. Eventually I realized the same issues are present for all similar styles of printers, and I just needed to spend time learning to identify the problem sources and how to correct them.

If I were experiencing your particular before/after ABS problem, the first thing I'd do: replace the nozzle. I have plenty replacements in a drawer and if an atomic pull or few is possibly not clearing the nozzle 100% then I toss the suspect nozzle.

Next thing would be use a caliper to check the size of the filament. If maybe it's 2.89 mm and the setting is for 2.85 mm, that could be creating the extra bumps because too much material could be getting extruded.

Then I'd try extruding using the Move Material function, at a few different temps 10-15C up and down from what I'd been using. It could be the temp is too low, so maybe the filament isn't flowing as well as it could, causing the extra blobs.... or maybe it's too hot, thereby making cooked blobs.

I hope some of this can help. My most common issues are the ones I look at first: clear nozzle, correct filament size, correct nozzle temp. I don't dig into deeper possibilities until these basics have been covered, because 90% (made-up stat but it seems about right) of the time the issue is one or a combo of those.

Chris

[gr5 2,167]

These are InMoov skull parts which are rather complex. Cura estimates 12-18 hours print time for objects like TopBackskullV1.stl, TopskullLeftV3.stl, or TopskullRightV3.stl. I can not babysit the UM2+ for that amount of time as I need to print out about 30 obejcts as complex as this. If the UM2+ can not operate unattended, it is just not suitable for this printing task.

So I looked at these stl files. The eyes have a gap all the way around them. That's not really printable. The Ultimaker can print millions of things but it can't print things in mid air. You will get spaghetti.

These in moov parts - many of them seem to have issues like this. The "jaw v4" also has a gap in the jaw. when it goes to print above it's going to printstraight across in the air and create more spaghetti. To 3d print these you need to do something unusual. Personally I would just remove those eyes and I would add a few pieces of support in the jaw to be removed later. But these parts as is are tricky to print on a FDM printer. On a powder plastic printer these are easy to print (but those machines start around $100k).

You shouldn't have to babysit the printer. I don't know what to tell you. I could print these in moov parts but they require some modifications.

I still don't know what's going on with some of your other parts - why are they getting spaghetti so early. I don't get this spaghetti - only if a part gets knocked loose which is maybe 1 in 100 prints. It happened to me a few weeks ago but was easily fixed by leveling a little closer. However you have some other issue I don't understand.

don't mess with the front screws once the first layer is done. Are you sure there isn't a 3rd screw hidden in the back? that's a major oversight by UM if there isn't however you seem to be getting parts to stick to the glass fine so maybe I should shut up about the leveling.

Maybe you could do a video of these "bumps" that created and then get hit by the nozzle.

This happens to me sometimes on certain very unusual parts but these inmoov parts look fine and don't have severe overhangs I don't think so I think this "bumping" you describe is something that I don't understand yet. So I think I need to see a video of what creates the bumps in the first place. You explained in words but I didn't understand it.

[asb 7]

My most common issues are the ones I look at first:  clear nozzle, correct filament size, correct nozzle temp. I don't dig into deeper possibilities until these basics have been covered […]

 

Thanks! That approach sounds very reasonable!

 

[…] replace the nozzle.  I have plenty replacements in a drawer and if an atomic pull or few is possibly not clearing the nozzle 100% then I toss the suspect nozzle.

 

Excellent idea as the UM2+ comes with a "nozzle kit" (other nozzle sizes and some small tools) which used to be called "Olsson block". Found this description, but can't get the nozzle out with the provided tools. It's stuck.

Is there a recommendation how much force I can apply? Should I try pliers?

 

Next thing would be use a caliper to check the size of the filament.  If maybe it's 2.89 mm and the setting is for 2.85 mm, that could be creating the extra bumps because too much material could be getting extruded.

 

Checked both filaments I have. The one supplied with the UM2+ is called "Ultimaker PLA Silver Metallic 2,85 mm", but actually it is 2,92 mm.

The "Form Futura EasyFil ABS 2,85 mm" actually is 2,96 mm.

If these deviations from the specs matter, how can I tell the UM2+ about it? I can not find a setting in the "Tune" menu and it's not mentioned in the PDF manual either. Oh, I hate devices without proper documentation :-(

[gr5 2,167]

In the filament settings - before you start printing - you can setup the filament diameter. You can also save it permanently if you want by doing "custom settings" set the values you care about and then "save" and then choose where to save it (PLA).

i don't think this has anything to do with your issues though.

Be careful removing the nozzle - brass is relatively soft. Make sure you heat the nozzle to 150C before removing it as PLA can be in the threads. You need about 2kg force at 3cm distance to unscrew the nozzle - more force than your fingers can produce alone but a very small amount of force if you are using a wrench.

I don't think you have a problem with underextrusion as it looks like you have plenty of extrusion! lol! So I think your nozzle is currently clean and unclogged but it doesn't hurt to clean it out. Once removed you can heat it in boiling water and then pull out the PLA with a toothpick or other tool. You can also burn out all the plastic if you want but be careful not to melt the brass.

[chrisw 70]

gr5 is probably right about your nozzle being clear already in this case, since so much filament is going through in your pictures. I'm just accustomed to swapping those out at the drop of a hat.

Whenever you do decide you need to remove a nozzle, here's the UM guide:

https://ultimaker.com/en/manuals/17869-swapping-nozzles

"Tune" menu is there when you're busy printing. But when the machine's idle there's a different menu available: "Material". That's where all the filament settings are located. Maybe change your PLA material settings to suit the brand you're using, and then try one of the quick and simple prints, like an ultimaker bot that came on the SD card for your printer.

