geert_2
-
Posts
2,084 -
Joined
-
Last visited
-
Days Won
34
Content Type
Forums
Events
3D Prints
Posts posted by geert_2
-
-
Tiny holes are very difficult to print to an exact size. Try pouring honey or yoghurt in an exact shape. Molten PLA is similar.
I think it would be best to post-process the model, and manually (!) drill the holes to the exact diameter. First print holes that are close enough, and then manually clean them up. I use a separate drill chuck for this: this gives a good grip and enough torque, and it allows a very slow and gentle drilling with good feeling, but without melting the material (which is a problem in electric drilling).
-
What about printing in PET, or NGEN? Would that be an option? They should give better layer bonding, and will probably give a stronger model than ABS? Or just plain PLA if it does not have to withstand temperatures above 45°C?
If tearing and cutting off the supports is too difficult, maybe you could drill them away? If you have a Dremel-style hand drill that can go slower than a Dremel, so it doesn't melt the plastic. Rounded bits as shown below, used in dental labs, work very well (their shaft is usually 2.7mm). You could even use a big drilling machine, if you clamp it to a table, and move the model around.
This short Youtube-video about making a dental retainer shows the idea and tool options quite well: https://www.youtube.com/watch?v=APRw1wgvwy8
-
I am not sure if ABS is a good choice for this: it often has poor layer bonding. PLA has better layer bonding, but may be less flexible, and it might not survive cutting away the supports? Maybe a more flexible material that can withstand some abuse might be a good choice? Such as nylon or similar? But I have no experience with these, so I can't tell for sure, just guessing.
You could try printing this in PLA without supports, if you print as cool as possible, and very slow? Maybe 20...25mm/s and 180...190°C? Just stay around and keep watching. But I think it is not going to be very smooth. Since it won't consume too much time and material, maybe the best is to just find out by trial and error?
It would be interesting if you could show us photos of the results.
-
While I also like the twisted vases, I think the butts are going to attract more attention. Especially from the kids. From the adults too of course, but they are going to do their very best to hide it. They will probably take a twisted vase in hand, act as if they carefully study it, and then glance over it towards the butts. :)
-
1
-
-
Just out of curiosity, why bright white?
Why not broken white, cream, or warm lightgrey, or so? Those softer colors also show the surface details quite well, without distraction, and they have a more natural color, closer to reality. But they don't show dirty fingerprints so much. :)
-
2 hours ago, Brunella said:
Hello
Yes you were right. For some reason I messed up in Solidworks. Anyway I Started printing but I have some problems with the build plate adesion. At half way of the printing process it messes up as it lifts form the build plate. I tried to put some glue and even increase a bit the buildplate temperature, unsuccessfully.
Suggestions?
Thanks
I would try two things:
1. Extend the flange at the bottom to the inside of the model too, so it is symmetrical around the vertical support column, and make it a bit larger like in my model. Steep overhangs tend to curl up, causing the nozzle to bang into them quite hard. This requires good bonding.
2. If printing this in PLA, try the "salt method" for bonding. See the manual at my webpage (it's a bit old, but still usefull):
https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/
Both combined should be okay.
-
30 minutes ago, Brunella said:
Hi
Thank you for your reply. Actually the support is 0.5mm thick. I don't know why it doesn't recognize it.
It appears that your suppport is way thinner than 0.5mm, maybe only 0.05mm? In my spring it is 0.5mm; and this shows up way thicker if I place both models next to each other in Cura. In my spring, the support does show up correctly in layer view. However, if I make my supports only 0.25mm thick, then it disappears just like in your model. So, could you check that?
There may be other issues with the STL-file too: it looks like the support column and spring are not merged into one single object? This might cause issues with printing in some cases. But I don't know anything about the internals of STL-files, nor about how Cura handles them, so I can't give much suggestions on that.
For comparison: support in right spring is 0.5mm thick.
-
I can load the STL in Cura (older version, don't have the latest). It shows up in normal view, but indeed the support is lost in layer view.
It seems that the support is too thin? It should be at least 0.4mm wide for a 0.4mm nozzle. But due to the triangular segments of an STL-file, some parts might become 4.1mm, and some other 3.9mm. And these last might drop out. So I always use 0.5mm width for such things, to have a bit spare.
-
6 hours ago, fbrc8-erin said:
The equivalent fan on your UM3 is the front fan. It should keep the upper part of the print core from getting too hot. If filament hairs get tangled in it, it could stop running.
