dgsharp
-
Posts
47 -
Joined
-
Last visited
Never
Content Type
Forums
Events
3D Prints
Posts posted by dgsharp
-
-
A more stable Z stage that doesn't bounce around so much. The current one is cantilevered out way too far, reducing print quality and compromising bed leveling.
Auto bed leveling. Much less necessary with a beefier bed but still very handy.
Feeder that doesn't jam so much. I designed the Hive76 Milled Bolt, a pinch-wheel with very positive grip. Once there was a filament jam with one of the printers with the filament spool hanging from the ceiling and the machine climbed right off the table. I'm sure there are lots of approaches that will result in a less finicky feeder than the UM2.
All metal hot end? I think it's clear that Teflon makes for fairly disposable isolators.
Dual material. I don't care about color, I can paint it if I wanted that, I need real support material so I can print functional parts.
I love my UM2, but there's lots of room for improvement!
-Dave
-
I did one or two test prints at 20um early on and they turned out ugly, fuzzy, etc. I didn't think it was worth trying to improve it because trying to get that sort of resolution out of this type of machine is basically a waste of time: it takes forever and just isn't worth it for anything I've seen.
-
As 3DMaker4U suggests, it all depends, especially based on the mechanical properties you need. That being said...
In my experience typically PLA is capable of capturing finer details. It also prints with a more glossy finish, whereas ABS prints with more of a matte finish that can sort of blur fine details out a bit. The glossiness of PLA can also bring out more subtle flaws and artifacts though, like 'ringing' from acceleration.
-
I found the same thing with NinjaFlex (I had to move the knurled pinchwheel onto the shaft further). Even with the snap-on part of Robert's feeder the filament still squishes through sometimes, so I have to watch the first 2 layers or so. If it fails I check things over, maybe add a touch of oil to the bowden (not very frequently), just check everything over. Cut off that section of filament if it gets really permanently kinked up at the feeder. Maybe pull the bowden out of the hotend and do a cold pull with PLA to clean out anything that might be getting held up in there. Usually I have to do this sort of thing a couple of times before it starts printing, and then it's printing fine as long as it's running (12 hours or more), but when I go to start up another print, I might have to do the whole thing again. Kind of a drag. I haven't totally figured out the tension on the Robert's feeder, I will sometimes try loosening or tightening it some and eventually it starts printing. I'm not super happy with the feeder though. I often use 30 mm/s, 0.15mm layer height, 20-50% infill, 2 shells, 225C head / 50C bed, and 120-140% extrusion. I wish there were fewer knobs to tweak!
-
Dirkels, one that frequently bugs me is that if I have a model that takes a long time to slice, and then I decide to switch the settings for Support Type or Platform Adhesion, I am forced to wait until the slicing is done before Cura will allow me to change settings. If I try and change these settings before it's done slicing, the dropdown combo box will not stay visible long enough for me to do anything. That gets pretty old. I often have to resort to something like scaling the object down to be very tiny (so it slices quickly), changing the settings, and then scaling back up.
-Dave
-
ALL RIGHT!!! Ok, I am tentatively saying this is solved.
I installed a new bowden tube, new heater block and nozzle (the old ones had totally seized up solid and broke while trying to disassemble it to clean it), and straightened out the filament. I was able to print one of Illuminarti's cylinders completely. I had my finger dragging slightly on the end of the knurled wheel so that I would immediately feel any pops of under-extrusion, and there wasn't a single one. Beautiful! Now if I can just get it to stay that way!
Thanks once again Illuminarti!
-Dave
-
I see. I've been using the latest firmware, and 20mm sounds about like what it's doing. Until I replaced my teflon piece today, 20mm retraction would be a "huge amount" and result in an immediate jam if allowed to cool that way. Now I realize that my teflon part was almost certainly the culprit of that particular symptom, allowing the large bump to form internally which then can't be effectively drawn back through the bowden or pushed through the extruder.
The material I'm printing is silver PLA, slightly under-sized and in the past it has printed very well for me. I am down past the halfway mark on my spool so it does have a bit of a tighter radius than it did to begin with. I did have a nasty issue with some black filament a while back and had to straighten it out, which helped some, but I think I had a couple of other issues going on as well. It doesn't take a lot of force to push the filament in or out of my bowden tube, I did that a few times this morning during the teardown and it's often how I load material. I will try straightening the filament on the next print. I'm currently about 10 hours in, and there was a little bit of under-extrusion several hours into it in one area, nothing like before but still an issue.
