Jump to content
Ultimaker Community of 3D Printing Experts


  • Content Count

  • Joined

  • Last visited


Everything posted by mrjohnk

  1. I have good luck printing ABS on borosilicate glass that is covered in super wide polyimide tape (Kapton). I just clip that right to the top of my heated bed with some large metal binder clips. Also, be sure to setup a "skirt" around the part to help it both stay attached and keep the air from flowing around the adhesion point. I use six lines on the skirt and that works well for me. -John
  2. To build upon what Owen said, you shoud not use PID/PWM controls on a mechanical relay. If you have a solid state relay or MOSFET switched power, PID/PWM can be used as those can take the rapid switching. The mechanical relay cannot. The Marlin firmware can control this via the setting documented below from the Marlin configuration file: // Bed Temperature Control // Select PID or bang-bang with PIDTEMPBED. If bang-bang, BED_LIMIT_SWITCHING will enable hysteresis // // uncomment this to enable PID on the bed. It uses the same ferquency PWM as the extruder. // If your PID_dT above is
  3. Use what temperature works. I'm printing ABS on mine at 255 degrees. I've heard that as long as you keep it below 260 degrees, you should be safe. As for printing PLA at that high, that is pretty well unheard of. I've printed PLA from 192 to 225. It may well be something with the material and less with the printer.
  4. I am printing in stand-alone mode from the ulticontroller as this one sits in my basement with no computer in sight. Power is fed from a dedicate 24V - 400W supply. Both the nozzle heater and heated bed are using PID control. I noticed the LED on the headed bed continued to flash even after the stoppage. I'm inclined to say that at least the Arduino and main circuit board were still functional enough to provide the PID signal to the heated bed. It is suspect that this happened right as the transition from the second to third part. So, I'm thinking something in the gcode (Cura) or the ulti
  5. My print job stopped at the end of the second piece out of six. Where the z-stage should have gone down and nozzle retract back to 0,0, instead, the nozzle parked itself on top of the part after the final layer was done and melted a spot in the top. It printed the first part perfectly. I used the Cura project planner to setup this print job. Any idea why the printer suddenly stopped and just parked right on top of the recently finished part instead of going on to the next one? I'm not sure if it is a Cura, Marlin or profile problem. I updated Marlin to the latest version yesterday. Cura
  6. Print looking good now. Updated to latest Marlin and changed settings. Thanks for the help. Link to pic below. Also changed the order of print to Perimeter > Infill > Loops in order to drop any potential blobs in the infill instead. http://www.flickr.com/photos/8465641@N07/8581931440/in/photostream [profile] layer_height = 0.2 nozzle_size = 0.4 wall_thickness = .4 solid_layer_thickness = 0.4 fill_density = 20 skirt_line_count = 6 skirt_gap = 0 print_speed = 50 print_temperature = 255 support = Everywhere enable_raft = False filament_diameter = 2.97 filament_density = .
  7. I figured out that I used the wrong syntax by eliminating the "S" in the passed parameter. Corrected syntax is working properly: M303 S250 John
  8. Since I'm getting ready to update my firmware to the latest Marlin version, I wanted to go ahead and do another PID autotune on the hotend so that I can get a better performance. I'm running a 24V power supply and printing ABS. Using default PID settings work, but the hotend overshoots the 255 degree temperature upon initial heat-up to about 265, but holds good at 255 after that. The goal is to reduce the initial overshoot. Using the pronterface GUI, below is the output I received. Any idea what caused it to fail the autotune? Thanks. John SENDING:M303 250 PID Autotune start ok T:46.
  9. Thanks for all the great suggestions. I'll work on getting the latest fork of Marlin compiled and uploaded to the board to start with for more realistic retraction results. I do indeed already have the v3 bolt, so just need the roller kit. I may end up printing that and try to make one per Bertho's instructions. Next print will be either smaller layers or slower speed. I may even enable the "skin" feature in Cura to still have fat layers for infill and nice skinny layers for the outter edge where the blobs are appearing. I'll just have to play with that and see how it goes. If those adj
  10. Currently, I have a stock v2 hot-end/nozzle. I also I'm running a custom compiled version of Marlin. I had downloaded the source a couple of months ago when I needed to setup PID control on my heated bed. Other than the bed settings, it is a default image from the source code. If only the newest version fixes the retraction issue, I can download and compile a new version. The current version states "THIS IS RELEASE CANDIDATE 2 FOR MARLIN 1.0.0". I had been running retraction at 70 mm/s a couple of days ago, but was still having the blobbing issue, then changed it to 45 mm/s with no chang
  11. Printing at 95 mm/s and travel at 150 mm/s.
  12. I've been having some issues with blobs and retraction. It appears that blobs are forming before a jump. The nozzle is still for a split second, blob forms, then nozzle moves. Slicing with Cura 13.07 on Win7. In addition, prints with high frequency of jump requirements (and retracting each time) are not completing as they end up chewing a spot in the filament eventually, even though it will print for over an hour before chewing into the filament. I have the standard delrin rub-block extruder, not the upgraded one with the roller bearing. Added a horseshoe spacer and zip to the bowden tub
