1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Partially Solved Solution to PETg Print Jams

Discussion in 'Troubleshooting' started by nickster, Jun 28, 2015.

  1. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    This is a fix to some PETg jams that end up with a zig-zag of filement spaghetti below the hobbed bolt and above the input hole to the base of the Wade extruder. I am running with an E3D-V5 hotend on a highly modified KickStarter vintage Robo (the Robo 3D logo is about all that remains) running 1.75mm eSun PETg. This jam problem seems to caused by the filament not be fully supported as it enters the hole in the Wade base. That hole is oversize like it was meant to accommodate 3mm filament. PETg, at least the first spool I have, seems to require more force to drive it thru the extruder, at least compared to ToyBuilderLabs ABS at 250C. Increasing the PETg temp to 260 did not really reduce the viscosity. PETg seems to be more flexible and less rigid than PLA or ABS, combined with the greater amount of extrusion force, makes it more likely to kink up and jam. The telling experiment should be that after you have a jam, to snip off all the zig-zag and try to extrude the remaining material in the the nozzle by manually pushing with a new piece of filament. If there is a solid blockage and nothing comes out, then you have a different problem, and this fix won't do anything for you. But if it can be forcibly extruded, this fix should help. This problem can happen on E3D-V6 if you did not use the PTFE tube or cut it too short, or drilled out you Wade base to be too large so the tube is not well supported.

    Exposing PETg to moisture causes cross linking which increases strength, but also increases its melting temperature (and possibly viscosity?). I was not religious about bagging up my PETg spool between use, and we've hand an incredible spell of rain here in Colorado. However my results are still pretty translucent, which implies it is not heavily cross linked? But water absorption may be a source of variability for different people.

    The original Robo KS springs that tension the roller against the hobbed bolt were way too whimpy. I did an upgrade way back using springs from a local ACE HW. (I think the specifics are burried in a previous post some where). Not very scientific, but filament tension on my Robo was enough to lift the entire X carriage with steppers and hot end off the Z nuts to the inside top of the cover when I had a filament spool jam one time.

    Next, make sure your hobbed bolt is centered. Shim/cut do whatever so that the center of the hobbed teeth are exactly aligned with the center of the feed hole. The final readjustment I made using a calibrated eye-ball was a shift of 0.25mm. Having access to a lathe, I undercut that amount from under the head of the bolt. If the hobbed bolt is driving the filament off center, it will naturally want to kink.

    From the pic below, the spacer between the gear and the bearing was replaced with a nut. This scheme does not rely on the tension of the aircraft nut to hold the gear to the hobbed bolt. The nut is adjusted to have a very small amount of axial play on the hobbed side, and the aircraft nut inside the gear is heavily tighten (forming a double nut) against the new nut. There is still some jiggle (that needs to be fixed) of the hobbed bolt in the 8mm bearings since the bolt is under size. Also from the pic, make sure that the teeth of the hobbed bolt are uniformly cut. Sight the indent profile of the hobbed bolt as it is rotated; the diameter of the hobbed section shouldn't visually vary. The original KS bolt was a joke, and other's I got from Amazon/eBay were not much better, so I cut my own. Hopefully there are quality suppliers out there. A non-uniform hobbed bolt can create Z artifacts.

    Back to the fix. Disassemble the extruder, and thread a 4mm bolt into the filament hole in the Wade extruder base. A socket head cap screw is preferred since it makes it easier to drive. Run the bolt through a few times and the result should be a tight 4mm threaded hole. Not sure I would recommend using a tap because could result in an oversize hole.

    Next take a (long enough 45-50mm?) piece of 4mm o.d. x 2mm i.d. PTFE tubing, and insert a 2mm drill into the center to act as a core. If you have an set of indexed drill sizes, pick one that requires a bit for force to insert. Goal here is to have a 4mm OD threaded, 2mm ID piece of Teflon. I found a cutting die just digs into the PTFE, (roll form die might work?) and I could not get a start. Instead, take a 4mm nut and start threading it onto the end of the tube section until you can get a thread started. Might want to sand a bevel on the end of the tube to help it start. Make sure the nut goes on straight and does not cross thread. I grabbed the PTFE tube with core in the center with a lathe collet to help hold it. You could try a 3 jaw drill chuck. Chuck will squish the tubing, but it can be cut off later. Run the nut down the entire length of the tube a number of times, cold forming a thread. Expect the tube to lengthen somewhat a the displaced thread material has to go somewhere. The nut should not cut material from the PTFE.

    Next with the now threaded tube still on the drill core, thread the PTFE tube into the Wade extruder base hole from the bottom. (I tried threading from the top, but because of the angled hole entry, I couldn't get it to start.) It should be a very snug fit, but the drill core provides enough support to keep the PTFE tube from shearing. Once to correct height relative to the hobbed bolt, (see pic), remove the drill and cut the tube flush with base extrusion from the bottom side.

    Try some 1.75 filament in the hole and run the drill through it as necessary to open it up. In my case, it was just fine without any tweaking. Reassemble.

    Just ran and 8 hour print with no jams where before, it would jam in 1-20 minutes.

    Backup plan if I still get zig-zag jams with this set up, is to replace some of the PTFE with 4mm stainless screw drilled out for a 2mm feed hole near the hobbed bolt.

    I am also hoping this setup will improve Z artifact quality (all filament types) since the feedpath of the filament will be tightly constrained.