After this print I'll try swapping out the bowden, straightening the filament, and cobbling together some way of more easily feeding off the spool. Thanks once again for all your input.
-
Another status update: about 2-3 hours into my print I started getting really nasty under-extrusion again (speed 50mm/s, height 0.1mm, temp 230C). I slowed it down to 50% (25 mm/s) and let it run for several layers and the under-extrusion didn't stop (feeder stepper kept skipping steps). I did notice that it seemed to lose steps around one particular area that had several retractions, so I lowered my retraction speed from 35 to 25 and distance from 5.5 to 4.5. I then bumped the speed up to 75% (37.5mm/s) and let it go a few layers, and it still seemed good.
I had to abort the print so I figured that was a good opportunity to try Illuminarti's cylinder test. The feeder skipped 4 different times during the bottom 3mm/s band. Looks like it's back to the drawing board.
So let's see what's left... I can tear it down again, take the nozzle off and thoroughly clean it out with acetone (ugh, I HATE taking the nozzle off). Replace the bowden tube. Anything else? I'm running out of ideas. I probably ought to print out something to help make pulling filament take less force, though the under-extrusion was still happening despite me manually providing excess filament.
Can we get Cura changed so that when you hit "Abort" it doesn't retract the filament a huge amount and then forget to ever re-feed it? Every time I need to abort a print (which is frequently these days) I have to either manually "Move Material" afterwards to undo what it did, or (what I often do as it is much faster) I just power the machine off mid-print and lower the build plate manually.
-Dave
-
Finally an update: I've received my replacement teflon part, and now that I have the two side-by-side I can see just how bad my teflon part had become. I'm attaching a picture showing them both holding a short section of identical silver PLA:
Despite the poor quality of my picture, and beyond just the obvious blackened tip of the old part, there is a very striking difference in how closely the parts fit around the filament. The new part actually holds onto the filament slightly here, due only to the slight bow in the filament. The old part is extremely loose, with lots of clearance between it and the filament (it's even more obvious in person). I think this greatly enlarged insulated cavity may be where my filament had been piling up, jamming my hot-end.
At my last hackerspace they had an old Stratasys machine they were trying to salvage, but the equivalent part (called the inlet buffer) was made out of polyimide and one of them was broken. It was ridiculously expensive stuff (McMaster has 1/2" round stock for $54.51 per linear inch!) but we were able to find some small blocks of it on eBay for cheap and I was able to make a few copies using my lathe and CNC mill (which of course I no longer have..). The polyimide was harder and more brittle than teflon but it machined pretty well. Teflon was kicked around as an alternative but it was deemed that the temperature ratings weren't high enough, at least for the Stratasys (IIRC their head temp was more like 270 or 275C?). Wikipedia is saying that the pyrolysis of teflon is detectable at 200C, suggesting that eventually all of our teflon parts will probably need to be replaced. Given this it seems like investigating alternative materials would be a good idea.
Currently I'm printing with the new teflon part at 50mm/s and 0.1mm layer thickness. I think I may have heard a few occasional pops of under-extrusion but I wasn't paying close enough attention to say for sure that they weren't just the bed dropping for the next layer (they're very different sounds but sometimes if I'm not paying attention I can't swear whether I heard one or the other). I forgot to do Illuminarti's cylinder test before starting a long print, silly me. This is definitely an improvement over before though, as I had it down to about 20mm/s and it was STILL under-extruding badly at 0.1mm layer height.
I'll post an update if I have any issues or notice anything interesting. I do wonder what I must have been doing to abuse this part so badly compared to others. All I can think of is using ABS and nylon.
-Dave
-
Thanks again for all the thoughts and PM.
I took off the heated build plate and poked around and saw no signs of blackened wires or loose connections anywhere. It did look like at least one of the screws holding the black rubber piece that attaches the cable to the build plate was loose though. My current theory is that because that was a bit loose it allowed the solder joints on the back of the build plate to cycle and wiggle just enough that one of the solder joints failed. I opened it all up and the surface-mount connector came off, quite possibly due to my handling. One of the solder joints looked different than the others (more shiny and a darker brown color) which may or may not be important. I soldered the connector back down and reassembled and it seems to be working again.
