Duet3D Logo Duet3D
    • Tags
    • Documentation
    • Order
    • Register
    • Login

    A Repeating Artifact

    Scheduled Pinned Locked Moved
    General Discussion
    6
    23
    1.5k
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • Sethipusundefined
      Sethipus @B0M0A0K
      last edited by Sethipus

      @b0m0a0k said in A Repeating Artifact:

      @mrsdelish Thanks for this. I had been researching what I could do by way of an upgrade to the extruder or the hotend but was a little undecided because everyone has a different take.

      My options so far are;

      Micro Swiss Mk10 All Metal Hotend Kit with cooling block. (Straight replacement)

      E3D Titan Aero HotEnd and Extruder

      Various other suggested mods.

      Don't get me wrong, I enthusiastically modded and upgraded my Wanhao (Monoprice Maker Select actually; it's the same printer), but I would like to opine that whatever was causing the behavior you saw in your initial posting would not be improved by upgrading the hotend, or the extruder for that matter. There's nothing wrong with the hotend and extruder (it's direct drive) on that printer, unless you want to print at nozzle temps higher than around 240-245 C or so for very long. I printed tons of PETG on that printer (including all the parts for my initial build of my D-Bot), and eventually replaced the teflon tube in the hotend and found it fairly blackened and charred. The problems you are seeing are 100% settings issues.

      To that end, I would recommend against tweaking extrusion multiplier and whatnot as a fix to your problems. That's a kludge that should be used last, after everything else is dialed in to what they should be. I'd set your e-steps back to what they ought to be. Using e-steps to try to "fix" problems is the wrong approach. Do the measurements and actually measure how many steps it takes, and use that value. Measure your filament and input that value. If it's within a couple thousandths of 1.75mm then it doesn't really matter here or there, but I recently was printing with some filament that was so much smaller I could actually see it with my naked eye, and measured it out at 1.62mm, which is definitely enough to make a difference.

      Once you're printing with relative extrusion, then dial in your retraction. Once that's done, then modify your acceleration/max speed/jerk settings, and you should be printing really nicely. You can do all this with Cura too. There's nothing magical about Slic3r compared to Cura that caused the massive improvement in results you saw. It's just settings, and you can change whatever you need to in both and get pretty much the same results. Massive tweaking of the extrusion multiplier will just cause havoc from filament to filament and print to print.

      Btw, I think I paid like $250-300 for that Maker Select printer, and by the time I was done with little upgrades and mods and whatnot I don't doubt I'd paid upwards of $600-700 total, including the Duet Wifi, which means I could have just bought a $600-700 print to start and probably been better served (other than the Duet Wifi, which I consider mandatory on any printer I will use). Oh well, I learned a massive amount with that printer, and so in the end it served me well.

      B0M0A0Kundefined 1 Reply Last reply Reply Quote 2
      • B0M0A0Kundefined
        B0M0A0K @Sethipus
        last edited by B0M0A0K

        @sethipus Thanks for this, that's a very well reasoned and useful set of instructions. I agreed with what you have said (I don't disagree with any of it) regarding the slicer used. I tried to print the Octopus after the successful cube test and it was so bad I had to stop the print.

        I will go back and put everything back how it was, but will use your sequential diagnostic to work through everything else. The printer is in bits at the moment because I took the extruder and the hot end apart whilst checking the condition of the extruder gear and looking for blockages in the hotend. Didn't find anything of note but I am sure I was hearing a click from the extruder gear as it was passing filament and wanted to investigate that.

        I'm going to go ahead with the Micro Swiss hot end because most folks seem to support this as a good upgrade.

        Thanks again, I very much appreciate you taking the time to put this together.

        ----------------------------------------------
        No fate but what we make ....
        Sethipusundefined 1 Reply Last reply Reply Quote 0
        • Sethipusundefined
          Sethipus @B0M0A0K
          last edited by Sethipus

          @b0m0a0k I almost got the Micro Swiss hotend at one point, but since I was successfully printing PETG through the stock hotend, I never actually bought it, and once I settled on building a new printer from plans the idea became redundant. There's nothing wrong with the stock hotend though. For people printing just PLA I'm not sure I see the point. I haven't looked at pricing on those things lately, but if the MicroSwiss hotend is in roughly the same ballpark as an E3D like the Titan Aero, I don't think you can beat the E3D. They're the gold standard. For one thing, the nozzles are nearly a universal standard, and the extruder is geared, which means (along with a great controller like the Duet Wifi) more steps/mm for finer and quieter control of extrusion, you can use a lighter extruder stepper, etc.

          I read about people getting clogs in their printer, and I simply never get them. I think I've gotten a legit clog maybe once ever in thousands of hours or printing, and if I recall it was an experiment with carbon fiber PETG filament, and the single-drive extruder I was using at the time had stripped out the filament so filament sat in the nozzle and cooked.

          I don't know if you've heard of this before, but most times when I'm reloading filament I'll do what's known as the "Atomic Pull." And when I switch between PLA and PETG or some other filament (ASA, nylon, etc.) I do this 100% of the time. Simply, an atomic pull is where I pull the existing filament out (leaving some melted filament down in the melt chamber) and push in a length of cleaning filament, until what's coming out is mostly the clear cleaning filament. It's not necessary for it to be 100% clear. In the Duet Wifi web control set the nozzle temp to 0. As the temperature falls below around 200 or so give one last little short push on the cleaning filament to push through another couple milimeters of it, then let it sit. As the temperature falls to around 90 C or so simply pull the filament out with a pair of pliers or whatever. If you do it with the temperature too high the cleaning filament will break off leaving the filament in the melt chamber, same if you let it cool too far below 90 C. If you pull it out in this sweet spot temperature zone you'll pull the cleaning filament out of the melt chamber, and you'll have a perfect "casting" of the melt chamber that will remove anything and everything in there, so the melt chamber is perfectly clear for putting in the new filament.

