SpiderBot Delta 3D Printer

My Spiderbot Build Log

posted in General
Friday, October 11 2013, 01:11 PM
Hi All, I've begun blogging my experiences building, calibrating, and printing using the Spiderbot. Includes my current Kisslicer settings for those that are interested. To read more, go to http://spiderbot3d.blogspot.ca/.

I hope this helps others out. Please comment on my blog, help me improve my own prints, and share your experiences.

djs
    Responses (20)
    • Accepted Answer

      Friday, October 11 2013, 01:13 PM - #permalink
      Nice! Thank you for sharing, reading it now. I also posted about my Spiderbot on my blog, although not as elaborate as you: http://www.cmhofstede.nl/hobby-vrije-tijd/overig/de-spiderbot-3d-printer-het-uitbreiden-van-zelf-fabricage-mogelijkheden/

      Having read your latest entry, I can confirm I had some similar issues and thoughts.

      - The end switches due effect repeatability in the order of +/- 0.2 mm on my machine. As far as I can tell anyway.
      - I have only printed ABS, which went fine and sticked to the tape that came with the Spiderbot just fine.
      - One of my rod ends also came loose yesterday, I thought I had broken it, but luckily its just a matter of fixing it back with new glue. I will make sure to check all of them.
      - The 580-600 milli voltages for the stepper controllers make them overheat quite fast on my machine also, I installed a fan. Works good so far.
      - Also, like you, I was expecting some issues with the 1.0 production run. So far things have been minor and easily fixed. Hope it helps the next generation of owners. :)

      I'll have a look at your auto starting to print settings when I get back from work.
      The reply is currently minimized Show
    • Accepted Answer

      Friday, October 11 2013, 01:13 PM - #permalink
      Thanks for the kind words Casper. Keep in mind my Kisslicer settings are for PLA, not ABS at present, using a 0.35mm print nozzle. There are also adjustments to z-axis lift in Kisslicer to accommodate the way I calibrate by bed.

      I just ordered some white ABS to try as well. I checked your blog, and it looks like you may be extruding too much ABS, or at a too high temperature, as your prints look 'goopy' especially first 1-3 layers. I'll try printing in ABS to see what happens, and see if we can clean up print quality overall.

      djs
      The reply is currently minimized Show
    • Accepted Answer

      Saturday, December 14 2013, 09:55 AM - #permalink
      I checked and read both your updates, the one with the fan duct and the nozzle calibration settings. Nice work! :D

      Can you elaborate a bit on the temperature measurement device you used (link?). I might consider doing something like this myself.

      The acceleration settings, you mention they are better at 1500, I had them set to 3000 at first and printed a lot of figurines at full speed in ABS, successfully that is. Then again, these were small figurines and not located at certain edges of the build platform.

      For testing your fan duct, you may have noticed there is a test script (Control_Fan.gcode) included with the installation to move along the border of the build plate. Thus checking if you can move freely everywhere. I did a fresh install recently but can't seem to find it... But I am sure it was there.
      The reply is currently minimized Show
    • Accepted Answer

      Monday, December 16 2013, 02:33 AM - #permalink
      I have a fairly expensive digital multimeter - an Instek GDM 393A. It came with a thermocouple attachment that can measure degrees fahrenheit or celcius. It's about $250.00, so likely not something you buy lightly.

      Here's the link: Instek GDM 393. Looks like it has been replaced by the GDM 394. Looks like your best bet now would be the GDM-350A, which is $15.00 + free shipping here: GDM-350A

      Re: 1500 vs 3000 acceleration: I think is a matter of preference- I was finding that the moves were so fast that any small blob that the head hit caused the magnets to release from the arms from the force of the blow. I was nervous. If you are extruding cleanly with no blobs, and you can raise the acceleration, go for it. I was having trouble with blobs from extruder pressure inconsistency printing PLA mainly, due to the way the fan was directing air without the new duct. I may try to increase speed again now that things appear to print cleanly with the new duct.
      The reply is currently minimized Show
    • Accepted Answer

      Monday, December 16 2013, 10:22 AM - #permalink
      Thank you for the thorough reply, I own a Fluke 115 myself. I will have a look around if there is a temperature probe available for that.

      Blobs make your print head assembly come apart? Wow! I have had it once with one of the rods, but that was due to the filament guide cord pushing it away from the top side.