I removed the Bowden tube from the hot end and did my best to make sure it was seated firmly against the teflon (with thumb screws already tightened). I then performed the "HAS-CAP" procedure a couple of times and got a very similar "shoulder" / bump on the filament. I trimmed it and put it back together and used Move Material to do the same thing with the Bowden tube bottomed firmly before adding the clip, and the results seem the same as before. It's almost as though my teflon piece is too short now or something.
I am almost tempted to apply a series of light but recognizable scratches to the end of my Bowden tube and inside the teflon part, so I can look directly on the filament after I pull it out, and the culprit will be the piece that made its imprint on the filament. I'm supposed to have a replacement Bowden and teflon piece coming so hopefully in a few days I can get it fixed either way. Anyway, fun for another night.
-
Thanks for your thoughts, I will investigate them all as soon as I get a chance. As for the heated bed, it gives no error, it just never prints, or the advanced option to heat up the build platform isn't able to affect the temperature. No wires seem to be loose but I need to fully tear it apart to get a better look at things.
My Bowden tube doesn't move at all, but I will be careful to try the directions you linked me to.
Thanks again, will report back when I can.
-
Well I put everything back together and left a print running at 210C. I came back after 9 hours or so (24h print) and halfway through one of the parts had popped off the build platform and it was printing air spaghetti. But spaghetti is better than caramelized PLA. I went to try re-printing it using the new raft option and.... my heated bed never got hot. (Presumably the heated bed died mid-print, causing one of the parts to let go of the glass.) So it seems like I have another issue in addition. Will update as I get a chance to uncover things. I haven't noticed any loose or discolored wires anywhere yet, but need to tear things apart more to look closer.
(I went to that web page and the ad was for an Ultimaker2. How it mocks me!)
-Dave
-
I've been having more under-extrusion problems (UM2). The highest I've ever gotten on Illuminarti's cylinder test since trying it the first time was about 7, and I tried the other day and it was popping multiple times per loop, utter failure.
Lately there has been some kind of gap somewhere causing filament to pile up, so that when I go to change material or when I do the "HAS-CAP" hot-end cleanout procedure it leaves a big nasty bump.
I tore down the hot-end (except the extruder) and it's puzzling. I do the "HAS-CAP" procedure and I can see daylight through the nice round hole, but it's giving a ton of resistance, even at 230C. I forced some PLA through with it all taken apart and was able to see it push the teflon piece up in the air, filling the new void below the teflon part with a puddle of PLA resembling the shoulder on the PLA ends in the picture. So it seems to me like what I'm getting is a combination of unusually high pressure in the hot end, and a teflon piece that is moving up under that pressure, causing further problems. I have a very strange lip/groove on the inside of my teflon part (which I've seen in one or two pictures on here before, attaching pic), and they're sending me a new one, but I'm not convinced that's my only problem. This filament is slightly under-sized, and has printed very well for me in the past.
Is it possible that the pressure is so high in my hot end just because it's coated in a layer of built-up burned-on PLA / ABS / nylon residue that is acting as a thermal insulator? And if so, is it expected that the teflon piece would push upwards and create a void underneath it, or should it be held in place more firmly somehow? I've got clips on the bowden tube and there is no play at all there, but I've been careful not to tighten the knurled thumscrews on top too tightly as I've read about here.. I'm sure people will say that printing multiple materials should be avoided, and maybe that's true..
Thanks for any suggestions! I'm dying here. I just tore it down again and put it back together, dropping the temperature back to 210C from my usual 230C to maybe prevent it from burning in there. Layer height is 0.1mm, speed is about 30mm/s, but at least it hasn't seized up yet this time. That's one thing about ABS, even when my best PLA is barely able to extrude at all, the ABS just comes pouring out..
-Dave
-
It could be something in the model itself. I've gotten wonky print files from people before and it turns out that the surface normals are goofy or there are unintentional internal faces and such. If you can get it into Blender or a similar program and really poke around to make sure everything is as you suspect, you may find your issue there. Usually when I have a strange disagreement like this between what I expect to print and what Cura produces, it comes down to the model. Might be worth checking if the settings don't fix it.
If you'd like to upload the file, I'm sure someone here would be glad to poke around at it to help you out.
