Jump to content
Ultimaker Community of 3D Printing Experts

mrjohnk

Member
  • Content Count

    41
  • Joined

  • Last visited

    Never

Everything posted by mrjohnk

  1. I finally got done building up my machine. It has taken about 5 evenings. In addition to the standard issue UM, I added a 400W power supply and heated bed with a glass top. Generally, I followed 3DTOPO's example http://www.thingiverse.com/thing:30655 for parts to use, but used borosilicate glass instead of aluminum. I've covered the glass with the kapton tape, installed the 4.7K resistor and hooked up both the thermister and power for the heated bed. I had tested all the standard parts yesterday using ReplicatorG on the included power supply before adding the new heated bed additions. All motors and switches worked. The heater came up to temp in the extruded. All was well. After adding the heated bed and new power supply, I tested again with ReplicatorG and was still able to jog around and everything was still good. I knew I needed to update Marlin to support the heated bed, so I got that together with the online Marlin builder and uploaded using the batch file. After loading that, it of course broke communications with ReplicatorG since the baud rate was faster. I lowered the Marlin baud rate to 115.2K to get ReplicatorG working again, however, this time, it didn't work. I could connect to the Arduino and generally send commands, but the machine simply didn't respond. I figured something was broke between ReplicatorG and the new version of Marlin I had loaded. Moving to the next step, I loaded up Cura and ran through the setup check which went fairly well, except testing the extruder. The heater block was only luke warm to the touch and the extrude kept running while testing, but just chewing a hole in the filament since there was just a cold extruder it was pumping into. Moving on, I loaded up PrintRun and noticed that it too could connect, however, not actually get the machine to do anything with the controls. At this point, I began to suspect the power supply. Upon inspection of the power supply, the LED was on, but dim and it was emitting a slight buzzing sound. I checked the voltage output and it was at only 4.7V. I disconnected it, along with the heated bed and went back to the power supply that came with the UM and was able to get PrintRun to operate the machine, so I was sure the 400W power supply was toast. So, I guess, after that long winded explanation, I'm looking to see if I did something wrong or the power supply I bought was just not any good. Thanks. -John
  2. mrjohnk

    Retraction and Blobs

    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 tube connector on the extruder drive to isolate movement here, but still getting blobs. Normal prints without the high-frequency retractions are just fine though. Retraction is set to 45 mm/s @ 4.5mm on jumps more than 2mm. ABS printing at 255 degrees. Picture linked below shows both blobs and the eventual failure of the print due to lack of extrusion. Thoughts on what I can do to help? http://www.flickr.com/photos/8465641@N07/8578127085/in/photostream/
  3. 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
  4. 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 the default, and correct for your hardware/configuration, that means 7.689Hz, // which is fine for driving a square wave into a resistive load and does not significantly impact you FET heating. // This also works fine on a Fotek SSR-10DA Solid State Relay into a 250W heater. // If your configuration is significantly different than this and you don't understand the issues involved, you proabaly // shouldn't use bed PID until someone else verifies your hardware works. // If this is enabled, find your own PID constants below. #define PIDTEMPBED
  5. mrjohnk

    my prints are now very brittle?

    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.
  6. mrjohnk

    Unexpected Stop

    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 ulticontroller had an issue. In this room, there are not any other running electrical devices as I unplug everything when it isn't in use and the light was coming through the windows. John
  7. mrjohnk

    Unexpected Stop

    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 is 13.03. I'm using a profile that worked properly yesterday for printing a single item. I wanted to upload the gcode file here for reference, however, I'm unsure how to upload and share that on the forum. Where shoud I look to troubleshoot something like this? Thanks. John
  8. mrjohnk