      At the moment I also have these PLA blobs that come out. But the situation it happens in is always the same, with large area prints with plenty of fill in. It seems, just before or during a large head movement (3-4 cm) the PLA gets retracted and then the first few lanes it tries to fill in are barely extruding and then a blob comes out. The next layer at the same location the same thing happens. If I print a small test cube, no problem as the movement distances are not that big. I am hoping this is a cooling problem that can be resolved with the air duct.
      The reply is currently minimized Show
    • Accepted Answer

      Monday, December 16 2013, 06:24 PM - #permalink
      I was getting the same thing with PLA- I think it's a two part issue- because the airflow is not great, the PLA sits in the barrel, cooks itself, and stagnates. This creates a partly solid plug of modifed sub-optimal PLA in the barrel. When there is sufficient pressure to force thru the plug, it results in a blob- you can see this as you say at the corner of a large infill area- barely no extrusion to start for the first 4-5 infill lines, then a 'spurt' as it pushes thru to a nice looking extrude again. This is the blobs I get that cause the print rods to pop off- but only if the rods are at extreme angle on a large print to the outside of the print bed when travelling at speed for infill. The second part: to counteract this, you increase the PLA temperature, which makes it more liquid, but this of course causes more heat to travel up the barrel, creating an even larger plug, and when it burst thru, its goopy liquid...

      Try the duct, and reduce your PLA temperature to the bare minimum you need to get to stick and lay down nice sharp corners.

      PS I bought some 3" wide foil backed fiberglass insulation and some aluminum tape used for metal house ductwork, and insulated the bottom of my heated bed (foil side away from heat). I am now doing a test print with ABS at 225 degrees celsius with a 110 degree heat bed, and have been able to maintain 107 degrees celsius. It could go hotter with the two resistors in the heated bed, but it looks like a larger power supply would be needed. It takes about 20 minutes to get to 100 degrees c. Crossing fingers the warping issue will disappear at the bed edges.
      The reply is currently minimized Show
    • Accepted Answer

      Monday, December 16 2013, 06:30 PM - #permalink
      The Tek Guy wrote: I was getting the same thing with PLA- I think it's a two part issue- because the airflow is not great, the PLA sits in the barrel, cooks itself, and stagnates. This creates a partly solid plug of modifed sub-optimal PLA in the barrel. When there is sufficient pressure to force thru the plug, it results in a blob- you can see this as you say at the corner of a large infill area- barely no extrusion to start for the first 4-5 infill lines, then a 'spurt' as it pushes thru to a nice looking extrude again. This is the blobs I get that cause the print rods to pop off- but only if the rods are at extreme angle on a large print to the outside of the print bed when travelling at speed for infill. The second part: to counteract this, you increase the PLA temperature, which makes it more liquid, but this of course causes more heat to travel up the barrel, creating an even larger plug, and when it burst thru, its goopy liquid...Try the duct, and reduce your PLA temperature to the bare minimum you need to get to stick and lay down nice sharp corners.
      Yes, I visited the hardware store today, but they didn't have any hex M3 bolts... I might order a set just for this thing, or redesign the duct.

      The Tek Guy wrote: PS I bought some 3" wide foil backed fiberglass insulation and some aluminum tape used for metal house ductwork, and insulated the bottom of my heated bed (foil side away from heat). I am now doing a test print with ABS at 225 degrees celsius with a 110 degree heat bed, and have been able to maintain 107 degrees celsius. It could go hotter with the two resistors in the heated bed, but it looks like a larger power supply would be needed. It takes about 20 minutes to get to 100 degrees c. Crossing fingers the warping issue will disappear at the bed edges.
      Interesting, can you provide a picture? I understood from Philippe that a shield in between the heaters and the bottom compartment of the printer would indeed also help.
      The reply is currently minimized Show
    • Accepted Answer

      Tuesday, December 17 2013, 02:58 AM - #permalink
      You should not need any new bolts- I countersunk the bolt hole used to remount the duct to the black standoff. If your hole is too small, ream it out a little bit- depending on how clean it printed, the countersink area may need some cleaning with a knife blade. I was able to use the existing screw that came off the black standoff to remount the duct on top of the standoff. Just push it in firmly and screw while pushing down, and it should grab and pull in.

      The other two bolts that mount the fan to the duct are the originals that mounted to the aluminum fan holder before. There is plenty of room- at least on my print- to reattach the nuts to the back of the screws.