  13. All was well with my UM, several layers of the current job put down and a few hours to go. So, I turned out the light and left my UM to do its job. I came back a couple of hours later to find it asleep on the job! Knowing that I had seen this symptom before, I found this thread and checked my connections. Sure enough, the thermocouple plug on top of the hot end assembly was about half out of the receptacle. I pushed the plug back in and used a small blunt tool to also push each of the terminals in the plug all the way down. It became clear that I probably had not put the stress relief po
  14. I have a few pictures of the bed in the following Flickr set. I don't have any pictures of the relay or power supply. In the pictures, you can see the pieces of aluminum that make up the adapter plates on the sides of the heater board. They simply extend the width of the heater and provide holes to mate the heater to the z-stage. One piece of aluminum is slightly longer to protrude past the heater and provides a place for the wires to get strain relief. You can see this where there is a zip tie. It holds up the wires leading to the heater on the right side. The thermister path can be se
  15. Yes, you will end up with two power supplies. One for the Ultimaker and one for the MK2 heater. The relay will be controlled by the main board. The power supply that came with the Ultimaker is 19V and the relay linked below works well at this voltage. http://www.findchips.com/avail?part=ALFG1PF18 Below is a small portion of the text I also posted on this subject in the Google Ultimaker group: As for wiring, It is fairly simple. The heated bed port on the main board (2 wires) connects to the coil side of the relay (2 connections). The positive of the 24V power supply goes to one side
  16. If you use a relay to control the power to the heater, there will be no noticeable power draw and it would allow the system to control the heater. The draw is about 70mA, so hardly anything. The relay is also very inexpensive and fairly easy to implement. -John
  17. Did you happen to install the 4.7K resistor to the board? It is required in order to read the thermistor correctly. -John
  18. I've detailed my setup on the Google group, but essentially, here is a summary of the high points: MK1 Heater 24V, 400W power supply bought from Ebay 14G or thicker wire, high temperature silicone insulated 18V relay connected to the heated bed port on the UM control board 100K EPCOS thermister from Ultimachine (very inexpensive) 4.7K through hole resistor (see wiki about this) healthy amount of polyimide tape (AKA Kapton tape) borosilicate glass top, covered in polyimide tape small bulldog clips attaching glass to heated bed I used the tape to attach the thermister to the bottom of
  19. After seeing the warning, I did back the overlap back down to 12% to see if that fixes it, but I didn't have any luck correcting the issue with that.
  20. I troubleshooting an issue regarding the toolpath generated does not call for infill on two layers in the middle of my model, but it does do a perimeter run on each. After two layers of no infill, then the next infill comes along and kinda droops and sags as it is printing in a relative void and it doesn't' really attach well to anything. I tried to use Netfabb to repair the part, if anything was wrong with it, but no help. I have attached the STL model file and the Cura profile I'm using. It is layers 16, 17 & 18 that I'm concerned with. The first two do an outline, then layer 18 sta
  21. A few years back, I bought a 100 piece drill bit set from a guy on Ebay. I believe it was the same ones that Harbor Freight sells. It has been the best investment. Just about any size you want right at your finger tips for not a whole lot of money. I also got some pin vices from http://www.widgetsupply.com/. They have lots of wierd things like this that are perfect for this sort of thing from dremel tools to dentistry. I think most of it is overstock stuff, so it is really inexpensive. -John
  22. I have a plate of glass that sits on top of the heater that I can remove. It just has four bulldog clips that hold it tight to the MK1 heater. It makes it easy to remove it, spray it and put it back. While the hairspray did work for one part, it wasn't all that reliable for subsequent parts. I'm back to using the yellow polyimide tape again. -John
  23. I recently bought my Ultimaker a few weeks back. I have since equipped it with the heated bed and have been printing ABS in the last week. ABS certainly has its own properties that are different than PLA, but once you get used to it, it is a nice material to print in. I wouldn't mind to share a working ABS Cura profile, but mileage may vary. I'm finding out that each plastic supplier has slight differences in the products they sell. Even different colors from the same vendor can need different settings for speed, temperature and other factors. Today, I removed my Kapton tape (or polyim
  24. While I'm no expert, I think there are a variety of things that can cause this. Things I would be looking at: Size of filament setting matching the actual size packing density minimum layer time minimum feed rate Fan on or off speed of printing layer size. Looks like pretty big layers. -John
  25. I've found that using the dwindle setting on something like this helps too. It slows the print down quite a bit as the layer size gets smaller. It would give the fan more time to cool the layers. I suppose it is similar to the minimum layer time too, but works well for me. I also have the fan set lower than 100% and I'm using a range, so as the printing slows, the fan picks-up faster.
  • Create New...