gr5

pro tip: If you override any settings - where it shows the profile drop down (for example "fast - 0.2mm") there is a star if you changed anything. Click that star and click on extruder 1 and extruder 2 tabs and anything italicized in the middle column is your overrides.

There is a feature called "draft shield". Maybe you turned that on by mistake? Always look at your print in layer view and scroll through all the layers to see if anything weird like this is going on. Actually I'm thinking those are maybe travel moves? Did you have retraction turned on? maybe you were printing it too fast - I recommend you use the recommended settings or *thinner* layers but not thicker layers until you have printed 50 to 100 prints as thinner layers and other changes in settings can cause massive overextrusion like this on travel moves. But still I'm thinking maybe retraction was turned off. EVERY time you do a print, please save it as a "project". In file menu: "save project...". That way you can post that and we can look at it. It will contain the model, where you positioned the model, the printer you are using, the scaling of the model/models, and all the settings.

You have 2 issues. You have "play" aka "backlash" and you have some underextrusion. Only maybe 30% underextruded. I don't know that printer well but the most likely issue of the underextrusion is you are printing too fast for that temperature. You get better quality at lower temperature so I recommend you cut the print speed in half. Be patient and you will be rewarded. If you don't care much about quality you can instead raise the nozzle temp to 230C. I'm confident you have backlash because you don't have symmetrical separation of walls (right side in photo is clearly worse). The second issue, play/backlash can have many causes. Most common is loose belts. If you have rubber belts, pluck them and the pitch should be typically 100Hz. You can use a guitar tuner to check. anywhere 75hz to 200 hz should be fine. It depends on how long the belts are for the pitch also. But high friction can also cause backlash or other things. Basically with power off push on the nozzle gently and see how much the nozzle moves before the stepper moves. That amount is backlash for that axis. Or push on the bed in the Y direction (it looks like your bed moves - not sure as I don't know this printer) and if it moves at all without the stepper moving - that is the amount of backlash you have in Y direction. If you printed this in the orientation in the photo then it's probably the X axis with the backlash problem (the axis that moves left to right over that print in the orientation in the photo). Oh - and don't try to fix this by increasing flow to 130% - that will just make the pressure in the head even higher and likely cause a complete failure. I always print flow=100% but it's safe to go up to 110% on most printers if you are really pushing the printer a little beyond it's limits (the colder the filament the more viscous it gets and the less you can cram through that tiny nozzle).

Short answer: I usually set "initial horizontal expansion" to around -0.25 to fix this. The most common name for this is "elephants foot" or I sometimes call it "micro brim". The problem is that "elephants foot" refers to another phenomenon as well. If you level such that you have no microbrim then your part won't stick as well. This is fine for small parts (less than 3cm in X and Y) but can be a problem for parts >= 10cm in either x or y. And having a part come loose can really cause a major headache and destroy the print head of the UM3 sometimes (that's called a "head flood"). So you can play it safe (recommended) and level the way you are doing now and use "initial horizontal expansion" to correct. Or if you aren't afraid of ruining your print head when the part comes loose you can just level higher up. I never do autolevel and when it's printing the brim or skirt I just adjust on the fly. I turn the 3 leveling screws exactly equal amounts until the brim is .4mm wide (if I want no microbrim) or nicely squished (if I want the part to stick really well).

@kmanstudios - I think he bought an AA 0.25. So he should have 2X AA 0.4, 1X BB 0.4. 1X AA 0.25

I never heard of "layerceramic filament". Do you mean laybrick? I tried googling but didn't find anything (was lazy - only 10 seconds searching). Tell us more about this filament. Did you try the existing feeder with this filament? Was there a problem? If you fight the feeder when the filament is only half way down the tube - how many kg of force does it take before the filament slips? Normally with PLA it's about 4-7kg. Is layerceramic extra flexible or does it have a hard outer surface? Anyway the DDG feeder for UM3 works nicely. You can buy the DDG for UM2 if you only want to do feeder 1. The UM2+ feeder and the left feeder on the UM3 are identical. Oh wait - maybe not. Because the instructions for DDG for UM3 have you change the steps/mm setting for both feeders - I can't see any way to change only one feeder steps/mm without also changing the other.