-
I was going to say you need to change the gcode flavor to reprap/marlin. I take it you did that as well? Or is that no longer a requirement? (That would be nice.)
-
I disagree with the "any 3D supplier" comment. I got 3 spools of PLA and 1 spool of ABS from Makershed. They sold me my UM2 so I figured it'd be good. Two of the spools of PLA and the spool of ABS measure over 3mm, which you will commonly find listed as the absolute max recommended filament diameter. The ABS prints great regardless (it's a much softer material than PLA). The two oversized spools of PLA are very difficult to print, I have to print them at 20mm/s to keep from jamming up my machine with any consistency. Extremely frustrating to wake up after an all-night print at 25mm/s and the entire build plate is covered in severely, comically under-extruded fluff. It gets worse the closer to the center of the spool I get.
I got 2 spools of PLA from ProtoParadigm, and they are great (around 2.8mm or so), and they list their tolerances (unlike Makershed). UltiMachine also lists their tolerances so I expect theirs is pretty good as well, and will be trying them soon. I've heard many good things about Printbl but the fact that the blog and Twitter were last updated a year ago and many of their items are listed as out of stock makes me wonder how reliable that will be in the future.
-
I believe I'm using the latest stable UM2 firmware and pause (in the Tune menu) works just fine for me. I used it a few times today while printing some ABS -- there were some steep overhangs that left some very ugly areas curling up sharply, and I just paused after that layer, carefully trimmed the curled areas down with a scalpel, and resumed the print.
I recall that Pause At Z wasn't working on the UM2 before (something about the relevant G-code not being properly implemented on the UM2). Anybody know if that works now? Manual pause is very handy though, still.
-
I had one spool that was giving me issues after only about half the spool was done. I tried straightening it manually by bending it back on itself as I pulled it through some paper in my hand -- it worked but it was annoying. So the next time, I took a big pot and set it on a cookie sheet, and then uncoiled the rest of the spool onto the outside of the pot. It was coiled tightly enough that it held itself on there tightly all by itself. So I popped it in my (electric) oven at the lowest setting (I think like 170F? that would be like 77C) and left it for 20 minutes or so before pulling it out. The cookie sheet keeps it from falling down onto the rack, but more importantly shields it from the direct radiant heat from the heating elements below that would burn it. The idea being that we "anneal" it so the stress is gone, and we give it a new preferred bend radius.
What I did next I doubt I would do again, but I wanted to try it -- while the filament (still wrapped around the pot) was still hot I unwound it and laid it all around my house, straightening it out. After a few minutes it had cooled so I was able to respool it back on the original spool. The problem there is that until it eventually regains a slight curve from storage, it actually wants to unwind itself automatically when you put in on the printer! I had to add a bit of drag just to keep it from ejecting all over. Next time I would probably just let it cool on the pot before respooling.
I wanted to do the ball bearing approach but I didn't have a pile of 608s like everybody else on here seems to. I would definitely do the pot/oven approach again though.
-
It appears that cool head lift doesn't do an un-retraction after the last layer of a part. This appears to be the case whether there is just one part in the print, or if it's a multi-part print. This results in the next part having nearly non-existent extrusion until it has extruded enough filament to make up for the lack of un-retraction. I have had to use cool head lift to prevent organic shapes with fairly flat but curved tops from getting all gooey and melty.
On a side note, why not do a cool head shift rather than lift? Whenever the extruder does a lift for me I get a long string or thread going up in the air (including at the end of each and every part). Then the head comes back to do the next layer and pushes that first string around, and then adds another one when it lifts again. This leaves a bit of a mess that doesn't clean up so well. By contrast, when the head leaves the side of a print it doesn't leave a big string. There is usually a tiny bit of filament on the tip that oozes out despite the retraction, but if the head comes back to start the next layer from the same direction and location that it did the time before, these little bits of plastic glob together into a single fiber that juts out to the side away from the model, and this single fiber comes off with a brush of the fingernail.
Hope that made sense.
-
You're making me jealous Burki! I am dying to do dual extrusion. It's ultimately the only way to really do it right I think.
