geert_2
-
Posts
2,084 -
Joined
-
Last visited
-
Days Won
34
Content Type
Forums
Events
3D Prints
Posts posted by geert_2
-
-
Up till your question, I hadn't even heard about PMMA as 3D-printing filament, even though I have used lots of two-component PMMA materials (powder + liquid) and PMMA-based light-curing materials.
So I googled it: indeed it does exist, and some people like it, but some complain it is very brittle (which is what I would expect, similar to two-component PMMA), and it shrinks too much, like ABS.
Is there a reason you need PMMA, and can not use PET, polycarbonate, some high-temp polyester, or even PLA (but only for moulds not subjected to temperatures above room temp)?
-
Or split the model in such a way that the seam lines sit in a natural indentation or other dimensional change of the model. Or create an indentation or decoration to hide it. So it becomes a "feature" rather than a defect. This may of course not always be possible.
-
Moisture might indeed prevent sticking. I had that too in the beginning when the weather was warm and wet.
Also, some cleaning aids (window cleaners or some alcohols) do contain soap or oils that prevent sticking. So, I would recommend to clean the glass with pure warm tap water only. You could use isopropylalcohol first (this should remove oils without leaving traces), but then clean again with warm water only.
For printing PLA, I always wipe the glass with a tissue moistened with salt water: gently wipe and keep wiping until it dries. This should leave an almost invisible mist of salt stuck to the glass plate: it looks just a little bit dirty. For PLA this greatly improves bonding. (This does not work for ABS, and thus probably also not for PC, nylon).
Apart from that, there could of course be other issues: nozzle temp too low, bed temp too low or too high, distance nozzle to bed calibration too high or too low, material flow too low,...
Store all filament in a sealed box with disseccant. I use bags with color indicator, designed to dry car interiors (so they have a high capacity).
-
In the standard UM2 firmware you can also manually heat the build plate, via "Maintenance > Advanced".
However, if you keep the temp high after print completion, then how are you going to get the models off? They do stick like glued, and you can lift the whole printer by pulling up a tiny model. While after cooling down they come off all by themself, without requiring any force. It may take you more time to get the models off a heated bed, than to wait until it is cooled and simply pick them off. You could speed up cooling by putting a fan in front of the printer, after completion.
And as said above: do not keep the nozzle heated without flow: the filament will burn and clog the nozzle in a few minutes. Printing temp is too close to burning temp.
-
More and more engineers in the USA and UK are switching to the metric system. Once you are used to it, it is sooooo much simpler and more logical than imperial units with their weird multiplying and conversion factors, and where some "units" have totally different dimensions, depending on where you are (such as gallons, if I remember well). It is just a question of time before everyone uses the metric system. So you can better be among the first than among the last. You will not regret it.
-
Geert, the word you're searching for is chuck or drill chuck. Have no idea why it's called that... have had PLA melt when drilling with a motor, but not generally other plastics
Thanks!
-
Maybe you could leave the holes the size they are: they will probably come out slightly smaller anyway. And then drill. Just try.
I would recommend drilling by hand, not with a machine: manually gives you much more feeling. A machine generates too much heat and may melt PLA or other low-temp plastics. Don't ask me how I know.
I have bought a separate drill clamp (I don't know the correct name in English) to mount my drill into, so I can get a good grip on it.Edit: added photo of drill clamp for manually drilling.
-
I don't have an UM3, so this is only a guess: from the photo there appears to be a way too big gap between support structure and bottom of the model. So the first layers of the model do sag. Also it looks like the support structure is not dense enough: not enough contact points between support and model for good support. I can't comment on the cause: a problem in the design? Or a problem with underextrusion of the support material? Or wrong settings?
-
1
-
-
If material is leaking out of the nozzle while jumping over gaps (thus causing strings), then indeed you could have underextrusion in the area after this. The software has no knowledge of these leaks, it doesn't supply extra material to compensate. If you print slow, you might try a lower temp, so the material is less liquid: 10°C less or so?
PS: don't leave a PLA printed model in your car in sunny weather, or don't leave it sitting in the sun on a concrete runway: it will warp. I don't know what effect warped wings would have on flight characteristics, but I don't think it will do any good...
-
The bed temp should be around the glass transition temp of the material: that is the temperature where it begins to soften and get flexible (which is very long before it is melting). For most PLA this Tg is around 55°C, if I remember well.