Keep the tension in the middle. If you examine the filament it should have 2 or 3 rows of diamond holes in it at this tension. I suspect your problem is primarily at the hot end and not at the feeder. That 3rd fan is definitely the first thing to check. Many things can cause the issue you are seeing. If you want to test the feeder all alone to eliminate it then insert filament half way down the bowden, move the filament so the stepper is engaged, then pull very hard down on the filament below where it enters the feeder. Try to gauge how hard you are pulling. The feeder should be able to handle 10 to 15 pounds of pulling force. If it slips with only 6 pounds of pulling force then it's the feeder. If it can pull with 10 pounds force then it's not the feeder. Send us updates (was it the 3rd fan, did the feeder test pass) and I'll send you a list of things that can cause what you are seeing.

netfabb free repair service is here - you have to create an account but that just takes a minute: https://service.netfabb.com/login.php Is it a sketchup model? Did you create the model? If yes to both there is a great guideline on using sketchup with 3d models.

The infill tends to be printed too fast and comes out irregular. Also the vertical harmonic resonance of the bed is around 10 or 20Hz and this can get amplified on moves over the infill pattern. This is fixed if you slow down the infill moves to 35mm/sec or change the infill density or move the print towards the back where the bed doesn't vibrate as much. If you want to print fast I recommend thicker layers and wider nozzles over higher print speeds. Maybe try the variable infill feature - I love this feature. And the reason it tends to get worse on taller prints is I think this harmonic resonance builds upon itself because if the bed is bouncing the infill pattern is varying in Z by a sine wave and each layer amplifies the problem more and more.

Did you select ABS as a material in Cura? It should set the bed hopefully to 110C. I find that 100C is too cold for ABS. Or do you mean the goal temp is 100C but the bed only reaches 80C? In this second case I recommend covering the front and top of your printer. I mean I don't know how to print quality abs more than 2mm above the glass without heating the air inside to at least 35C and to do that you need to cover front and top. If you raise the air by 15C you should be able to heat the glass another 15C hotter as well. I'm very lazy so I just took a 1 gallon plastic food bag and taped it to the front of the machine and took one of those cardboard boxes that holds photocopier paper and it's the perfect size to place on top of the printer (no tape required - no modifying required). The nice thing about those boxes is it leaves a gap in the rear for the bowden to come up in through and the amount of air that can leak out is small enough to raise air temp to 35C if room temp is 20C. Make sure fan is off or no more than 3% for ABS. If you have fan at higher value the print will look great and feel great until you break it and realize it's very weak along layer lines. You can't print on plain glass. You need ABS juice (google it) or thin layer of pva. For example put down some glue stick (which is pva) and then take a wet tissue and remove most of the glue and spread it around, when it dries it is almost invisible. Prefect for ABS. Also make sure you use the brim feature in cura. Why ABS? It's a pretty difficult material. It's not really any stronger than PLA (it is slightly tougher but UM TPLA is just as good). And if you need higher temp material maybe CPE/PET/nGen will work for you (work fine up to arond 80C).

@Ghene has a UM3. Well you need two different STL files - one for each color. You load both of them and then right click one of them and assign one to the second extruder then select both of them and merge them together.

Well don't let the PVA stay hot in the nozzle for more than 10 minutes or it tends to bake itself into a semi permanent gunk. You need to do repeated hot and cold pulls (do it from the UM3 menu) to get that core cleaned out again. Not sure what the original cause of your grinding would be but if the filament is too dry it will be brittle (break off the end at the end of the spool to test it) and if it gets too wet then it boils and pops and snaps and I could imagine maybe it can't keep up with demand (too much pressure in nozzle) and would grind to dust back in the feeder. Try to keep the filament in the bag with the dessicant whenever you aren't using PVA or Nylon. other filaments aren't so sensitive to humidity.

Oh wait - you didn't say you had a UM3. You would need to at least buy a hardcore so you have a place to put the nozzle. So, if you can't afford a hardcore right now then... nevermind. But if you were thinking of buying an AA 0.8 and haven't bought one yet, then get the 3dsolex hardcore instead!

You can possibly get a custom nozzle for free - 3dsolex is always creating new products and you can be the beta tester. Tell him you want a 1.5mm or 2mm nozzle or whatever it is you want.

Look straight down from above the printer - do the two rods going through the head look perpendicular? Push the head around with your hand to the center of each of the 4 sides. Do the end blocks seem to stop at the same time or is one block offset?

When you get up to 1mm and larger you also want the 3dsolex "race" technology which means the filament flows through two paths inside the nozzle to heat the filament more thoroughly. 3dsolex has a 1mm race nozzle and a 1.2mm sapphire nozzle (which is expensive). I was going to say you could drill it out to 2mm if you wanted but that would be difficult because it would be too easy to remove the internal structure inside the nozzle. I guess if you had a drill press and went in from the tip and went slowly. Maybe you could ask 3dsolex to make you a 1.5mm or 2mm custom nozzle. The guy there, carl, is busy but loves to come out with new products (much to my dismay).

