BMG + Mosquito + Tool Changer

Discussion in 'Tool heads & ToolChanger' started by Ntesla66, Oct 24, 2019.

  1. irumbo mateene

    Joined:
    Oct 6, 2020
    Messages:
    7
    Likes Received:
    5
    Hi,

    I am using your designs to build a BMG + mosquito. The for putting time and sharing it, I am taking this build slowly and hope to have something working by end of year.

    I opted to use the shaft that came with Bondtech BMG it is much shorter than the 24 mm cut I would have made from the 32mm shaft.

    Is there something I should be aware of if I go this route. cutting the 32mm shaft I had was a bit difficult so I opted for this temp.solution.


    Thx
     
    garethky likes this.
  2. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Yes: The shorter shaft will deform the PETG used to print the parts. It's too much force over too small an area. This was one of the failures I had with the prototypes. I up-sized it to match the shaft Prusa uses which is a tested design. The Bondtech extruder is some SLA plastic that is tougher than PETG (Nylon?) so they can get away with the shorter shaft on the idler.
     
  3. irumbo mateene

    Joined:
    Oct 6, 2020
    Messages:
    7
    Likes Received:
    5
     
  4. irumbo mateene

    Joined:
    Oct 6, 2020
    Messages:
    7
    Likes Received:
    5
    Thx for the update, I opted to use Prusa PC Blend filament to print the parts but one of my tools is using PETG for the main body.

    I wil reprint the parts since I had used 15% infill for the PC blend. My toolchanger is half way built, also studying up on how to program this printer something I will need to master once the hardware is fully assembled..
     
    garethky likes this.
  5. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Let me know how it goes with the PC Blend. I have some as well but I wanted the design to be easy to print. I may convert mine over to PC blend when I do an enclosure.
     
  6. irumbo mateene

    Joined:
    Oct 6, 2020
    Messages:
    7
    Likes Received:
    5
    I will do so -- I am just about finished building up my tool changer --- I will let you know before end of year .. sorry I have been rather slow in getting this off the ground..
     
    garethky likes this.
  7. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Happy New Year! I want to make 2021 the year I print more on my Tool Changer than my Prusa. So as promised, some 6 months ago :oops:, I've updated the tool design to support the 40mm Noctua fan. I think this is the first tool to have both a 5015 part cooling fan and the 40mm hot end fan on it so its now probably the quietest tool you can build:

    mosquito extruder Revision 3.jpg

    All of the parts had to be adjusted to support this. You can read the full details and get the .stl files in this commit on the development branch of the project. Its still not an official release as I haven't updated the build instructions and photos but it should be very close to final.

    I would really be grateful if at least 1 other person built one so I could have my blind spots revealed to me. For my part, I'm going to have this one print all 3 of its siblings and then a bunch of parts for my CNC machine.

    Happy Printing in 2021!
     
    irumbo mateene likes this.
  8. copperbricks

    copperbricks Member

    Joined:
    Aug 1, 2020
    Messages:
    3
    Likes Received:
    1
    I actually have two of your tools built! They've been functioning quite nicely on my homebuilt toolchanger. It uses a different docking mechanism but I designed the tool plates to be compatible with E3D ones so your design only required a little bit of modification. I actually just finished adding a 40mm fan shroud as well, the mosquito fan was driving me crazy. Thanks for the great design!

    Images:
     
    garethky likes this.
  9. blarbles

    blarbles Well-Known Member

    Joined:
    Aug 10, 2019
    Messages:
    108
    Likes Received:
    68
    How heavy is the whole thing?
     
  10. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    ~400g.
     
  11. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    @copperbricks I'm happy to hear that you built one!! (I cant see those images?)
     
  12. mkudzia

    mkudzia Active Member

    Joined:
    Dec 26, 2019
    Messages:
    43
    Likes Received:
    24
    Thinking of building one to replace one hemera tool (which give me headaches)

    Do you have pictures of any prints and settings to share?