    Retraction and Blobs

    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 = .92 machine_center_x = 100 machine_center_y = 100 retraction_min_travel = 2 retraction_speed = 30 retraction_amount = 3 retraction_extra = 0 travel_speed = 150 max_z_speed = 3.0 bottom_layer_speed = 40 cool_min_layer_time = 15 fan_enabled = True bottom_thickness = 0 enable_skin = True model_scale = 1.0 model_rotate_base = 0 flip_x = False flip_y = False flip_z = False swap_xz = False swap_yz = False extra_base_wall_thickness = 0.0 cool_min_feedrate = 20 fan_layer = 1 fan_speed = 17 fan_speed_max = 75 raft_margin = 5 raft_base_material_amount = 100 raft_interface_material_amount = 100 support_rate = 25 support_distance = 0.5 infill_type = Line solid_top = True fill_overlap = 12 bridge_speed = 100 sequence = Perimeter > Infill > Loops force_first_layer_sequence = False joris = False retract_on_jumps_only = True enable_dwindle = False dwindle_pent_up_volume = 0.4 dwindle_slowdown_volume = 5.0 retraction_enable = True plugin_config = hop_on_move = False model_matrix = 1.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,1.0 object_sink = 0.0
  9. mrjohnk

    PID Autotune

    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
  10. mrjohnk

    PID Autotune

    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:46.42 @:127 ok T:46.42 @:127 ok T:46.91 @:127 ok T:48.34 @:127 ok T:50.05 @:127 ok T:52.61 @:127 ok T:55.66 @:127 ok T:59.57 @:127 ok T:63.66 @:127 ok T:68.27 @:127 ok T:73.21 @:127 ok T:78.58 @:127 ok T:83.89 @:127 ok T:89.32 @:127 ok T:95.21 @:127 ok T:100.92 @:127 ok T:106.60 @:127 ok T:112.30 @:127 ok T:118.32 @:127 ok T:124.11 @:127 ok T:129.82 @:127 ok T:135.44 @:127 ok T:141.42 @:127 ok T:146.94 @:127 ok T:152.31 @:0 ok T:157.96 @:0 ok T:162.78 @:0 ok T:166.99 @:0 PID Autotune failed! Temperature to high
  11. mrjohnk

    Retraction and Blobs

    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 adjustments don't cut it, I'll start hacking away at the temperature a little at a time to see what I can get away with. John
  12. mrjohnk

    Retraction and Blobs

    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 change perhaps this version of Marlin still has the limits. As you can see below, I am not using "hop on move", but I can give that a try. @gr5, I can try a lower temperature and I agree that lower may be good if I can get away with it. I ended up at this temperature originally to offset layer adhesion issues where the model would split horizontally due to poor layer adhesion. Thanks for the help. John =========== Full Cura Profile Below ======== [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 = 95 print_temperature = 255 support = Everywhere enable_raft = False filament_diameter = 2.97 filament_density = .85 machine_center_x = 100 machine_center_y = 100 retraction_min_travel = 2 retraction_speed = 45 retraction_amount = 4.5 retraction_extra = 0 travel_speed = 150 max_z_speed = 3.0 bottom_layer_speed = 40 cool_min_layer_time = 15 fan_enabled = True bottom_thickness = 0 enable_skin = False model_scale = 1.0 model_rotate_base = 0 flip_x = False flip_y = False flip_z = False swap_xz = False swap_yz = False extra_base_wall_thickness = 0.0 cool_min_feedrate = 20 fan_layer = 1 fan_speed = 17 fan_speed_max = 75 raft_margin = 5 raft_base_material_amount = 100 raft_interface_material_amount = 100 support_rate = 25 support_distance = 0.5 infill_type = Line solid_top = True fill_overlap = 12 bridge_speed = 100 sequence = Loops > Infill > Perimeter force_first_layer_sequence = False joris = False retract_on_jumps_only = True enable_dwindle = False dwindle_pent_up_volume = 0.4 dwindle_slowdown_volume = 5.0 retraction_enable = True plugin_config = hop_on_move = False model_matrix = 1.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,1.0 object_sink = 0.0
  13. mrjohnk

    Retraction and Blobs

    Printing at 95 mm/s and travel at 150 mm/s.
  14. mrjohnk

    stopped again ??

