Jump to content

yellowshark

Dormant
  • Posts

    1,759
  • Joined

  • Last visited

  • Days Won

    6

Everything posted by yellowshark

  1. Whatever your problem is/was I agree with cloakfiend, your extruder temp is way way too high. Mind you this is always a tricky one, whilst I agree with his temp +/- 5 for 0.06, I would consider his temp for for .2 at 50mm/s to be too high, I would use 205-210 normally.
  2. I have not read the link "this" but any circle will always give you greater inaccuracies than a straight line; it is inherent in the way a 3D printer prints a circular geometry. And yes as mentioned other print settings will affect the percentage error. The only thing you can do is to increase the design diameter until the desired diameter is achieved. Irritating yes but once you have done it a few times you should find you will get it right within three or four attempts - and reduce the height in Cura until you have got the design dims to where they need to be
  3. On the basis that the failing filament is stopping in the same place every time, I doubt that you have a problem with the filament. Similarly if another PLA filament and ABS are OK, I doubt you have a problem with the drive/feed system. I assume you used exactly the same settings for all three filaments, apart from increasing the temp. for ABS? Although, I am guessing here as I do not have an Ultimaker and do not have those "smart" options with "standard" settings behind them, but if you say you are using ABS I guess that that the standard settings for ABS change the temperature. But if you ARE using the same settings for both PLA filaments then it does not make a lot of sense. I suspect your problem is your layer height. You do not say what your "initial layer thickness" is but I guess that is 0.3mm I do not think anybody uses 0.6mm layer height, with 0.3mm being the normal maximum; the help tip on the Cura box for layer height says that 0.25mm is the maximum - my printer is fine with 0.3mm. Either Cura sees 0.6 as the layer height and just goes into a terminal abort "does not compute" mood so that no gcode is generated after the initial layer. Or, as you are using "default temperature" I suspect this is not hot enough to melt the filament enough to allow it to push through double the plastic required for 0.6mm and it just clogs up in the nozzle. But as the printer stops, rather than carries on printing with no filament being extruded, I suspect it is the first point. Of course what is confusing about your post is that it works OK with a different PLA filament - confusing if, of course, you are running that with a 0.6 layer height. It would be worth using a layer height of 0.25mm and see what happens.
  4. Yes, they were glued. The “glass” was made to the same dims. as the window frame and simply glued to the back of the frames. For the square windows, the white walls around the window apertures have a small recess built in on both sides and top & bottom, so the window frame was inserted from the interior side and glued to the recess; this stopped any gaps appearing between the window frame and the wall. I use Revell’s Contacta professional glue, chosen on the basis that Revell has been making plastic kits for decades! It needs the parts to be held in place for 5 minutes or so and seems to cure within a couple of hours or so.
  5. Hi, well you wanted to see some pics. I am not sure exactly what you wanted to see but here are some. If they generate questions then feel free to ask! This was printed some considerable time ago when I was very much a beginner. The first pic shows the difference! On the left is the original patio area and on the right is the same part that I printed a couple of days ago for comparison. I am sure if I reprinted the apartment block again I would achieve a much better overall quality. [media=9516][/media] The next two pics show the completed apartment block; along with a normal sized mouse to give a feeling of size/scale. The final pic shows the pieces ready for final gluing; there are four parts, plus the roof and patio parts, not shown. The base (A) was printed separately as part of the effort to avoid really long print times. On the left (B) is the single story part of the building and as this is easy to glue to the rest of the model I printed it separately. Then as you can see, the rest of the apartment block was printed as two parts – again to avoid a lengthy print time. The front wall alone was originally printed as seven or eight separate parts and was probably the main area I was having problems with gluing it together satisfactorily. Another point to mention is the roof line. I do not know what the architectural term is but you can see (I hope) that there is a stepped lip running all the way around. Originally I printed this as one piece, intending to stick it to the roof line on final assembly. But again this was a part that I found difficult to glue satisfactorily. If I remember rightly I was not sure, with the overhang, how it would print vertically, so printing it separately meant I could print it upside down. In the final print I printed it integrated with the walls, i.e. right way up with the overhang. I should say that all the walls in the final version were printed vertically. When I originally printed them as a variety of pieces I had considered printing horizontally – it would have reduced the print time considerably. But I did not like the idea of the walls have diagonal lines (from the infill) rather than horizontal lines from the layers, so printing vertically was an early decision I made and which I stuck to. If I were to print it again I think I would print parts C & D as one piece (in the region of 15-20 hours) to avoid the join lines; I would also probably either print it upside down to get a better roof line, or try printing the stepped roof line on its own again and glue it to the roof which should be a lot easier.
  6. Ok I have posted a new thread, Architecture, with pics and comments
  7. Yup agree with George - if a specific colour is not a requirement I tend to use various blues, reds/burgundies, greys and ColorFabbb Dutch orange has always delivered excellent results for me - which is not to say there are no other good colours, just my personal experience. I will try and post some photos tomorrow
  8. I doubt very much it is under-extrusion; it is almost certainly printing a flat surface across a void. As a test I suggest you use the "cut off bottom" function on the advanced tab. I am a looking at your 2nd pic and what I assume is a shoulder that has the error. Using 1.2mm as your solid infill setting, I would "submerge" your model so you have about 3.5mm of the shoulder above the surface. Change your infill from 0% to 26% and run the print - sure stop it once the shoulder has printed. You should find you have a much better finish on the shoulder. A pic of the result would be useful, if it too does not print well