Ah okay. I don't have an UM3, so I didn't know that fan has changed location. In my UM2 I once found that it had sucked up filament hairs: it looked like a coloured PLA spider web, although not enough to stop the fan yet. The bearings of these little fans also tend to wear out fast, much faster than bearings of big fans. First they get noisy and slow down, they wobble visibly, and then they totally stop due to too much friction in the damaged bearings. This was a very common problem in early Pentium computers (1995...2000). This could also happen if dust from clothes accumulated in the fan bearings, or inbetween the blades. So it is something I still look out for, if overheating might be one of the causes of a problem (but of course, there could also be other causes).
-
Personally - but this is just my opinion - I don't think it will contaminate your oven and food if you let it cool down with open door, and wipe the oven afterwards. Cleaning aids you use to clean the oven, will contain worse chemicals. And the smoke and coal produced by baking meals will be a bigger health risk. The main risk of kitchen ovens is their poor temperature control: it may easily get too hot, or overshoot the set temp, and destroy the filament.
So, using the UM glass bed as oven, or a dedicated filament dry box, or an electronically controlled laboratory oven (commonly available in schools, universities, labs), is better.
-
1
-
-
I would say: buy a few colors that you like in PLA, one spool each. Further, maybe buy one spool of NGEN or PET too. And experiment with these until you are used to it. Don't buy huge amounts. Some colors of the same brand and base-material (polymer) do print better than other colors. Not sure why, maybe due to additives such as the pigments?
ColorFabb sells samples. I don't know if Ultimaker has samples of its colors?
Note that "transparent" filament does not stay transparent after printing it: it becomes translucent like frosted glass, due to the air entrapped inbetween the sausages. If you need translucency this is fine of course. But otherwise opaque materials will usually be more beautiful. At least, according to my taste. But feel free to see things differently. :)
Don't buy excess PLA rolls: after sitting for one year, PLA gets harder and more brittle in my experience.
Avoid special particle-filled materials (wood-fill, metal-fill, carbon-fill,...), avoid ABS (poor layer bonding, poor bed bonding, poisonous smell), avoid high-temp and soft materials (difficult to print), etc..., until you are well familiar with the process. So start with a gentle learning curve, that will be steep enough.
Also: buy a few big storage boxes in PE or PP, and buy big sacks of dessiccant (you can find that in car shops to dehumidify car interiors), and store all filament in these boxes to keep it dry.
-
I like the idea, but doesn't it cause extra friction in the tube?
-
For us it was just a little bit too small. The height was okay, the X- and Y-size not. And it did not have a heated bed: if that had been included from the beginning, I think it would have had more success.
-
Thanks for the feedback.
When using dilluted wood glue for PET, indeed I have had chips being taken out of the glass. I already heard weird loud cracking sounds while it was still cooling down, before removing. So I think that was the moment the glass chipped. Mostly I print PET on bare glass, without salt. But in both cases (salt or bare glass) I have to use as little cooling as possible, otherwise large models tend to warp, so for PET bonding is not as good as for PLA. But results may differ a lot from brand to brand.
For PLA, since 2 years I always use the salt method with success, although here too there are differences in bonding from brand to brand. The biggest advantage for me is its simplicity. But I still don't really know *why* the salt method works, on a chemical or physical level. I guess it has to do with surface tension: soap and oils reduce surface tension and reduce bonding. Salt on the other hand increases surface tension, and by observation increases bonding (of PLA, not for some other materials), so it might have to do with surface tension too? Or would mechanical grip around the salt crystals, on a microscopic level, also play a role?
So if there would be any chemists around here, feel free to add your viewpoint. Educated guessing is also okay for me, as long as you tell so.
-
Blender is inderdaad erg ingewikkeld, en meer geschikt voor organische vormen dan voor geometrische vormen, dus niet optimaal voor dobbelstenen. Fusion 360 ken ik niet, dus daar kan ik niets over zeggen. Misschien kan het ongeveer hetzelfde als DesignSpark Mechanical? Iedereen heeft zijn eigen voorkeur inzake programma's en user-interface, dus je moet iets zoeken dat je ligt; en dat genoeg mogelijkheden biedt. Youtube filmpjes vergelijken kan al een indicatie geven of het je ligt, en of het genoeg kan.