If you print PLA on bare glass, after wiping the glass with a tissue moistened with salt water, then the effects are as follows on my system:
- no heated bed: model does not stick at all, extruded sausages curl up and come off immediately,
- bed temp way too cold: printing starts fine, but model pops off suddenly during print,
- bed temp a bit too cold: corners lift due to insufficient bonding,
- bed temp good: no corners lifting, PLA sticks like glue (at least for Ultimaker and colorFabb PLA, some other brands stick a bit less),
- bed temp too high: whole model stays too soft, and corners lift due to the model being peeled off the glass. Also: model sags and gets severe "elephant feet".
Different bonding methods might give different optimal glass temps.
So, your best option is to design a test model, for example a ruler of 1cm x 1cm x 10cm, laying down (not vertical), and 100% filled. Then start with the bed at the Tg, and then try steps of 10°C above and below this.
If you want an even more critical test model, try an inverted prism: a small base of 5mm wide x 30mm long or so, and a wide top of 15mm x 30mm. This gives huge warping forces on a very small bonding area.
To prevent the primed "sausages" from being pulled into the print, I use a plied piece of steel wire around the priming area. It is inox spring steel (like used in bridges in the mouth), so it lasts forever. It is a metal version of the 3D-printed "filcatch" idea of another user here (I can't remember any names...).
-
Fascinating. Well @ultiarjan the same symptom is true of the newer formulations of Ultimaker PLA. If you leave it in the bowden for 8 hours it becomes very brittle! And it has nothing to do with moisture but has to do with mechanically straightening the filament and holding it in that position for many hours.
I really don't know why it does this. I'm guessing you get micro-fractures occurring slowly over many hours. Or maybe something down at the chemical bond level? What REALLY surprises me is that this same thing happens with non-PLA. I did *not* expect this to happen with PVA.
...
This is a late reply, but may still be useful.
I am pretty sure it are micro-cracks indeed. When you manually bend a piece of translucent or transparant PLA, or when you straighten a bent piece (like a tight-wound piece near the end of a spool), then you clearly see white micro-cracks growing. Some colors show these cracks better than others. If you stop bending the filament and release it, the cracks stop growing. But if you keep it under stress, I guess they will keep growing until the material breaks or almost breaks.
Near the end of spools, I always unwind a few meters of PLA, manually bend it around a skater wheel (7cm diameter) in the opposite direction, and then release it again, so it is sitting very loose on the spool. Then it feeds much easier into the bowden tube and nozzle, and it does no longer cause underextrusion on my UM2 printers (non-plus). But I always see those white micro-cracks indeed.
Most plastics have this, as you see in old plastic parts near areas that have been under stress (typically around screws or clamps), even where sunlight does not come (so the cause is not only UV). ABS, polystyrene and polycarbonate definitely have this. I don't know about PVA, but I wouldn't be surprised if it has this too.
In transparant plastics like polycarbonate you can see the stress by holding a polarising filter before and behind the part. Then you see the typical rainbow colors: the closer these color bands are together, the higher the stress, similar to height lines on a map.
-
This is a normal effect on all similar 3D-printers. The filament is still molten or somewhat flexible when laid down. When the nozzle is printing a corner, as in tight circles around holes, it pulls the molten filament to the inside of the curve.
In small holes this is even more visible: a 3mm hole may come out as a 1.5mm hole. It also depends on printing speed and temp.
Solutions could be: change the dimensions in the design, or go through the hole with a drill.
-
1
-
-
To prevent curling up, have you tried wiping the nozzle with a tissue moistened with silicon oil? And/or with PTFE oil?
This reduces build-up of molten filament on the nozzle, but it also reduces the tendency to curl up when priming the nozzle. So it might help during the print too?
If temp conductivity of a steel nozzle is much worse, maybe you could try increasing the temp by 10°C, and see what happens? If the filament is too cold when exiting the nozzle, it also won't stick well.
Further: did anything change on the glass plate? Different glue, different preparation or cleaning, different climate conditions (e.g. moist weather),...? In the very beginning I also had this problem in moist weather, especially after cleaning the glass plate with window cleaner: then nothing would stick. Now I only clean with pure warm tap water (no soap, no additives), and then I wipe the glass plate with a tissue moistened with salt water: this seems to increase surface tension and greatly improves bonding of PLA. For other materials (polyester, PET) gr5's method with dilluted wood glue works well.