          Sorry to be so chatty, just got kind of enthused on the topic of that particular type of printer, and all the many upgrades and mods people do on them. With the Duet Wifi you've already done the best possible mod on that printer bar none. No more shuttling an SD card back and forth between the computer and printer, massively quieter printing, nearly unlimited configuration simplicity, etc. Good times.

          ETA: one other topic is on the extruder itself. With the Micro Swiss hotend you'd be stuck with the stock extruder, which is functional, or if you went Titan Aero you'd get a nice geared extruder along with the E3D hotend in a nice, compact package, but it's still single drive (one toothed gear pulling on the filament). I'm now using dual drive (two toothed gears pulling on the filament rather than one toothed gear with a flat bearing pushing the filament against it) and dual-drive is quite simply fantastic. I'm using a Bondtech BMG clone that works amaziningly. If you can find or design a print carriage for your printer that combines a Bondtech-style extruder with an E3D hotend in a compact-enough package you'd have the best of all solutions. With the Micro Swiss hotend you're locked in to their nozzles, and it's just a hotend, leaving you with a simplistic stock extruder. If you're gonna go, go big.

          B0M0A0Kundefined 1 Reply Last reply Reply Quote 0
          • B0M0A0Kundefined
            B0M0A0K @Sethipus
            last edited by

            @sethipus Hey, thanks again for the detailed response. I am aware of the Atomic Pull and have used it before. Your info on the E3D like the Titan Aero is most interesting and I may take another look at that. I have just completed putting everything back together and have re-aligned everything. I am going to go and look again at my config.g and undo the Steps per mm change I made then fire up Cura and start over with attempting to tune my prints.

            ----------------------------------------------
            No fate but what we make ....
            1 Reply Last reply Reply Quote 0
            • B0M0A0Kundefined
              B0M0A0K
              last edited by

              I was wondering if anyone had suggestions for good test print pieces that do not take ages to print and will allow for speedy test prints? I am currently using the Octopus print but it takes over an hour for each test.

              ----------------------------------------------
              No fate but what we make ....
              1 Reply Last reply Reply Quote 0
              • mrehorstdmdundefined
                mrehorstdmd
                last edited by

                Some people prefer to test everything (?) at once by printing a tugboat or other objet du jour, and then try to sort out the problems. I usually design test prints to test specific things that I'm trying to test or tune up. The prints get right to the point and often finish quickly. For example, if you want to check squareness of axes, steps/mm calibration, print a large 1/2 cube. For testing seam quality, print multiple small cones or cylinders with the slicer set to align the seam. For bridging performance, print an object that has multiple bridges of different lengths oriented in different directions, printed at different speeds. For testing bed level/flatness, multiple square or circular objects scattered over the bed surface. It helps to get a little familiar with gcode so you can go into and edit the files for specific properties like speed, temperature, etc.

                https://drmrehorst.blogspot.com/

                T3P3Tonyundefined B0M0A0Kundefined 2 Replies Last reply Reply Quote 1
                • T3P3Tonyundefined
                  T3P3Tony administrators @mrehorstdmd
                  last edited by

                  @B0M0A0K I agree with @mrehorstdmd and once you have the basics dialied in with cubes etc then there are good objects on thingiverse that have, for example, a bunch of different bridge lengths, or a bunch of different slopes for overhang testing.

                  www.duet3d.com

                  1 Reply Last reply Reply Quote 0
                  • B0M0A0Kundefined
                    B0M0A0K @mrehorstdmd
                    last edited by

                    @mrehorstdmd OK, I took your advice and focused on using calibration cubes to set retraction first. Here are the results and it's clear that the farther away from 1mm the worse it gets.
                    0_1567265411154_Calibration Cubes Retraction Tests.png
                    One strange artifact that appears on all these cubes is the "tail". It typically appears on the first cube and varying lengths in between but always presents again on the last cube.

                    0_1567265422778_Calibration Cubes Retraction Test - Tails.png

                    Any thoughts on what might be causing this?

                    ----------------------------------------------
                    No fate but what we make ....
                    1 Reply Last reply Reply Quote 0
                    • B0M0A0Kundefined
                      B0M0A0K
                      last edited by

                      .. And it doesn't actually help if I don't tell you the settings I am using in Cura. Sorry, here are my Cura settings.

                      Resolution: 0.2mm
                      Print Temp: 200 and later 195
                      Print Temp Initial Layer: 200 and later 195
                      Initial print temp: 190 and later 185
                      Build Plate Temp: 60
                      Flow: 100%
                      Retraction enabled
                      Retraction Distance: 1mm
                      Retraction Speed: 25
                      Print Speed: 60
                      Travel Speed: 120
                      Initial Layer Speed: 30.0
                      Cooling Enabled
                      FAN Speed: 100
                      FAN Regular Speed: 100
                      FAN Max Speed: 100
                      Support: No
                      Build Plate Adhesion: Skirt
                      Skirt Width: 3
                      Print Sequence: All At Once

                      ----------------------------------------------
                      No fate but what we make ....
                      1 Reply Last reply Reply Quote 0
                      • B0M0A0Kundefined
                        B0M0A0K
                        last edited by

                        Actually, I kept increasing the retraction and eventually the "tail" disappeared. I needed to increase retraction to 1.3mm before this happened.

                        ----------------------------------------------
                        No fate but what we make ....
                        1 Reply Last reply Reply Quote 0
                        • First post
                          Last post
                        Unless otherwise noted, all forum content is licensed under CC-BY-SA