Bij goede 3D-software moet je de uitsparingen voor de ogen in je dobbelsteen niet apart ontwerpen. Ontwerp eerst de steen. Ontwerp dan de ogen in een andere kleur, op de juiste plaats in die steen. En trek daarna de ogen af van de steen. De naam van die functie verschilt, soms noemt het boolean operations, soms subtract, soms cut... Op die manier heeft je steen automatisch de juiste openingen op de juiste plaats; en dan passen de ogen daar perfect in.Wil je erg ronde hoeken op je dobbelsteen, dan kan je een bol maken die iets groter is dan je steen, maar waar de hoekpunten er nog net uit steken. En dan die uitstekende hoekpunten buiten de bol eraf snijden.
3D-ontwerpen is totaal anders van design-filosofie dan werken met klei, of dan draaien en freezen in hout en metaal. Het is even wennen. Het helpt om veel Youtube filmpjes te bekijken, gewoon maar om die workflow te zien. Het komt erop neer dat je moet denken in termen van "simpele basisvormen optellen en aftrekken", om zo tot zeer complexe vormen te komen. Dat kan zeker in DesignSpark Mechanical; en wellicht ook in Fusion 360.
Na het ontwerp kan je de steen (=dus met uitsparingen) opslaan als apart bestand, en de ogen opslaan als een tweede bestand. En beiden exporteren als STL-bestanden om te kunnen printen. En deze dan in Cura laden en elk definieren als aparte kleur. Hoe je dat in Cura doet, laat ik aan Sander over, want ik heb geen UM3 of andere meerkleurenprinter. -
1 minute ago, Ghene said:
As some of the fabrics are thick, do you think I should calibrate the build plate with the fabric on or is that a stupid idea? ?
Yes, that seems a good idea. The fabric of our posters is ca. 0.5mm thick I guess (I don't have it here at the moment), so it sure won't fit under a 0.1mm calibrated nozzle distance. Some painting canvas might be even thicker. I think you will need to establish the best plate calibration by trial and error. Maybe the nozzle should just very lightly touch the fabric, or be maximum up to 0.1mm above it, depending on the roughness, thickness and compressibility of the fabric? Anyway keep watching carefully. If you have success, let us know what worked best.
-
The molten filament has to be able to flow into the pores of the fabric, and to saturate it for best bonding. So the fabric should not chemically reject the filament, and it should not contain soaps and other stuff that rejects filament and destroys bonding. I think you definitely need a non-stretch material. If you would use stretch, then (1) or it is going to wrinkle afterwards if you print on it while stretched, (2) or it is going to move all places if you print on it while not stretched.
Recently we had scientific posters (120cm x 90cm) printed on canvas, instead of on paper. This canvas is quite thick and very stable. It does not sag when you hang up the poster on the display panels. But I don't know what material it is. Maybe you could search in that direction? Or try oil painting canvas? If you would have an artist's painting shop in your environment, it might be a good idea to go there for advice and to feel the materials?
-
2
-
-
Nice piece of art.
Maybe you could simulate or replace the ceramics with gypsum or fine concrete, or something similar? You could make a negative of this model, cut it in half and add flanges to make a mould that can be opened. And then fill the mould with the gypsum, concrete or whatever else of your choice? So-called "hard gypsum" as used in dentistry is quite strong, and is available in different colors, or you could color it yourself. On Youtube you can find lots of good videos on mould making and casting.
PS: what does the front side of the vase look like?
-
1
-
-
12 hours ago, aubustou said:
Yes, I understood quite a bit. The filament is thick over a large distance along it, around 5cm starting from the hot end. I never had this problem for over a year on an Ultimaker 2+.
Well my Ultimaker 3 is brand new. I bought it in past May. Therefore, I don't think it's a lack of maintenance because I used it less than 5 hours.
I definitely think there is a problem with cooling in the hotend, but the fans are working correctly...
Did you also check the third fan, the little one at the back of the nozzle? Your description of phenomena is exactly what happens when this fan fails: heat traveling way too far up into the filament, and softening this before the nozzle, so it can't get into the nozzle anymore. Maybe some piece of plastic got stuck into the fan? Or a loose electrical wire?
-
3 hours ago, drayson said:
hmmm.... I just put them in a cloth mesh to prevent them from escaping ?
...
Yes, this seems the best idea to me. Socks like to play hide and seek. If you just bind two together, they may still play their little game, but then as a siamese twin.
Anyway, if you make clips, make sure they are way bigger than the holes in your washing machine, also if they would break or deform. Otherwise these parts might go play hide and seek in the pumps and valves of the machine...