There's tons of causes of grinding of filament back at the feeder but this one - filament grows to 3mm - is not something I've heard of much before. Are you sure that's the issue? how do you know? when it happens is the filament stuck in the core? Can it slide up and out? This growth in filament size may be unrelated is what I'm trying to say. There are about 15 causes of failed extrusion on my list but not this cause as I've only seen it be a problem when the print is over. Maybe we should look into other issues? maybe I should give you the complete list?

That's almost always caused by a loose pulley. Each axis has 6 pulleys with belts around them and connected to steel rods. There is a little set screw to connect the pulley to the rod. That needs to be tightened. A lot. So much that the hex tool is twisting. 90% of the time that is what causes these steps. Figure out which axis is slipping and then tighten the 6 pulleys for that axis. It's almost always the pulley on the stepper motor which is the hardest to get to. If you need to remove the cover over the stepper it's held on by only 2 screws.

My first suspicion, the power brick, is now my latest suspicion. It will probably get gradually worse and power cycling it may help. Or it may be a fluke where the cpu in that power brick got confused about what voltage it was supposed to be putting out and power cycling the brick helped.

If you take that black sliding block apart you will almost certainly never get it back together. Basically taking it apart breaks it. But you can buy new ones from your reseller if you contact them (probably not in their store but they can get them).

Well PLA gets quite soft (think of clay) around 55C so I would avoid that anywhere near boiling water or steam. I don't think it shrunk necessarily - it may just be moving slowly towards a sphere in shape. Or a saucer. I don't know much about plastics and food. ABS seems like it wouldn't be good near food (I could be wrong) but it softens at 99C. Nylon probably wouldn't be good because the high humidity would likely turn it into a very floppy object. Or maybe it would be fine because most of the time humidity would be low. PET/CPE/nGen I think might be mostly the same thing? I think they soften around 80C which is significantly better than PLA and almost as easy to print. Things that soften at 100C or higher are going to be hard to print (you'll want to enclose the printer and make sure fan is at 0% to get good layer bonding). Polycarbonate might work. ABS will definitely work. here is some info on materials. Unfortunately the "softening temp" is wrong on many of these because, well, measuring this with a standard measure is going to give you the wrong answer. But it wil give you some ideas of where to start - you want the second graph which shows temperature related stuff: http://gr5.org/mat/

I would contact either your reseller, or fbrc8 which covers USA. If it's a large reseller like matterHackers they are pretty good but fbrc8 is even more knowledgable. fbrc8 made your printer and they are experts. If you are at all worried about voiding warranties, you simply need to get permission to remove covers before doing so. But they won't know if you removed the cover of the larger board without you telling them. Good point about Z and E axes - but the X and Y axes may be set up to run closer to their limits of acceleration and velocity and may be more likely to miss steps (that's what this is - missing steps - the steps don't have enough power to move the X and Y axes). Oh! that gives me an idea. Maybe you have high friction (due several possible things including no oil). With power off you should be able to push the head with just one finger on each end of one of the rods that goes through the head. Try adding some light oil - sewing machine oil or 3-in-one oil or any light mineral oil (not vegetable oil which will go rancid, not wd-40, baby oil will work). Put very little oil on - one drop on each of the 6 rods in the gantry. Push the head around to spread the oil. High friction can be caused by too much belt tension (unlikely but I've seen it) also.

Lots of great answers for someone who may have never come back here to read the answers ? As a mod I can see @Brunella never logged back in since the original post. But that doesn't mean brunella didn't read the answers without logging in first.

I really don't know. I don't have enough experience to form an opinion. I've seen this exact problem on UMO and UM2 but I don't know the plus feeder as well. I've only taken it apart 2 times. Most of the stuff in there isn't very adjustable. I did have a feeder where one of the parts (gear) kept sliding off the motor. fbrc8 sent me a replacement. I think they sent the entire stepper motor and gear assembly. They service USA customers. If you do get a new feeder, look at the sharpness of those little pyramids on the gnurled wheel and compare the new versus the old. Maybe post a picture if they are different. FYI - the UM2 plus feeder is identical to the UM3 feeder. I have both printers.

If you look at the first post in the above link, this is for the old black UM2 feeder (not the plus feeder) which notoriously was on the weak end of things and even the worst of the 3 feeders tested there was at least 4.5kg or 10 pounds. I have to tell you I am very happy that we have narrowed this down to your feeder. Please repeat that 7 pound experiment now that the feeder has been rebuilt or look at how the experiment was done in the above link to get more accurate results. I urge you to get a new feeder. The amount of work to figure this out versus just getting a new one... Especially if your business is paying your wage versus the cost of a new feeder. I think at this point you should just get a new feeder.