-
I think printing little test plates of 10cm x 2cm x 1mm would do. This doesn't waste too much material. Design them with a hole so you can hang them on a line, like a decoration.
Even better: print two copies of each, and keep one stored inside in a dark space, to compare.
Then every month you could compare the plates visually, and flex them to feel if they got brittle and if they snap under pressure. In some materials UV-damage may be visible, but in some it is not, and the item just falls apart when you flex it.
-
I don't have any experience with replacing the heater and sensor on these printers. But I did build a couple of electronics machines in the past. So my comments below are educated guesses, but not necessarily correct or complete.
Probably the heater got red hot because you disconnected the sensor from it. So the feedback-loop was broken, and the system kept on heating and heating, because it did not sense any change in temp. I don't know how long these heaters can withstand this without burning out?
If - after assembling everything again - the temp overshoots a lot, with identical components as before, then I would guess that: or the heat does not flow away fast enough (so the applied heat overshoots the set temp), or the sensor does not sense the changes fast enough (maybe bad mechanical connection).
Things you might want to verify:
- Does the little fan at the back of the nozzle work? It should take superfluous heat away.
- Is there good mechanical contact between sensor and nozzle?
- Good mechanical contact between heater and nozzle?
- Is the aluminum fan mount not touching the nozzle? This can be hard to see, so you might need a mirror and magnifying lens. (However, if the aluminum plate would be touching the nozzle, it would rather work as a heat sink, and cause the temp to go up too slow, or not reach the set temp at all, instead of overshooting, I think.)
- I guess you did not update the firmware, and did not change any firmware settings? Different firmware might behave differently. But if you did not touch this, this isn't going to be the problem.
If the system still less or more regulates the temperature, I guess the electronics are still okay. Otherwise I would expect the heater to be full on or full off, but not regulating.
I do not know what is inside these heaters, but probably just a resistor. In that case, it doesn't matter in which way you connect the wires (the "plus and minus"), since there is no plus and minus. Just make sure there is no short circuit and the connections are good, so the heat is developed in the heater only, and not anywhere else along the traject in a bad connection or in a short circuit.
But as said, these are guesses...
-
To improve UV-resistance, maybe you could use black or highly opaque materials? This prevents the sunlight and UV-rays from penetrating deeply into the material. That was a reason they originally started to make car tires black by adding carbon in the old days. Originally they had the light brown natural rubber color. Of course, manufacturers use synthetic rubber now, and they add lots of other things into the mix (e.g. silicon), so I don't know if they do still add carbon.
Disadvantage of black is of course that it accumulates more heat. So it definitely won't work with PLA.
I guess you need to expect that you have to reprint it anyway in a few years. Apart from car dashboards (vinyl?) and rear light lenses, I haven't seen many plastics that survive in sunlight. Even not headlight lenses: you see a lot of lenses that get brown or get cracked, even in expensive cars.
Maybe you could print a small test plate in any material and any color you have, and put them all out in the sun and rain? Then, in a year, you can compare?
-
Concerning PLA smoothing, if you haven't done so, I would recommend that you read the thread about PLA smoothing created by user cloakfiend. This has lots of good info and photos (so you can verify the results). He is the specialist on PLA-smoothing. This thread contains about all that is known about this subject.
It should be this:
https://ultimaker.com/en/community/10412-acetone-finishing-on-pla
-
You can import SketchUp files (skp-files or something, not STL or OBJ) directly into the freeware 3D-editor DesignSpark Mechanical ("DSM").
If you do that with fonts created in SketchUp (with the characters as vectors only, or as surfaces, not yet 3D-models) and then you try to make 3D-models out of them in DSM, you will see that in a lot of characters the vectors-ends do not match: there are tiny gaps. It are not closed shapes. In DesignSpark Mechanical this can be repaired manually by zooming in on the offending characters and vector-ends, and manually adding small vectors to close the gaps. Then these characters can be "3-Dimensionalised" as expected. But sometimes this does not work well.
The same happens with walls and other design parts in SketchUp.
I guess that if SketchUp makes an STL- or OBJ-file out of its defective vector- or surface-elements, that STL- or OBJ-file also contains these defects. Which is guaranteed to cause problems in 3D-printing.
I am not sure if you can repair those defects in SketchUp itself in the source files.
It would be better to do the design in DSM (or another editor) that is designed for solid 3D-modeling and printing.