Another option is that you don't mind wearing socks of different colors. Some people have that as their little game. :)
-
In ieder geval: gebruik zeker *geen* SketchUp, want dat is een garantie op zware problemen bij het 3D-printen. Dit komt doordat SketchUp geen waterdichte, solide 3D-modellen maakt, maar eerder een soort "papieren" modellen, met spleten tussen de naden. Zulke eindeloos dunne papieren wanden zijn niet printbaar. SketchUp is nooit ontworpen voor 3D-printen, maar alleen voor visuele representatie van gebouwen op een computer; en daarvoor is het wel goed.
Ik zou voorstellen dat je op Youtube eens kijkt naar DesignSpark Mechanical ("DSM"). Dat is freeware, en vereist enkel registratie. Verdeeld door de elektronica-distributeur RS Components. Het is een afgeslankte versie van het betalende SpaceClaim, maar het is meer dan goed genoeg voor 3D-printen. Kijk eens dat je bevalt? Iedereen heeft zijn persoonlijke voorkeur inzake user-interface en workflow.
Er zijn op Youtube een massa goede demo-video's en tutorials te vinden. DesignSpark Mechanical kan je op enkele dagen leren gebruiken, in tegenstelling tot de maanden bij veel andere programmma's.
DSM heeft een eenvoudige user-interface en werkt handig en snel. En in de 2.5 jaar dat ik het gebruik, heb ik nog nooit problemen gehad bij het printen, veroorzaakt door de software. (Wel natuurlijk door ontwerpfouten van mijzelf, maar dat is iets anders.)
Edit: hoe je in meerdere kleuren moet printen, eens je het model ontworpen hebt, weet ik niet. Ik heb alleen printers met één nozzle, dus voor eenkleurige prints.
-
2
-
-
I have a PLA sift sitting in the sink of my laboratory, just to test how well it would survive this harsh environment of soaps and diluted alcohol and detergents. Currently this is in colorFabb PLA/PHA. It does not show any damage after one year. The previous one was Ultimaker light blue, and that got a little bit dull after a year. I replaced it because the newer design is more handy to take out and to clean (and by then my UM blue filament was of course used up). I would say: just try it, it is a good experiment. :)
-
On 5/31/2018 at 7:15 PM, smartavionics said:
...
I find that with PLA some sharp corners do tend to curl up and adding a skirt that's almost a brim is a great help. My theory being that it's something to do with the temperature gradient on lower layers and having the brim helps keep the temperature uniform across the bottom of the part (this could be complete rubbish but that doesn't stop it being effective ?)
I guess that might make sense indeed? A theory is as good as it produces results.
I found that the cooling fans do reduce bed temp by 10 to 15°C in the area where they are blowing; measured with an IR-thermometer gun.
So maybe such a brim or wide skirt prevents the glass from cooling down too much and thus improves bonding? And/or maybe it deflects a little bit of the airflow away from the bottom layer? So that the first layer is sitting in the shadow of the wind?
Anyway, I am gonna keep this theory in mind and try it on some of my next models with lots of tiny holes and sharp corners.
-
The supports have to catch the lowest part of each winding, otherwise that area will print in thin air and fall down like spaghetti. So I think it might be best to have the support run through the center of the windings. As shown in this cut-out. This should print, but the lowest layers of each winding will still be ugly due to the overhang. And then in CAD merge both support and spring, so it is in one piece, otherwise it might not slice correctly. After completion, cutting out the support and sanding will be a pain in the ass, but yeah...
Also, make sure you have some soft of baseplate to make it stick to the glass. Otherwise it will come off during printing, roll away, and produce spaghetti. Big overhangs tend to curl up, so the nozzle may bang into them. This requires good bonding to the glass to survive these bangings.
These are the two reasons why I would rather try to print it vertically. But I have never printed such a spring, so it is all a bit of educated guessing, and it may take a few tests to get it right.
Overlap in Layers
in UltiMaker Cura
Posted
It is very hard to print transparant with FDM 3D-printing. Due to the air gaps inbetween the extruded sausages, it usually looks like frosted glass. So, out of curiosity, how did your internal cylinders come out? Were they visible after printing? Maybe you could post a photo?
If I want to make internal details visible using transparent PET, they should sit no more than 0.5mm below the surface. And even that only works on the flat surface facing the glass plate. Not on the top layer. This is good for a logo or a mm-scale ruler. (Since I have only single nozzle printers, I can only make hollows, not two-color prints.)
Text caps height is 3.5mm, sitting 0.5mm below the surface, character depth 1.0mm.