Jump to content

gr5

Moderator
  • Posts

    17,513
  • Joined

  • Last visited

  • Days Won

    372

Everything posted by gr5

  1. Oh - you could test the astigmatism versus grain issues by rotating the object you are cutting to see if it moves (then it's grain) or rotating the laser (then it's astigmatism if the problem moves).
  2. The fan-turn on command puts it at 100% briefly to get it to start spinning before starting the PWM phase. This is a flag in Configuarion.h which you can disable. Regarding X versus Y directional cuts being different depths. It could be grain related but it could also be a focus issue where the point of light is focused (thinner) in one direction and blurrier (wider) in the other. Maybe? I don't know if that's a common problem with lasers. An astigmatism.
  3. @foehnsturm - that's beautiful but it looks so "milling machine" constrained! Now you have a 3d printer and can make those fins be non-flat!
  4. It's fine! Stop worrying. The NFC chip on the core will not even know it is connected to anything if you leave the middle of the 5 connectors unconnected.
  5. This was with bondtech. I was asked to test out the bondtech - if I wasn't always asked to test feeders I'd still have the original black feeder on there which I think worked good enough for me.
  6. Printed just fine. I printed your gcode. This is ngen at 230C bed at 80C. sorry my pictures aren't as clear as yours. This is a small part. Click on second/middle image to see more details. It does look like possibly a little underextrusion (dashes of gap maybe?). Hard to tell.
  7. okay - will print this with ngen now...
  8. But this is my favorite print on the UM3. PLA with dissolvable support. The blades are paper thin - very delicate. I tried to make this with PLA support years ago. My daughter spent an hour with a dremmel carefully removing the support under the tips of the blades at the outer edges. it came out pretty good but this print with the UM3 was much less labor! Just dropped it in a bucket of water over night. The blades look equally good from both sides. I can't tell which is which by looking at the blades only. I have to find the surface that touched glass to know which way it was printed. It's so delicate I have already chipped it a few times - when you blow down on it from above it spins and then "hovers" and moves quickly sideways and hits something every time. sometimes the blade gets a chip taken out. My current plan is to use my UM2 for 80% of prints but we'll see. Time will tell.
  9. These really not fair comparisons because they aren't PLA. But here are some prints showing overhang is at least reasonable without doing anything special (didn't tweak fans or temperature or print speed): In the first white/green print it is made from ABS. The red arrow shows good overhang. The black arrow shows that the green color dripped a bit at the tip of south america. In this second print which is upside down here it was made from Ultimaker CPE (aka UPET). The black arrow is showing extreme overhang angle - probably 70 degrees from vertical and it printed fine. This is the underside of the print where I didn't care about quality.
  10. I only printed one single nozzle print so far and it looks quite good but it's a mechanical part that doesn't need to look good. I haven't tried to print something extra fine quality yet. The overhang ability is similar to um2 and the quality appears similar to um2 to my eyes. But I haven't tried any fair comparisons yet (e.g. at least print the same part with same filament).
  11. Oh - also I think the bondtech recommended steps/mm might be a bit low. Maybe try increasing flow by 10% or increasing steps/mm by 10% over what bondtech recommends.
  12. I forgot you were printing nGen and not PLA when I said 60mm/sec at .2mm was too fast. I don't know how different nGen viscosity is versus PLA so it could be that even 30mm/sec at your printing temp is too fast. Maybe you need to increase the temp. I looked at your gcode file with the website: gcode.ws and I can see that your outer wall at 15mm/sec is 2X slower than inner walls. I recommend you print all at one speed to get more consistent quality. But you have to decide on strength versus speed tradeoffs. I'm too lazy to get off my butt and get the UM2 out of the basement. Lately I've been using only my UM3 and my um2go. But I need to print something else on the UM2 this weekend so maybe I'll print your part also. Using nGen. What temperature are you printing nGen? Unrelated - note that colorfab recommends 50% fan speed on ultimaker 2. I hope you are doing that. If not you may have very weak layer bonding which is difficult to notice unless you purposely stress your part until it breaks. If it breaks along layer lines then you need to lower the fan and/or put a front cover on your printer.
  13. The A20 connects to the ultimarlin board through (I believe) that very same USB interface (I could be wrong). Now if you want to send just a single gcode without sending an entire gcode file you have to use the REST interfact I think. Anyway you are definitely correct that the USB is gone. Now if you want to print something just send it to the printer over the network. If you want to update firmware the printer can update itself.
  14. It could be. You would expect the head to move a smaller distance every 8 or 16 steps and so the laser would rest longer on that spot. It's easy to ground this pin but also easy to destroy your stepper driver. Only do this if you or a friend are experienced with working with and soldering SMT. Someone on the forums destroyed their stepper driver already when they attempted this change. Cohen at Ultimaker who chose to set these boards this way felt that the potential error was smaller in importance to the additional torque you get but I know he wanted to do experiments and measure the change in torque at each of the 16 microsteps in both modes of the stepper driver. Not necessarily a simple experiment.
  15. I have no problem with "autoslice". It's the "pre-slice" aka "preparing to slice" that kills the user interface for me (for 30 seconds with some models). I keep my house cool so all extra CPU energy ends up keeping my tiny office warm and allows me to heat the rest of the house less. So at least half the year it doesn't waste any energy for me.