[asb 7]

Thanks to you both! I will try to remove the nozzle as I supect that it is somehow misaligned and causes the dirty prints. If I can not remove it, I will have to talk with the vendor and ask for advice how to resolve that issue without damaging the nozzle. Since they especially recommend this ABS filament for Ultimaker printers, they should be able to provide suggestions to make it work, without building a new case. If ABS can only printed safely with an enclosed case, then the UM2+ is not designed for ABS.

Regarding the InMoov files… for sure it might be possible to enhance them with most excellent skills in 3D design; they are Open Source, so everybody is invited to improve them and share the results. However, this project is running for a couple of years already; it's goal is to 3D print a robot on cheap hardware. Meaning, 3D printers in the $450 DIY to $1500 price range. Considering the hardware hundreds of other people around the globe have managed to print their InMoov parts, the UM2+ should be a high end machine, guaranteeing best results with least effort. InMoov is definitely not designed for industrial 3D printers. So if something doesn't work, I'd blame the STL files last.

My apologies for this post; currently I'm just extremely frustrated and I'd rather to not even have to look at this device anymore. The reason to buy an expensive "plug & play" printer was the assumption not having to bother with stuff I'm absolutely not interested in (how a 3D printer works, how to make buggy software work, how to work around incorrect specs…). I just wanted to get the parts done as quickly and smoothly as possible, it was supposed to be recreational fun, not tiresome debugging work. Obviously I was very wrong, and I'd seriously recommend the UM company to reconsider their misleading advertising. They need to communicate clearly that it takes weeks if not months to get the UM2+ operational, and they have to seriously enhance on the documentation. And it's just bad if they ship a filament that does not match their own profile for this filament.

[gr5 2,167]

Um - if ultimaker says it's "plug & play" then I agree - false advertising. But every printer out there - including the million dollar printers - need a lot of training. Like a lathe or a milling machine. This is nothing like a 2D printer. Hopefully some day someone like HP will get into the market and eventually they will have plug& play printers.

I think anyone with a $450 printer will have EVEN MORE trouble with those in moov parts. I can't beleive people are having no trouble with the eyes - they probably just ripped out the eyes and spaghetti afterwards and threw them away. There may have been 100s of downloads but not everyone was fine.

The cheaper printers tend to have "support everywhere" checked off. That's a cura feature also. You could try that. Personally I avoid support generated by any slicer including Cura but people seem to like it. Examine the part in slice view first to see if it makes any difference in the "eye gaps". I prefer to put the support right in the cad model and anything I publish has the support built in. So right there I know the in moov designers are not designing super friendly and easy to print parts. A well designed CAD support will work beautifully and snap off with no tools or simple tools such as needle nose pliers.

Again - don't just blindly check "support everywhere". Look at it in slice view to see what it's going to do. And there are quite a few tweaks you can make to the support settings. Some models apparently are so different than anything ever imagined before that Cura may not create supports at all. maybe. You won't know unless you check slice view.

[gr5 2,167]

Regarding ABS - putting a front door on the printer violates a patent I believe. Hopefully that patent expires soon.

You can print ABS just fine without any enclosres. But the experts who want extra good ABS quality prints always enclose. Particularly if you want the parts as strong as PLA parts. I've printed many ABS parts without any enclosure and they come out great on my UM2 printers.

[asb 7]

Be careful removing the nozzle - brass is relatively soft.  Make sure you heat the nozzle to 150C before removing it as PLA can be in the threads.

Yay! Got the nozzle out. Heating it to 150°C is the key to get it loose. That's how it looks like:

Boiling in water doesn't dissolve the PLA. Solvents like Isopropylen or Acetone don't either. Took a lighter and a wooden toothpick, scratched the heated PLA from the nozzle's thread, and pulled remaining PLA out of the nozzle with a plier. Here is the "cleaned" nozzle and the PLA fragment:

The PLA fragment looks similar to what I got with the "atomic method".

Screwed the nozzle back in, leveled the build plate once more, now it prints…

Edited March 7, 2016 by Guest

[neotko 1,417]

The nozzle looks 'bend'. Did you use the anders olsson helper to avoid over tighten it?

https://www.youmagine.com/designs/nozzle-torque-wrench

Edited March 7, 2016 by Guest

[asb 7]

Did you use the anders olsson helper to avoid over tighten it?

No, I did not use an "anders olsson helper" and I do not even know what that is. It can not be important as my UM2+ did not come with it and not even the manual mentions it.

By the way, the whole point of this thread is that I can not print any complex object with the UM2+. So even if the manual would have mentioned that thing, I could not get it printed.

Back to the topic.

As expected, the print failed, this time after about an hour. No other parameters were changed.

But there are a couple of noticeable differences: 1) The print managed to include ~1/5 of the object. Before, it was ~1/10 or less. 2) Also, the sides are much cleaner and the UM2+ is no longer "incontinent".

In Cura I don't see any difference between "support material everywhere" and "support manterial only on build plate" (this was selected for the above print).

This is the object SideHearV2.stl in cura. Next time I generate new Gcodes, I will check if it makes any difference for the print.

[neotko 1,417]

Well. If the nozzle it's bend it won't work well. And I don't know what the manual says or doesn't but if you print the helper (when tou can) you can avoid damaging future nozzles.