geert_2
-
Posts
2,084 -
Joined
-
Last visited
-
Days Won
34
Content Type
Forums
Events
3D Prints
Posts posted by geert_2
-
-
12 hours ago, tinkergnome said:
Yes indeed. :)
Or see for the full PDF-manual and some other stuff here: https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/
-
2
-
-
Another method to prevent models from sliding around stuck to the nozzle, is to add a sort of wall with a big base plate around the model, unconnected. If the model would come off the glass, the wall will prevent it from sliding around. This will still ruin the model of course, but not the nozzle, hopefully.
As in this example: the "warptest" is very likely to come off, due to the very small baseplate and very high warping forces, and the curling up of the overhangs. But the big baseplate around it has a good bonding and will keep it less or more in place. This could be a solution for difficult or tiny models where a brim is too difficult to remove.
You could of course use a normal brim too, or design mickey mouse ears into your models, but then the models need more post-processing to cut these off.
STL-file:
Be sure to stay with the printer while trying this testfile, and carefully watch what happens.
Concerning cleaning aids: recently I saw that some window cleaners do contain soaps, some contain alcohol, and others contain acids (vinegar), and even other stuff. This might explain the huge differences in bonding performance after using them. Soaps will reduce bonding, alcohols should be okay I guess, and I don't know about acids. Anyway, it should never hurt to wash the glass with pure warm tap water only, after using any cleaner, to remove traces of soap or other left-overs.
-
2
-
-
Wow, designing them is one thing, but printing people in scale 1:100 (thus ca. 17mm high) with lots of detail on an FDM-printer, that is going to be tough! Your print-settings would have to be absolutely perfect to be able to even recognise them as a man or girl, or even as human. Maybe cloakfiend could do it, but else?
I don't want to ruin your contest, but I would rather buy unpainted sets of figurines in HO-scale from Preiser or similar companies. There are sets for leisure people, business people, working people, sitting people, musicians, animals, farmers, and lots of others. They are all very detailed, and modeled after real people. In total they have thousands of different models. It must have been an incredible job to model all these manually on such a small scale, and to make injection moulds for it. Usually they come in white or pastel colors. There is no way you can beat that with 3D-printing.
As a kid I used to paint them myself. But for architecture maybe it is better to keep them white?
In addition to their HO-scale series (=1:87) which seems to be the biggest assortment, they also exist in scale N (=ca. 1:150), scale O (=1:43), architecture scales (1:200, 1:100, 1:50), and others. But I think HO people would also be suitable for architecture buildings at 1:100, since the difference is not that big.
Google for: ho scale people preiser unpainted set
And then view the images.
Preiser homepage:
https://www.preiserfiguren.de/
Or see the PDF catalog of Preiser's unpainted HO-sets:
https://www.preiserfiguren.de/download.php?file=055 - 16325-16400.pdf
-
STL-file of the above model:
Print this with 100% infill, for maximum warping forces and test effect.
Stay with the printer while printing this, and watch carefully what happens.
If the bonding is not perfect, this model is likely to come off the build plate. Usually the corners are lifting first. After it comes off, it might stick to the nozzle and get dragged around, encapsulating the nozzle in molten plastic. Or it might be thrown aside after which the printer will print spaghetti. Be prepared to stop the printer when this happens.
Phenomena like warping of the bottom layer, or the edges of overhangs curling up, will also happen with other "easy" models, but in a lesser degree. With this test model you can compare various bonding methods, and try which works best for your materials. If it works well with this model, it is likely to work well with easier models too (=with less overhangs, and a larger bottom area in contact with the glass).
If you can not print this model at all, if it keeps coming off, you may need to find a better bonding method to be safe. Or add flanges or "ears" to your models.
-
1
-
-
On 7/4/2018 at 11:49 PM, SteveCox3D said:
...
Someone who I showed it to was very interested in it as a potential application where they were trying to automate strawberry picking which needs a damage-free grip. So-called "soft robotics" is a new area where robots become less mechanical but have more of a human touch and maybe this would be appropriate for that.
...
Yes, I could see the use for such applications with fragile materials indeed: biological samples, glass bottles,... I hadn't thought of that.
For some applications, maybe you could even print these pliers out of one material, and by just varying the geometry get both the desired stiffness and softness where needed? For example, by adding a similar pattern at the clamps, it would get a softer grip around the sample, like rubber?
Interesting field of research.
-
To me it looks like there might be multiple issues?