  16. The "clip" system has an orange thing that you press quite hard to get the heater wires in and out. You have to remove the wood panel to get at those orange things that you press. Anyway - it looks like your heater is shorted. Lets get terminology correct: temp sensor - plugs in to left side of photo above. It has a resistance of around 108 ohms at room temperature. This is working fine - don't mess with it! heater - plus into right side of photo above. It has a resistance of around 24 ohms normally. This is not working for you. Wires probably shorted in the photo but can't see it as they are covered. I know your temp sensor is working because in the video it said 0/25 meaning it is reading 25C which is perfect. Anyway I think your heater wires are touching each other. You should take them out and put them back in and try again. Yank on them to make sure they are in there good and won't come out easily. Also make sure the metal of the wires is touching no other wires or no metal on the circuit board. If it all looks good and still reboots when you turn heater on then I would get a new heater. I mean maybe first find a friend who has a multimeter and make sure it is at 0 ohms as I suspect. It should be around 10 to 25 ohms depending how many watts of a heater you bought. Maybe you got the wrong type of heater? for example a 35W heater designed for 12V will be 140W with 24V power supply and do exactly what is shown in the video. I recommend you get heaters from 3dsolex. I think there might be a reseller in UK.
  17. What sander said. But if the glass is tilted by 0.1mm across the cube you are printing then 5 sides will be square but the 6th side - the bottom will be tilted by that .1mm of error that was compensated for in auto leveling. It's really not a problem since you should have .1mm accuracy across the entire bed with manual leveling and across a small cube the error should be much smaller. But a cube that completely fills the machine would be slightly out. Anyway I'm happy with my manual leveling techniques and didn't care about auto leveling at the time. I will try it some day.
  18. And there's much more. I hear great things about tinkercad (another web service) and moment of inspiration(MOI) and there are hundreds more great things out there. Morphi is fantastic - designed for kids but like tinkercad is quite full of features.
  19. Probably onShape. It's free if your designs are open source. It's by the same guy who did solidworks so if you know one it's easy to learn the other. I don't like to use web services as you never know how long a company will last before it goes out of business but onShape is extremely well funded right now and you can export your design as step files. I personally use DSM (design spark mechanical). It's free and installs on my pc so I don't need internet to use it. It's incredibly advanced and easy to use compared to say sketchup. don't use sketchup as sketchup is great for visual modeling but not for actual 3d objects. Quite bad for 3d modeling real things to be printed.
  20. The problem seems clear to me. Cura has more than one speed. It has separate speeds for outer shell, inner shell, and infill. Somehow your inner shell seems to be printing too fast as I can see the outer two shells are touching nicely and the infill looks solid but the inner two shells are probably both a little underextruded (printed too fast). So go into the more advanced settings in cura. 15mm/sec is crazy slow. Try this: 35mm/sec (all printing speeds - travel speed should be 150-300mm/sec), .2mm layer height, 220C. ALTERNATIVE EXPLANATION: make sure your shell width is an integral multiple of your nozzle size. so if nozzle size is .4 and you want three shells than shell width must be 1.2. If you make it 1.3 you can get underextruded shell.
  21. policies keep people from doing their jobs. I don't like policies unless they are just guidelines and it's okay to ignore them if you have thought it through.
  22. Yes there is still a Z switch at the bottom identical to the way the UM2 homes. I didn't enable the automatic leveling on my machine. I'm very good at manual leveling so I decided to skip it. There are cons to using automatic leveling. For example it is moving the Z motor while printing the first 5 or 10 layers (1mm or so) with less and less compensation as you go. So if you have quite a bit of tilt to your bed it will show up in the final print where the bottom will not be flat but tilted. So I don't want to use auto leveling to fix a tilt to the bed since I can fix it with the 3 screws but I *do* want to use it to fix any wave-shape to the bed - like if it's saddle shaped or cylindrical. But my understanding (not certain) is that it only does tilt compensation - not 9 point compensation (or more) like some other printers. Delta printers need at least 16 point leveling because of the way errors creep in. UM3 shouldn't need this. I'd rather just bend the entire bed until a straight edge says it is flat.
  23. Mine is not yet damaged The part that is incredibly delicate is the part wth that thin neck in the top center with the long threads on one end and the hex head on the other. Anyway - to answer the question, yes there is some TFM style expensive teflon in there as you can see near the spring. As Daid said it goes inside the aluminum part with fins. Actually it goes in the threaded steel part (top center in picture) that then goes in the aluminum part. As someone else mentioned it shouldn't get above 100C with the nozzle at 250C. Even if it does get soft there is no significant pressure on it like there is on the UM2. So it should last "forever" unless you start printing at 350C.
  24. REMOVING EVEN JUST THE FIRST SCREW THAT HOLDS THE TEMP SENSOR IN PLACE CAN EASILY DESTROY YOUR CORE. If you don't understand why, don't try it! Even UM employees have broken many of these - even when being careful! If you don't understand why, don't try it! Even UM employees have broken many of these - even when being careful! The "neck" of the core is the same thickness of two sheets of paper.
  25. @arjan - I looked at the wiring very carefully in the Core and the 3 electrical components are isolated. so the heater uses 2 wires which don't connect to the other 2 components. Same for the temp sensor. So you can simply decide to not connect to the eeprom and simply never power it up if you mount a core on the UM2.
×
×
  • Create New...