As far as I understood, SketchUp was never designed for solid (watertight) 3D-modeling: it was only meant to create superficial 3D-looking models to represent buildings in Google Earth and Maps. In which case gaps in the models don't matter at all, as they are only needed for crude visual representations on-screen.
But now Google has developed another way to turn its photos directly into 3D-models, so it didn't need SketchUp anymore and sold it.
For example, in Google Maps in your browser, if you zoom in to the port of Antwerp (Belgium, Europe), and then use the Control-key and mouse buttons, you can rotate and fly around in the 3D-world. The 3D-effect is not perfect, but still impressive and fairly accurate. Similarly, you can also fly through the mountains in the Alps in 3D.
-
As far as I understood, the basic reason is that ABS requires a very hot temperature to melt and fuse well. The new layers tend to cool down too fast, thus without proper fusing to the previous layers. You need to keep the temperature of the old layers way up too: close the front of the printer, switch off cooling fans, high bed temp, and don't print too fast. In combination with the huge shrinkage of ABS, this causes the splitting, or the coming loose from the build platform. ABS is good for injection moulding, but not so good for 3D-printing.
If PLA would not be suitable for your model, you might consider materials that have less shrinking and warping than ABS, but that print more like PLA. For example NGEN, PET, PETG? Try one spool, or even a sample, before buying a lot.
-
I store all filaments in huge sealed boxes with a big (!) bag of disseccant, so they are dried and then stay dry. I use the sort of disseccant that is sold in car shops to dry car interiors and prevent condensation on the windows. They have a blue indicator that turns pink when the bag needs to be reheated in a microwave or other oven.
For nylon, you should also put the spool in a sealed box with disseccant *while printing*: otherwise it may absorb too much moisture in only a few hours. This may require some bricolage: you need to make a spool holder, a small exit hole, and a holder for the disseccant in the sealed box. Polypropylene food storage boxes do exist in suitable sizes.
-
Another option would be to take a ruler and measure it. And then draw the thing in whatever CAD program you want.
-
I have printed 10mm x 10mm x 10mm test cubes in that way: just set the wall thickness to more than 5mm indeed (in my case), and it spirals layer per layer. Then it prints only walls without infill, until the whole object is filled with wall. If your object would be larger than 10mm diameter, adjust wall thickness accordingly.
However, the center of the spiral does not come out well: the nozzle stays too long in that little area, develops too much heat, and it has to slow down too much to go around the tight corners. This results in ugly blobs and overextrusion in the center of the spiral. The rest is okay. Printing multiple test blocks at the same time helps a bit of course, but does not eliminate the effect.
So I prefer the diagonal infill pattern.
-
I tried all sorts of keychains in PLA, PET and NGEN, just to get an idea of the durability of these materials in daily use. PLA gets harder, more brittle and breaks under too much stress. PET deforms and then also breaks. NGEN snaps also. They seem to be about equally strong in normal use.
If you want the part to last, you need to design in methods of stress relief. For example, give all parts enough room to move around freely, so they do not get stuck in weird angles and break, when you sit on it, or when you pull it out of your pocket with too much force.
I haven't tried flexible materials like nylon yet.
-
Have you tried this? Remove the bowden at the print head. Heat up the nozzle manually to the normal print temp. Check if the little back fan works indeed. And then manually feed filament into the teflon coupler and nozzle directly, without bowden tube. Keep an eye on the temperature. Does that go well, and keep going well? Or does that cause jerks, stops, irregular friction, or whatever? If you can feed 10cm of filament through smoothly in this way, that should exclude the nozzle and coupler from the cause, I think.
And what happens if you disconnect the bowden tube at both ends, and manually feed filament through it? Does it slide smoothly? And then connect it again at the nozzle, and feed filament manually all the way from the back to the nozzle? Does that feed well and extrude smoothly?
I could also imagine that some flexible filaments do get too sticky and slide very difficult when warm. Like soft rubber or siliconerubber which does not slide against a surface smoothly, but always jerks.
I have bought a separate drill clamp (I don't know the correct name in English) to mount my drill into, so I can get a good grip on it.
general forum feedback
in UltiMaker.com
Posted
I also have this sometimes: the auto-logout function kicking in after one hour or so. The best thing to do is write the message in Notepad++, and then copy and paste it into the browser. But of course, I usually do not follow my own advice...
It would be good if we could set the default auto-logout time in our preferences. I would probably set it to 4 hours. And then I would lose the next reply at 4:05h.