I have the impression that the bed is too high, and thus the first layer is not squeezed into the glass enough and does not bond well?
Maybe your bonding method also needs adjusting? For example if you use the glue stick, wipe it with a wet tissue after applying, to get a very thin layer? Or clean the glass and apply the bonding more often?
And then of course there is underextrusion at irregular intervals, which could have lots of causes: partially blocked nozzle, too high printing speed, too low temperature, spool near the end and wound too tightly (this acts as a spring that resists unwinding), deformed teflon coupler, wrong feeder pressure, dirty feeder, wrong filament diameter, dirt in the bowden tube, or other things along this line...? But I am not familiar with the UM2+, so difficult for me to say, it is a bit of guessing.
-
What bonding method do you use, if any? If you print on bare glass without bonding method, results may vary wildly: in dry winter weather it may stick very well, while in moist summer models might come off unexpectedly.
Also, glass temperature plays a role: for me bonding is best between 55°C and 60°C. At lower temps, the models might suddenly pop off the glass. At higher temps, the models stay too soft and the warping forces tend to peel the model off the glass. This effect will differ from material to material, and from brand to brand, within the same material.
Also, I would suggest you have a look at all available bonding methods, try them and select one that works well and that you like:
- The "official glue stick", and maybe wipe and spread the glue with a wet tissue after applying.
- Hairspray (neotko's preferred method).
- Dilluted wood glue 10% in water (gr5's preferred method).
- Salt method: wipe the glass plate with a tissue moistened with salt water, prior to starting a print. So that there is a very thin mist of salt stuck on the glass, almost invisible. This is my preferred method, but it only works for PLA, not for other materials. And I do *not* recommend it for high models, or models with big overhangs, because these might get knocked off. For me this worked well with Ultimaker PLA and colorFabb PLA/PHA, but less with ICE PLA, so they seem to have a different composition.
- 3DLac or similar sprays.
- Other?
Be very sure to *stay with the printer* during the first tryouts of any bonding method, and try small and difficult test pieces first, which don't waste too much time and material. For example something like this: it exerts very high warping forces on a tiny adhesion area.
-
2
-
-
Cool indeed. It would turn the printer into a light CNC machine.
Can the holder withstand the occasional getting stuck or slamming of the drills? If not, maybe you could consider fixing the clamp with wing screws, or so, instead of magnets? This also makes fast assembling possible.
Just make sure you level the bed correctly before starting to drill. :)
-
All four of them now... :)
Scale: 4cm = 1m
-
1
-
-
Indeed not my office; it's for female collegues. Meanwhile there are four walking around in the model. :)
But I planned my own laboratory in the same way, a couple of years ago, when we had to move to another campus. I represented each piece of lab equipment we had in plasticine. It made the discussions with the project leaders a lot easier, since I could clearly demonstrate what furniture I needed, and why and where I needed it. But then of course I made male figurines, although without all details. :)
You can recognise the hydraulic pump, test bench, vacuum cleaner, ladder, weight balance,... Note that the scale is half of the above one: 1cm = 0.5m.
I like 3D-editing and printing, it is great, and we can now develop things we couldn't have done in the past. But I do not go along with the hype in the press that "it is the best solution for everything". Sometimes it isn't. Like in these cases.
-
1
-
-
Maybe it's not 3D-printed, but it is still 3D-editing. Sort of... I hope I can post this here?
Most people are not used to reading 2D-plans on paper, and they have difficulty interpreting them correctly. So, for the reorganisation of an office, I made all furniture out of plasticine. This allows rearranging and trying lots of layouts in a very short time. Moving around the furniture in plasticine goes a lot faster than moving around 3D-models on a computer screen; and looking at plasticine models from all angles goes faster than orbiting around a 3D-model on-screen. Further, you can easily walk around and show this to collegues, without the need of a computer with 3D-editing software.
It looks like Flintstone-stuff, I know. But actually, everything is quite accurately on-scale: 1cm = 25cm, thus 4cm = 1m (the floor is standard mm-paper under a plexiglass plate). This is advanced 3D-editing à la Flintstone. :)
All this took ca. 2.5 hours to make, most of which was spent warming up and softening the plasticine, as I forgot to put it in the oven at 40°C. I could have designed it all in CAD and 3D-print it, but that would probably have taken more than a week to print, especially if I wanted the parts to be solid enough to sit stable when moving around the whole scene. Plasticine weights a lot and is sticky, so no problem here. Modeling the girls in CAD would for sure have taken more than a week, maybe more than a month, since I would need to learn Blender first. Now it took me ca. 15 minutes per girl.