      I'll post some pictures tomorrow of the insulated bed- it can eventually get up to 110 degrees celcius now. There is only a very small amount of warp now on large prints- but I am printing on glass only. Am waiting for kapton tape to see if that cures warp for good.
      The reply is currently minimized Show
    • Accepted Answer

      Wednesday, December 18 2013, 12:50 AM - #permalink
      Pictures and instructions on how to insulate the print bed can now be found on my blog here:

      In search of a hotter bed for printing ABS
      The reply is currently minimized Show
    • Accepted Answer

      Johan
      Johan
      Offline
      Thursday, December 19 2013, 08:03 PM - #permalink
      Nice job! I will definitely give this a try.
      The reply is currently minimized Show
    • Accepted Answer

      Thursday, December 19 2013, 08:18 PM - #permalink
      Printing on glass, at 110 degrees c, I was still getting warping (although much less than with a coolder bed) around the edges of very large flat prints (those that exert a lot of force as the plastic cools and contracts). I've now been able to minimize this by following the 'ABS juice' method- dissolving some ABS plastic in a glass jar of acetone. Before each print, with the bed cool, clean the bed with acetone and then smear this juice on bed with a kleenex or cloth. This acts as a very thin raft of ABS to help the printed ABS deposited on top to bond to the bed. You still need the bed as hot as it can go to minimize the stress placed on this raft, or you'll notice the raft itself (looking like skin) lifting off the bed along with your print. That's about as good as I can get with an ABS print to-date. I really hate printing a full ABS raft- a pain to remove, so am trying as much as possible to find a method that does not require one.

      I will be trying Kapton tape out as well, as I hear that bonds at temperature to ABS possibly even better than ABS juice.

      djs.
      The reply is currently minimized Show
    • Accepted Answer

      Friday, December 20 2013, 09:40 AM - #permalink
      Wow you are one fire David! Nice! :D

      I am curious though, do you really need to print large ABS parts ( I do for one ) or is this just pushing for innovation and enjoying the experiments? :p
      The reply is currently minimized Show
    • Accepted Answer

      Monday, December 23 2013, 05:55 PM - #permalink
      I find that ABS is much easier to sand/machine than PLA, is more flexible (important when used in designs that will vibrate a lot), and has a much higher point of plasticity (PLA goes soft at 50C, which is easily attainable on a hot summer day). Therefore I prefer to print 'serious' parts in ABS, and everything else not so critical in PLA. I do print almost to the maximum of my bed, and I can see from your blog you are printing so large you have to 'slice' your parts to fit the bed.

      And you are right... I also enjoy a challenge.... :D
      The reply is currently minimized Show
    • Accepted Answer

      Wednesday, April 30 2014, 06:01 PM - #permalink
      Hotend Clogging:
      We have had our machine for a while now trying to get it to run well with PLA. We are noticing the same thing with stalled extrusion followed by a burst of excess material as the clog clears and it spews forth the built up material. Even more concerning however is that the hotend will completely clog whenever it is doing a print that calls for a lot of short moves and retracts. If we turn retract off it will not clog, but we get lots of stringing. With the retract on, if it does too many in too short of a distance then the PLA stops flowing completely. The stepper can't move the filament and just makes a barking sound as it forces filament into the bowden anyway, and then spins backwards a bit as it skips steps and the springiness of the bowden pushes the filament a bit back through the extruder. If we pause the print and try running the extruder it remains clogged unless we retract about 30mm then extrude again. This always clears it. When we retract in this scenario, we can see the bowden wind up before the filament pops free of the hotend as we retract. We tried a lot of things to fix this, including:
      o a very nice fan shroud to keep the cold part of the hot end cool
      o even added a piece of insulation material between the heater block and cooling fins
      o bought an entirely new hotend / bowden assembly, thinking we had a bad unit
      We are using the stock .5mm nozzle. We were doing 0.2mm layers and upped that to 0.3 which seemed to help a bit.

      Has anybody else experienced this issue? Like I said, it only comes up when there are a lot of retracts. If the extruder is allowed to keep extruding without pause it is fine.

      -Paul

      CasperH wrote:

      ...At the moment I also have these PLA blobs that come out. But the situation it happens in is always the same, with large area prints with plenty of fill in. It seems, just before or during a large head movement (3-4 cm) the PLA gets retracted and then the first few lanes it tries to fill in are barely extruding and then a blob comes out. The next layer at the same location the same thing happens. If I print a small test cube, no problem as the movement distances are not that big. I am hoping this is a cooling problem that can be resolved with the air duct.


      The Tek Guy wrote:

      I was getting the same thing with PLA- I think it's a two part issue- because the airflow is not great, the PLA sits in the barrel, cooks itself, and stagnates. This creates a partly solid plug of modifed sub-optimal PLA in the barrel. When there is sufficient pressure to force thru the plug, it results in a blob- you can see this as you say at the corner of a large infill area- barely no extrusion to start for the first 4-5 infill lines, then a 'spurt' as it pushes thru to a nice looking extrude again...
      The reply is currently minimized Show
    • Accepted Answer

      Friday, May 02 2014, 01:45 PM - #permalink
      It seems there is a lot of difference between PLA.
      Last week we tried one other manufacturer and it has been impossible to print a complex part with.
      The same GCode file with an other PLA works fine and print perfectly.