    Another tip is to use 0.014" steel wire available on line to clear out debris from your 0.4mm hot end tip. While it is hot, run a 1 meter piece of wire thru the tip fed from the bottom. Feed the wire into the tip using needle nose pliers; and be careful not to touch the hot end. Alternate feeding the wire, and pushing some new filament into the extruder and quickly withdrawing to pull out debris. Should never have to take a hot end apart to clean it. At one point, running ABS, there was a deposit that built up that causing back pressure issues. In free air, the filament was extruding at a 45 degree angle as it exited the tip. Wire cleared it out. Haven't had much luck with cleaning filaments.


    DSC02847a.jpg
     
    Frankn, Mike Kelly and mark tomlinson like this.
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Almost the same solution I used for NinjaFlex (except I extend the tube up and over the hobbed bolt).

    Good work.
     
  3. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    Mark, Sorry I missed the link to your solution. Can you post a link or pics? All I have seen with Google searches on PETg are mysteries and specific no resolutions.
     
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
  5. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    Update:
    Had a print jam caused by the eSun PETg getting flattened as it went past the hobbed bolt and pinch bearing. The filament flattened out to 1.85mm. Fix was to drill out the PTFE insert to 2mm. Also read on line, to use just enough pinch force to keep the filament from slipping which kind of goes against the need for additional filament force compared to ABS/PLA. Reason for the additional force remains unclear. Specific heat and thermal conductivity of PETg and ABS are similar.
    http://www.makeitfrom.com/material-properties/Glycol-Modified-Polyethylene-Terephthalate-PETG-PET-G/
    The squishing of the filament was probably aided by running in a heated enclosure which was probably getting close to 50C similar to what I run with ABS.

    Using small diameter wire to clear material from the nozzle works sometimes. But I stand corrected that you never have to disassemble a nozzle to clear it. Wasn't thinking very much, was lazy, and drilled out the PTFE with a pin vise without disassembling the extruder. The drill also nicked the top of the E3D V5 aluminum hot end tube, creating PTFE and aluminum metal shards. I gave up trying to clear out all the debris, and finally had to disassemble the hot end to clean it out.

    Good news though is I ran a 7 hour PETg print on some parts with tall vertical walls that were problematic to print in ABS because of bed lifting. Results were fantastic with no lifted edges using Elmers Extreme glue stick on glass. This was done 90C bed temp 255C hotend, full fan, minimal pinch spring tension, room temp (no heated enclosure). Don't know if this matters, but somebody said to run the bed to 110C to cure the glue stick and then run the print at 90C.

    I don't see PETg replacing ABS for higher temp applications, but it is certainly better than PLA.
     
  6. Frankn

    Frankn Member

    Joined:
    Mar 1, 2015
    Messages:
    261
    Likes Received:
    22
    Note: PTFE expands over time.
    I am a trained observer and have watched the movements as the Robo runs.
    Here is something that I have never seen mentioned. The drive bolt has a tendency to twist or spin the filament as it is driven.
    You will only notice this on long runs (HOURS). You can observe it between the filament supply spool and the bolt.
    This twisting motion just might built to the point of 'folding' softer filaments.
    Also, with that 'Funnel' PTFE lead in, it makes for easier starting, but it may also funnel unwanted material in.
    Just some thoughts, Frank
     
  7. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Polycarb, or frankly, other better choices exist if you need something better than ABS.
    Each has its own quirks :)
     
  8. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    @ Frankn - Interesting observation. I am not experiencing any filament twisting, but I can believe it could happen if the dish on the hobbed bolt is not centered with the filament feed or the pinch roller is not parallel to the bolt. Had to re-spool a filament roll one time using a hand drill. Seemed like the filament tended to twist one way for 30 turns, and then the other way. Not a clue how that could happen.

    @mark - Yup. Choose your poison. Alternatives I know of shrink like crazy or are extremely hygroscopic. I remember about a year ago, some experiments with in-line heater/dryers for polycarb. Glass transition temp on ABS is ~102C; PETg ~81C. Off topic, but what's your favorite high(er) temp goto filament for printing large pieces?
     
  9. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    At the moment one of the Nylons but usually Allow 910 (depends on what properties I want). Both of those are easier to print and tolerate heat well. In my cases alloy 910 pretty much replaced any need for ABS. I haven't even reached for a spool of nylon since I got the 910 working either.
     
  10. nickster

    nickster Member

    Joined:
    Oct 7, 2013
    Messages:
    82
    Likes Received:
    52
    What kind of drying are you needing for Alloy 910?
     
  11. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    None at all.
     
  12. Bulent

    Bulent New Member

    Joined:
    Jun 17, 2016
    Messages:
    1
    Likes Received:
    0
    Same problem here with my custom made delta. Here is my observations: esun PETg is slightly softer than others and not smooth. The 1.75 filament easily shapeshift right after the drive bolt. And this make it easy to twirl before the exit hole. What I found is, PETg from esun is a little more viscous when melted. I use 235C and no heated bed. So, I try to lower the nozzle as closely as I can, with the paper method.

    With all this combined, the nozzle is low so a tiny space is between bed and nozzle, extruded PETg can not go out easily because of high viscosity and easy to twirl because the filament is soft.

    So I doubled the nozzle space and boom! No more spaghetti!
     

Share This Page