The main difference between these models and traditional 3D-models is that I make these *before* I draw any plans. Usually it is the other way round. All collegues involved can freely move around the furniture, walk the figurines around in the scene, and try out and discuss what looks best to them. Only after the final layout is chosen, we would draw any plans. Although often plans will not be needed anymore, since the model is clear enough by itself. So a few photos with quick annotations about part numbers and dimensions will do.
It is not 3D-CAD design, but rather 3D-MAD design. Manually Aided Design... :)
-
4
-
-
Thanks for this interesting post with practical examples.
I especially like the digital sun clock, because it shows an inspiring and creative new way of thinking, breaking out of the box. This is what we need to learn more. At the same time this concept reminds us of the ancient astronomers, who also used such ideas in their temples.
Then I have a question concerning the orange plier: this is an interesting concept, but how long does such a mechanism survive in real life? And how well can it handle a load, if you want to fix a bolt or pipe for example? Can it really clamp with force? And how do you come to such a structure, thus how do you develop it? It would be interesting if you could describe the workflow, or the idea-flow.
Looking forward to your next posts.
-
On 6/25/2018 at 10:16 AM, marcogiers said:
Dank je wel kmanstudios. Prachtig werk!
Ik heb mijn 3dpen gekocht en ik ben er mee aan de slag gegaan. Ik denk dat het mooi wordt.
Kan je bij gelegenheid eens een paar resultaten tonen? Nadat je wat ingewerkt bent. Het kan altijd nuttig zijn voor andere mensen met diezelfde vragen.
In modelbouw, om bij een treinlandschap bomen, takken, struiken, en hekken rond een weide te maken, kan ik het nut inderdaad wel zien. Daar mag je wat grilligere en onregelmatigere vormen hebben. En dan heb je unieke modellen tegen een fractie van de prijs van de commerciële modelbouwspullen.
De 3D-pen die ik gezien heb, was iets met "doodle" in de naam, en je moest er rechte staafjes filament bij kopen, dus mogelijk de 3Doodler die kmanstudios beschreef?
@peggyb: misschien zou je het laagjes-effect in je modellen weg kunnen krijgen met het aceton smoothing van cloakfiend? Bij hem zie je zelden laagjes, maar er gaat toch niet veel detail verloren. Maar ik weet niet of dat met moldlay en soortgelijke ook werkt? Of inplaats van aceton een ander oplosmiddel? Bij wax zou eerst diepvriezen en dan de buitenste laag kort verhitten met een hete-luchtblazer misschien ook werken? En dan direct terug afkoelen.
-
1
-
-
13 hours ago, kmanstudios said:
These would fit well on the bar of a classy restaurant, as an eye catcher, I think.
-
1
-
-
10 hours ago, kmanstudios said:
I do not think silicone will stick to the PVA enough to use as the support material for your models.
This is a good point, I hadn't thought of that. Sticky silicone like used in bathrooms and pipings (=often acid, and cured by moisture in the air) might work, but this has a curing time of 24h or more. But I don't think you can print in sticky silicone: it would string like mad. Mould-making silicone definitely won't stick, indeed, otherwise it would be useless for mould-making: you would never be able to get the model out of the mould. Here too, printing will be difficult: if liquid, it will leak away, if paste, it will string.
Mould-making silicone has curing times between a few minutes (for dental use, but these do not have much strength) and 24h (for traditional mould making, much stronger). If you would print in silicone, then you would have to wait after each layer until the previous layer is cured enough so that it does not sag or leak away. But it must still be tacky, otherwise the next layer won't bond. If you would print with a syringe system with mixing tip, then the silicone in the mixing tip will be cured too during this waiting cycle, and you would have to replace and prime the mixing tip for each layer. Unless you keep printing somewhere else to keep the flow going, wasting a lot of material.
All these things add a lot of complexity to the system.
Another point to consider: do the models really have to be in silicone? Or would that just be the prefered material, because of its resistance against fracture when dropping? In the last case, maybe you could consider printing in nylon too? Nylon may not be easy to print, but it is compatible with PVA supports.
You might even be able to make the mould in PLA, if you pour in a silicone that is heat resistant enough. Most silicones can handle 250°C for a short time. Then you could melt the PLA away. Or heat it to 100°C and peel it away.