      Currently we try to understand what can happen, here we are :

      1 - We don't think it is a cooling problem, the cool area of the head is always well below the glass transition temperature of PLA (often ~30°C)

      2 - We suspect it is that : when the Slicer retract 4 or 5mm filament, the filament has a small blob at end (due to the diameter of tube which is bored at 2.0mm and filament at 1.75mm).
      This blob is very soft and when the slicer push again the 4 or 5 mm filament (to start a new print zone), the blob rubs into the tube and adheres to the wall of tube. The more the extruder pushes, the more the filament adheres to wall...

      We have a spool of filament doing that, we work on it.
      I have the Differential Scanning Calorimetry of the filament which works. I will ask a new DSC of Filament which not work (the manufacturer seems not able to provide the DSC of his material) .
      The problem is it can take time to get the DSC... Today, unfortunately we do not have solution.
      The reply is currently minimized Show
    • Accepted Answer

      Joules
      Joules
      Offline
      Friday, May 02 2014, 08:39 PM - #permalink
      Reduce your retraction to 1mm. I have just been battling this problem for the last day having changed filament.

      Joules
      The reply is currently minimized Show
    • Accepted Answer

      Friday, May 02 2014, 09:41 PM - #permalink
      Thanks for the insight Philippe & Joules.

      We were really wondering if we were the only people having this problem. Does this problem happen with ABS? Is there a supplier of PLA in the USA I can buy from that has "good" PLA?

      -Paul
      The reply is currently minimized Show
    • Accepted Answer

      Saturday, May 03 2014, 03:07 AM - #permalink
      I have had zero problems using the standard Spiderbot extruder (mine is 0.35mm) and ABS filament. Most of my issues arise with PLA. There are few observations I can conjecture on use with PLA:

      Not all PLA filament is exactly 1.75mm, even on the same spool. The closer this gets to the 2mm bore of the all metal hot end, the more pressure and friction is created. PLA has a higher coefficient of friction than ABS. Most of my jams are created with crap PLA filament that measures 1.8 mm+ instead of 1.75mm. Even if I adjust this in the slicer, it still means a much tighter fit in the barrel of the hotend. With this tighter fit, I expect temperature is too high, and cooling is too much, trying to compensate: the filament is melted almost to liquid at 240 degrees to lower its viscosity, but is simultaneously being cooled in the too-tight barrel to the point where it is partially adhering to the wall of the extruder further up. The two are fighting each other. This is pure conjecture, but as an experiment, I designed and printed a Spiderbot mod. for a standard j-head hot end, which is partial all metal instead of full all metal, which uses a teflon/PTFE tube to the tip as a barrel liner, with all aluminum exterior. The make is Aluhotend v5, which can be found here: aluhotend v5. A very modest fan is used to cool, and the PTFE liner drastically reduces friction. Temperature I am now printing ABS at is 230 degrees, but for PLA I am now printing without issue at 210-215 degrees. Most material on the 'net suggests these temperatures for PLA, yet with standard Spiderbot we are printing PLA at 240-245 (likely to lower viscosity). I have only had one issue with this j-head extruder since: the thermocouple wire detached from the hotend, and the hotend heated to 350 degress c or so, cooking the hotend and the PTFE liner- it was black and charred. I had to disassemble the entire thing and replace the PTFE liner. This was pure accident- I was fooling around too much with the hotend, and pulled the wires loose- but did not notice until too late on my next print.
      The reply is currently minimized Show
    • Accepted Answer

      Tuesday, May 20 2014, 06:59 PM - #permalink
      I just got the DSCs of 2 different PLA : One working fine and the other one not working with our extrusion head.
      There are big differences in the comportment of material with temperature.
      We will do some modifications this week on one hot-end.
      If it works and if you want to test Paul, I can send one hot-end in one or two weeks...
      The reply is currently minimized Show
    • Accepted Answer

      Friday, May 23 2014, 06:56 PM - #permalink
      Thanks for looking into this.

      Two things I wish we knew from day one- don't try PLA and don't bother with open source slicers.

      We switched to ABS and bought Simplify3D and all our problems went away. Combating curl is the only thing we have to worry about.
      The reply is currently minimized Show
    Your Reply