    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 portion together correctly as it would hold the plug down if it were flipped over. Unfortunately, I will need to disassemble the hot end upper assembly to flip that over as it has a nut on the inside. Hopefully this will help someone else who finds themselves in a similar situation. -John
  15. mrjohnk

    Printing ABS On Ultimaker

    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 seen by following the blue and yellow wires. The clips on the corners hold the glass to the heater. http://www.flickr.com/photos/8465641@N0 ... 166760275/ -John
  16. mrjohnk

    Printing ABS On Ultimaker

    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 of the contact portion of the relay. The positive side of your heated bed goes to the other contact portion of the relay. Hook the negative of your heated bed to the negative of the 24V power supply. Essentially, you are interrupting the circuit between the power supply and the heated bed by switching it through the relay on the positive side. Of course, you can switch the negative side if you want to as well. Polarity normally isn't too important on the heated bed except if you have an LED that is only working on one way since it is just a really long, flat wire on a circuit board. In my case, I have the MK1 heated bed circuit board with two LEDs mounted in two different direction, so either way you send the power in, it will light one or the other. -John
  17. mrjohnk

    Printing ABS On Ultimaker

    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
  18. Did you happen to install the 4.7K resistor to the board? It is required in order to read the thermistor correctly. -John
  19. mrjohnk

    Printing ABS On Ultimaker

    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 the MK1 heater. Use a lot of tape over a wide area since you really don't want this part coming off and causing a run-away heater. The tape is also used to provide strain relief for the heavy gauge wire connecting to the relatively small soldering spots on the MK1. I did have to fabricate some adapter plates that transition from the holes in the MK1 heater to the spring holes in the Z-stage on the UM. I just used some aluminum strip that I bought at the local hardware store. It is probably 20mm wide 3mm-4mm thick (estimated). It is mounted to the edges of the MK1 heater to extend the length as the heater is slightly smaller than the original acrylic footprint. Then, holes are drilled in the aluminum where the original spring loaded screws mount to the Z-stage. This allows me to continue to use spring mounted adjustments to set the level of the heated bed too. Also, I did mount my heater sideways so that the wires came off the side of the board instead of the front or back. Doing it this way, I did have to add some shims to the X & Y end stops to be sure that the hot end stays within the bounds of the heated bed since it is a little smaller than the original size of the bed that came with the UM. Using 24V seems like a great way to go. When I get ready to print, I simply set the heated bed temperature using the "control" part of the Ulticontroller to 115 degrees, then navigate to the card menu and select the file I want to print. the bed easily heats up faster than the hot end and is ready to go without slowing me down. From what I've read, some 12V implementations can take up to 10 minutes to fully heat and be ready to print on. I'm printing ABS at between 245-255 degrees on this setup. -John
  20. mrjohnk

    Missing Infill Layers

    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.
  21. mrjohnk

    Missing Infill Layers

    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 start the infill again. Once the parts are off the printer, several of them separate easily at this point while I'm trimming them up, however, they are supposed to be solid. Any thoughts on the best way to keep this from happening? A little background on the part: It is a button. Essentially a cube attached to a cylinder. I have added support structures around the edge to hold up the larger cube that is sitting on top of the smaller cylinder. These support are meant to be break-away, so they barely intersect into the model. It is at this intersection point of the supports where the infill layers are not present. Yes, some of the settings in the profile are kinda wonky like using a nozzle size of 0.5, overlap of 65% and filament size of 3.2mm. These are to control some of the fine detail and keep from over-extruding. Each tiny button is "engraved" with a number so getting fine detail is challenging and these seem to help. As far as I can tell, these setting are doing their intended job of getting the engraved portion done, but not causing lack of infill issue I'm having with the middle of the model. Thanks for any help. -John
  22. mrjohnk

    What tools do you use to cleanup prints?

    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
  23. mrjohnk

    Printing ABS On Ultimaker

    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
  24. mrjohnk

    Printing ABS On Ultimaker

    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 polyimide tape if you prefer) from my glass platform and started printing directly to glass. Initially, I didn't have a lot of success. I tried naked glass, sanded glass and sugar/water coating solution, but none worked well. I just completed a print using glass sprayed with aerosol hairspray. This worked well. I needed to turn the temperature of the heated bed down from the usual 115 degress to 75 to keep the hairspray from smoking and burning, but it worked well and held the part securely. I've yet to remove it from the glass, but at least it printed well. -John
  25. mrjohnk

    improving print quality, any help/experience?

    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
×

Important Information

Welcome to the Ultimaker Community of 3D printing experts. Visit the following links to read more about our Terms of Use or our Privacy Policy. Thank you!