This is why making a traditional mould (in PVA, breakaway, wax, PLA, whatever...), and then pouring non-stick silicone in it, seems the best option to me, if it has to be silicone. If you use a low-viscosity silicone with long curing time, you have plenty of time to remove the bubbles by shaking and vacuuming.
I would suggest you watch a couple of "silicone mould making and casting" videos on Youtube. There are lots of good ones, made by real professionals, and they give a wealth of info.
Anyway, we would appreciate if you could give feedback on the results, of whatever method you chose.
Fast curing silicone (1...5min), with mixing tips. These are a paste when extruded, like slimy pudding.
Fast curing (ca. 5min), paste-like when extruded. See the string at the back where the mixing tip took off. It was a long string, but it folded back on itself.
Medium curing (15...60min) liquid silicone for pouring moulds. This is liquid like thin sirup or liquid honey. These come in bottles of 1 liter, to be mixed 50-50. The orange took ca. 1h to cure and was more liquid. The blue was a bit more viscous, and took ca. 15 min to cure: this was too fast to allow the bubbles to rise to the surface, so they got trapped. I poured it in the hard shell, and around the model (not shown). So you would need a similar liquid silicone to pour into empty canals in the mould.
Shown here are dental silicones, but you are cheaper off buying artist silicones. It has to be platinum cured silicone. (No tin-cured silicone, which may decompose over time).
-
No, I rather thought about only printing the negative (thus the surrondings) in PVA. No PLA at all. So you would have a hollow dissolvable PVA mould, with empty canals. In which you could pour the liquid silicone.
But this would only work if you can orientate the model so that there are no steep overhangs. For example let's take the "heart" symbol shown next to each post here on this forum, which you click to like. Imagine this is a 3D-model. Then you would only print the grey ball in PVA, and leave the heart itself hollow. So you can fill that hollow heart with silicone. But if you print the heart as shown, it would not work because of the overhangs. However, if you print it upside down, with the point facing upwards, and add a canal for filling at that point, it could be done. At least, it would be worth trying.
If you would want to print the model in PVA, and the surroundings in another removable material, then maybe some "breakaway" material might work as surrounding? Then the PVA-model could be dissolved, leaving the breakaway-material as a stable mould for pouring the silicone in? And then, after curing, break away the "breakaway" material?
Or maybe print the model in PVA, and the surrounding mould in wax, which is water resistant? Then you could first dissolve the PVA, next pour the silicone model, and then melt away the wax?
Or something else along these lines? An imperfect method that does exist and can be used immediately, might be better than a perfect method that does not yet exist.
If not clear, feel free to ask. I know what I mean when writing this, because I have the pictures in mind. But I am not sure the meaning comes across. If not, I will make a quick drawing (but not today).
-
And how is strength, out of curiosity? Can it flex without breaking, and without delaminating?
-
A couple of months ago this has been discussed on the forum: one of the regular users gave a good and complete overview how he did his calculations, and they included almost everything. But I don't remember who it was, nor the name of the topic. So you may have to search a bit. Anyway, that description could be a good starting point.
-
2 hours ago, kmanstudios said:
Oh, tool function is of utmost importance whatever it is called.
Oddly, nobody noticed the old fart rambling about bad language usage whilst inventing words himself.....slopification anyone? ?
I looked it up in Google. It appears that "slop(p)ification" is already in use, although not very common: I got 116 results for "sloppification" and 199 for "slopification". But the latter also includes the meaning of "slope" (side of a mountain). So I would rather prefer the version with double "p" in this context, since it is derived from "sloppy", like "sloppiness".
It is a nice word with a clear meaning. I am against sloppification too, unless it is in the context of girls wearing leggings (which is considered sloppy by some). :)
Anyway, I think this thread could well create a new trend in 3D-printing: the buttification of vases and other decorative items and tools...
-
Have you considered printing an inverse (negative) of your model? For example a solid block, where the model is subtracted from, leaving the shape of the model as hollow canals in the block? And then fill those canals with liquid silicone? If you print that model in PVA, you can dissolve it, and you get the positive silicone model out.
I am not sure it would work for your models, but it might be worth trying?
The main issues would probably be:
- Big overhangs would cause sagging of the top layer. Thus causing rough top sides in the casted silicone model.
- You may need to design venting canals in the model, to allow entrapped air to escape. This may require some thinking and trial and error.
- A cured silicone model is hard to post-process, at least the silicones I used, because they are too flexible. It's hard to grind and polish.