I did a test today using TweakAtZ to print a solid tower, alternating the material flow percentage between 100% for 1mm, to a lower number (starting at 50% near the bottom and ending at 10% near the top), resulting in a number of 1mm-thick plates connected by fluffy, severely under-extruded support (I've been calling it "CUE support", for "Comically Under-Extruded" support).
I found that after tearing away the clumps with a pair of pliers, the underside of the 30% section (the side that was built on top of 30% fluff) looked the best of all of them. It's not smooth, but it is relatively uniform (comparable to Cura's built-in support at 25%). The top side (the side with 30% fluff sitting on top of it) cleans up quite nice, better than the corresponding side of the test print I did using Cura's built-in support set at 25% (though maybe tweaking the Z offset could help here) and not that much worse than the top side of a solid print.
I don't know that this is a viable approach on its own (is this how Netfabb does its "fluff"? I've never seen it), but it could come in handy for some specific situations where the material is easily accessible for removal with tools. It would probably work better with a direct-drive extruder as the pressure in the Bowden takes a certain amount of time to dissipate, resulting in a slow ramping up or down of material flow following a setting change.
I tried the new/old "lines" support in the Cura 14.02 RC5 today and wonder why the lines aren't connected into zigzags at the ends. That's more like what the Dimension machines did IIRC and it worked pretty well, probably easier to remove all at once. I like how easily it removes but the bottom (supported) surface left something to be desired for my purposes. I will see if I can do some more experiments, maybe playing with Z offset can help as you suggest.
-
I have some parts that require supports, and I'm trying to find a reasonable solution. Normally it's fine to try and design such that you require no supports, use Cura's built-in supports, put in custom specific supports where necessary, or even chop the model up into pieces to be assembled later, etc. MeshMixer's supports can be extremely helpful, but don't do much on a fully overhanging surface. Ideally I would use a second extruder with a soluble support material (and hopefully we won't have to wait too much longer before this is an option! <nudge nudge>). But there's got to be something else to try. In particular I am making some small experimental propellers. They need to be as strong as possible, so gluing parts together isn't ideal. Some of them have a flat, pure-overhang face on the bottom of the hub that absolutely needs some kind of support, and the default supports leave a really nasty surface finish. Most of the blades can print reasonably well without support but a couple of areas inboard require support and come out with a really nasty surface finish when you use it.
I was getting ready to start some experiments to try and find a support method I like, and thought I would ask here to see if anyone has any good ideas. Let's say the design criteria is that it has to be able to support a full overhang (a horizontal surface with nothing beneath it), and the surface finish needs to be as good as possible after removing it.
Here are some ideas:
- A series of closely spaced pegs. In a simple case maybe just a single extruder width wide if possible, or to make it easier to both fabricate and cleanly tear off, they could be wider (like 0.8mm or 1.2mm diameter) and then taper down to a single extruder width point of contact or close. By varying the spacing I could try and find a good balance between nice flat overhanging faces and minimal pock marks where the supports were torn off.
- Very narrow (ideally single-extruder-width) zig-zagging vertical supports.
- Wonky idea inspired by some comically under-extruded parts I made when my UM2 was having trouble with very stiffly coiled PLA: print a relatively "solid" section but force the g-code to produce them severely under-extruded (presumably by writing some kind of plugin). This produces a sort of wispy lattice of diagonal blobs that you can see right through. Not sure how well this would work when mixed in with areas of proper extrusion (might just clump up in areas or something).
How can I guarantee that a small feature I create geometry for comes out in the g-code? I tried making features that were 0.4mm wide and they get ignored most of the time. I found that even 0.6mm didn't always make it through to the toolpath. Should they? How can I design support structures that are minimally small/thin, but which come through to the printed model? Is there a straightforward way?
Thanks for any thoughts you may have! Hopefully that second extruder will be an option in not too long.
Edit: Not mentioned was the option of just tweaking Cura's native support settings. Know of any particular settings to try?
-
I love my UM2 but definitely would not put it in the category of 'professional' machines. In my experience a professional machine just works, all the time, all of them have a second extruder for a support material, the software changes very infrequently... The results are reliable, but you pay for it. There are some things my UM2 can do that most pro machines can't, but they're just not the same thing.
-
In case anyone is curious, here's how I found the X/Y resolution:
- I created a sketch in FreeCAD showing a pie wedge with the radius of my sphere, as measured with calipers (the radius shown is in fact half the diameter I measured).