  9. Well as a general statement, no 30-40 mm/s is not too fast for .200 layer. Running as cool as possible will help you with the stringing on your bracelet. If I were printing something intricate like your bracelet then I would drop down to 20mm/s in search of a good quality finish, but I tend not to print those type of models so I have not done any testing on that subject. I do suspect that dropping to 20mm/s, irrespective of layer resolution, will improve your result - if you have the time
  10. Sorry this may be trying to teach your mother to suck eggs but if you are a slicing in Cura then the model will be coloured Yellow if it is located within the dimensions of the bed and coloured Grey/Blue if part of it is outside of those dimensions
  11. As a reference point the last architectural job I did was for a two story very modern apartment block containing four apartments at a scale of 1:100. It certainly was not four walls with a roof. It had a total print time, at .200 layer depth, of about 40 hours, in white. The floor area was around 300cm * 250cm, so just fitted onto my bed. At that size bed temp. is not enough to ensure a perfect 1st layer and corner adhesion, straight or curved. Personally I use extra strength hairspray but choice from a number of good adhesives is a personal thing. I too have had trouble with both colorFabb and Faberdashery "whites" but I have found Faberdashery Arctic White to be flawless. I decided to cut up the model into a number of pieces as I did not want to get 80% of the way through a 40 hour print and then have a failure plus there were several aspects to the architecture where I was not sure if they would print OK if I printed them in the orientation dictated by a single print. Big mistake, there was no way I could get a satisfactory gluing together of about ten pieces making up the walls. So I had to redo it, another 40 hours, having remodelled into three pieces. George's comments are all spot on, he knows his stuff. The only thing I would say is that for a surface area close to the size of the bed I use 65c and have gone to 70c. If you have that size and have not done it before, then I would also add - do not forget that your bed temp. sensor is probably measuring the bed, not the glass plate, so you need to check it with a laser thermometer. Also it takes longer for the perimeter areas to get up to the same temp. as the central area of the bed, of course depending on where the sensor is measuring, so you need to check this. If the perimeter areas are 50 - 55 when you start then that large surface make not stick.
  12. Not easy to absolutely sure from the photo but looking at the walls you do not seem to be suffering any under-extrusion. The first thing I would suggest is to try another filament. I once had a similar problem with a Faberdashery white - not Arctic white - with gaps between wall and infill. A different colour fixed it immediately. I have no experience in printing letters but I would be inclined to try it real slow, 20mm/s max - maybe use 100% infill, I suspect impact on time and material used would be minimal with that geometry.
  13. Great print qmaker, does not look like you need any help!!!!! I was put off by Meshmixer initially - BUT persevering I am comfortable with it now and it is a step up from Cura for supports, much better. You can manually add your own supports and delete auto generated supports if need be.
  14. I do not see layer depth mentioned; personally I would use .1mm, but that is a matter of taste, and print speed of 20mm/s. Fans full on and print bed temp. as low as you can - certainly reduce it below 60 after layer 1. I suspect your "under extrusion" could be coming from the print bed heat, that is a guess. As already said, extruder as cool as possible; certainly with settings above I would not exceed 200 but it depends on your printer and the make/colour of the PLA.
  15. The wall thickness may well be adding to your problem. It should be a multiple of your nozzle width - of course if your nozzle width is 0.3 rather than 0.4 then you are spot on
  16. LOL get your order in I am running out 8)
  17. You do not say what infill percentage you are using prior to the top layers - that has an impact. But if you are running less than say 40% I think you will need a lot more than .6mm top layers when using .100 layers
  18. I use the Arctic White all the time in 3mm, it prints really well and for me is the right colour white. I have tried ColorFabb too which is OK but Faberdashery AW does edge it for me. The other day I printed .300 layer at 50mm/s at 205-210c
  19. OMG for the first time since the summer I have just this minute received a bunch of email notifications! Someone has been playing
  20. I have a non UM printer and have used the usb connection for the past two years without a problem. I have no idea whether it makes a difference but whilst I slice with Cura I run my printer with Reptier Host software.
  21. Yup it is normally a dirty nozzle but the filament should lay down OK once it has settled; i.e. the first few centimetres may not stick; but some judicious splatting and scrapping with a long screw driver should see you off and running.
  22. Also do not forget that the temp. sensor is measuring the bed not the glass, there w ill be delay in the glass reaching the indicated temp. On my printer when the temp. reading hits 60 the glass measures circa 45. Also there is a further delay between the centre of the bed hitting X and the rest of the bed hitting X - initial temp difference can hit 10; of course not an issue if the model has a small base.
  23. I pretty much always use Fab. On first use I take it from the packaging and unwind/rewind it, making similar sized loops to how it was delivered. It then lays on the floor whilst being used. Personally I find that 50metre lengths is the largest I like to work with; recoiling 100meters is a bit tricky, but that may just be me, probably easier with two pairs of hands. It can occasionally twist up again(I think that is a function of how well you have recoiled it) but if you lay it on the floor so that the top loops are spread horizontally towards the printer you should spot it before it becomes a problem. Having said all that I still prefer this physical method to using a spool of filament from other manufacturers.
  24. Not sure about that one; we use the air on our remote controlled touring cars and the idea is to get the dry crap off the mechanicals not on to them!
  25. Lol well I have a 3mm bearing on my filament and it has never caused me a problem, apart from preventing blockages from bad filament
×
×
  • Create New...