Advantages are:
- You can use any liquid silicone of your choice, as long as its viscosity is low enough to fill the canals.
- You can add any color to the silicone mix.
- Curing time is no problem, you can use silicone that takes all night to cure (these are often stronger than fast-curing ones).
- Degassing to remove bubbles is no problem (unless this would damage the PVA support?)
- And shaking the mould to remove bubbles is no problem.
If the overhangs would be too large to print well without distortion, maybe it might be possible to cut the (negative of the) model in multple pieces, print these separately, and then stick them together? Wet the edges of the PVA, and glue parts together? PVA is water-based glue, so it might work?
First try this concept on small models which do contain all critical aspects: overhangs, entrapped air, fine details... But which don't waste too much material and printing time, until you get it right...
-
If this happened on both printers in a short time, could it be the connecting cables were bent in the wrong way? So that they got stuk, or got a high mechanical load or twist when the plate was moving up and down? I am not an Ultimaker employee, and not familiar with the details, but I read somewhere that these cables need to be bent in a special position, to avoid distortion?
While this won't repair your issue of course, maybe it might prevent it from happening again, if it would have been a contributing factor to the cause?
-
I guess 32GB might be too big for small controllers in 3D-printers?
If you would have some knowledge about partitioning and formatting, you could try formatting the card as "floppy disk", or as "hard disk", and try both. There seems to be a difference. Or you could try partitioning it into a partition of only 4GB, and leave the rest unused. And then only format that 4GB partition in FAT32. But all this may take some trial and error, and it is not sure it will work. You will have to Google a bit.
-
I was wondering, your butts (well maybe not yours, but the ones you print), are they actually based on 3D-scans? Or are they sculpted from scratch in Blender, Sculptris, or so? Or scans from clay models?
-
Ik heb nog niet zelf met 3D-pen gewerkt, maar het al wel zien doen in de winkel. In mijn ogen is dat waardeloze brol. Een paar jaar geleden was dat een hype, maar nu hoor je daar niets meer van. Dus ik denk niet dat die gebruikers daar veel voldoening uit gehaald hebben. Het resultaat is erg bibberig, met veel gaten en slechte verbindingen. Ik denk niet dat je daar een deftig afgietsel van kan maken. En als het wel lukt, gaat het model wellicht kapot bij het verwijderen (kapotkloppen) van de mal.
Als je het model liefst wil maken met handwerk, (ipv. op de computer), dan kan je het beter in tandheelkundige was boetseren. Daar bestaan honderden kleine basisvormen en dunne draadjes in was, die je zo aan elkaar kan smelten en kan plooien. Dat gaat veel mooier zijn dan een 3D-pen. Juweliers zullen ook wel zulke basisvormen in was hebben die je aan elkaar kan smelten, denk ik?
Wil je liever 3D-printen, dan zou ik zeggen: ontwerp een testmodel dat alle delicate aspecten bevat die typisch zijn voor jouw ontwerpen. En laat dat model dan op verschillende soorten printers printen. In België kan het bij Materialise in Leuven, en bij Melotte (denk ik), of in de UK bij Shapeways (?). Beetje Googelen.
Daarbij kan je ook de productiemethode kiezen. Dus je kan gaan voor lasersintering (aan elkaar gesmolten poeder), voor laserstraal in vloeistof, FDM (zoals Ultimakers), lichtbak in vloeistof, gips; en andere technieken. Voor kleine voorwerpen is de kostprijs daarvan nog beperkt, omdat die meestal berekend wordt per gewicht + setup kosten. Misschien is laten printen in wax ook mogelijk?
Of je kan het ineens laten printen in metaal, met diverse soorten afwerking (ruw, gepolijst,...). Dat kan ook handig zijn voor unieke stukken. En dan zelf evtl. nog verder polijsten, vergulden of verzilveren.
Op die manier kan je alle productiemethodes naast elkaar vergelijken tegen een betaalbare prijs.
Alleszins, zelf je model in was maken, of (laten) 3D-printen, gaat veel betere kwaliteit opleveren dan een 3D-pen.
-
1
-
Very frustrated with Ultimaker 2+ layer issues
in Improve your 3D prints
Posted
If it happens mainly with the 0.8mm nozzles, a factor could be temperature: it requires 4x more heat to melt enough plastic in the same time for a 0.8mm nozzle than for a 0.4mm nozzle (twice the diameter = 4x the surface). So you might want to reduce speed and/or increase temp. Try adjusting these on the fly on a test piece, and watch what happens.