- With calipers, measured the diameters of two concentric rings. In this case I believe they were the 2nd and 3rd rings, as the first one seems to be gobbled up by the backlash in the belts (25 microns of backlash ain't bad!), but it doesn't actually matter which rings they are as long as they are adjacent.
- Added constraints to the sketch in FreeCAD showing everything I know: the diameters of the two concentric circles (shown in the sketch as the vertical distance between points along the arc). The two arcs are both symmetrical about the X axis.
- At this point the sketch is fully defined, so we can measure the horizontal distance between the points of the smaller arc and the points on the larger arc (the smaller and larger concentric rings on the 3D print). This comes to about 26.5um. If I could measure the ring diameters with a microscope it would probably come out closer to 25um (just a guess).
UPDATE:
Ok, something is fishy with my results, the horizontal distance between those two slices appears to be 265 microns in that diagram (0.265mm, the dimension at the top), I goofed a decimal place in reading it off. I re-checked the numbers in SolidWorks and they agree with FreeCAD. I checked the horizontal distance to the next pair of concentric rings and it appears to be 170 microns. This suggests that we are seeing something other than the X/Y resolution of the machine. It is much more coarse even than the layer height.
I performed this again using a different printed sphere model, which was printed at 100um layer height. Using the approach described above on the top surface I was able to measure the layer height (Z) to be about 0.11 mm (110 microns), very close to expected. This was identical for two pairs of adjacent concentric rings, as you'd expect, since we know the Z step height is consistent. I think this shows that my approach works for finding the radial/planar distance between two adjacent concentric rings that appear on a spherical surface.
I'm re-printing a test model I made at 20um layer height and 20mm/s, it's 1" diameter (25.4mm) and has about 100k faces (many many more faces per mm^2 than the model that started this thread). I have one done at 20um but it was probably printed at 35mm/s and the surface, while very nice, isn't quite as crisp as the newer part. If I see rings that are the same planar spacing on the new 20um part with gobs of faces as on the 50um part with fewer faces, then it can't be related to the model itself, and apparently has to be in the slicer after all.
I'm guessing it is something in Cura like an angular deviation threshold. It's not actually snapping the points to a consistent X/Y grid as I'd first assumed, but just creating a new point when the angle between them is greater than some threshold. I just re-checked the points in the G-code and while the angle differences between them aren't identical, they are all pretty close to 5 degrees (some as low as 2.7, some as high as 7.1). Perhaps the rationale is that having too many move commands in a tight space would make the SD card a bottleneck, requiring slower prints to be consistent? This sounds very plausible to me.
-Dave
-
How much infill did you use? Are that the points were the Infill connects?
I used 20% infill for this, but this is not related to infill. I have a different model of a partial sphere I printed with 0 infill and it exhibits the same exact thing.
Your print really looks great.
In Cura, go to layer view. You might see the concentric rings also there when choosing the right view angle. I think it's a numerical artefact from slicing as it is acutally producing a bunch of straight lines and no arc for a curved wall. But I think Cura is slicing with a resolution which fits to the UM resolution, at least I hope so.
I've confirmed that Cura's output resolution appears to be 10 microns. It uses 2 decimal places, but more importantly than that, the shortest distance between two moves along the X or Y axis is exactly 0.01 mm.
We are almost undoubtedly seeing the effective X/Y resolution of the system, and it's right around 25 microns. Cura clearly isn't the limit if it is commanding moves of as small as 10 microns. So that leaves something on the hardware / firmware side.
Here is a picture I pulled from Google Images of a sphere as it would appear in Minecraft:
Clear as day you can see the concentric rings exactly as I've described -- one set of concentric rings vertically oriented at the max X, max Y, min X, min Y, and of course also the min and max Z. And if you measure the diameter of the sphere and the diameters of two adjacent concentric circles, you can figure out how far the rings are separated along the X or Y axis -- the limit of the resolution.I don't see another way this could be interpreted. So where is this 25 micron resolution limit coming from? It's a nitpick, totally, but.. where's the other 50% of the resolution going?
Printing with PLA - no luck
in Coffee corner
Posted
I just wanted to echo this. For a long time I had a worn isolator and I spent a lot of time looking at the wrong things. ABS was the only thing I could print. When I finally realized it and got a new isolator, it was a whole new machine.