    Regards
    Marcin
     
    garethky likes this.
  13. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    The last thing it printed:
    [​IMG]

    Issues I've identified and am working on:
    1. The depthing between the drive gear and reduction gear is a little sloppy. I'm printing experimental parts now that bring them 0.25mm closer together. The other cause of loose fit is the generous clearance around the boss on the stepper. I'm experimenting with tightening that up to 0.05mm radial clearance (0.1mm over spec diameter). This loose fit causes an annoying clicking noise on retractions.
    2. The part cooling fan is causing clogging in the nozzle. You can see this in the print above as gaps in the outer perimeter. But at the same time I'm seeing inadequate cooling on bridges :confused:. My fear is that the part needs a complete redesign and its my least favorite part of the whole thing.
    3. I think I see some part curling happening on my 40mm fan shroud. Its ok if it stops where it is, but if it keeps going the design is going to need a re-think.
    #1 I am confident I can address with a few test prints. #2 has a lot of possible solutions but they will take time to try out, mostly CAD time. #3 I will monitor as I make more test prints.

    I'm also neck deep into a re-write of all the tool changer g-code using the RepRap's expression language and a deep dive on machine noise. I swapped steppers and am messing around with spread cycle verses stealth mode. The above print was in stealth mode and I think that's why it has more ringing. This work is paying off but it means reliable printing on the machine is still a little ways off.
     

    Attached Files:

  14. dc42

    dc42 Well-Known Member

    Joined:
    Aug 16, 2016
    Messages:
    484
    Likes Received:
    216
    Do you observe much hot end temperature drop when the fan turns on? If so then the feedforward heater control in RRF 3.2 may help.

    [​IMG]
     
    garethky likes this.
  15. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Yes! I'm on 3.2 final now.

    I tweaked the nozzle design and re-ran the problem print and it was able to finish without issue. I even turned up the bridging fan speed from 50% to 80%. Here is the bridging top layer:

    IMG_5913.jpg

    But like you pointed out, I think the root cause may be with bad heater tuning on my part. I don't think I re-tuned when I upgraded to the duet 3. So I tuned the heater with the fan at 30%. And now I get stringing all over the place like its a totally different hot end! I had to go from 250 down to at least 240 to get the stringing under control. Its also now much more aggressive at recovering temps when the fan kicks on. I'll have to completely re-tune temps, retractions and fan cooling before I can say what is going on.

    But when I tuned I didn't know about Feed Forward and I didn't use the T parameter so it didn't use the fan at 100%. I need to tune it again! :D

    Also I managed to get the gear depthing in just 4 prints. The result is all of the clicking I was hearing on retractions is gone. That's a big relief for me. I'm sitting next to the machine and it is no more objectionable than my i3 MK3.

    And finally I am getting super flat solid layers with pressure advance set at 0.035. (you can see the flat area in the photo above)[/QUOTE]
     
  16. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    M303 T2 S250

    Screen Shot 2021-01-10 at 11.59.29 AM.png

    Sooo... that fan/nozzle combo can overpower an 80 watt heater cartridge :rolleyes:

    I think this means that one or both of these is true:
    1) I need to re-design that nozzle
    2) The fan I am using has too much CFM.

    I lean heavily towards #1 as 4.5CFM is supposed to be the right amount for a print fan.
     
  17. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Wins:
    • The stringing was wet filament.
    • I designed a new nozzle that 100% does not point a the hot end and was able to tune the heater without issue.
    • With that I was able to tune in PETG, PC and PLA filaments
    • Fan whine from the 5015 is gone by setting the PWM frequency to 10Hz
    • Tool head 1 was able to print all the parts for a second toolhead in both PETG and PC.
    • Got the bearing clearances dialed in so the hobbed gear contacts the bearing and is centered on the filament path.
    Outstanding issues:
    • The 5015 fan impeller resonates at specific frequencies making a horrible noise. I need to explore soft mounting the fan.
    • The new nozzle still doesn't provide optimal cooling for PLA. Curling and distortion in overhangs are visible on some prints. Further design iterations are required there.
    • The hot end fan shroud is twisting upwards slightly. It doesn't impact functionality but its not expected. Still more work to be done there.
    Print settings:
    I can't get retractions longer than about 0.5mm to work well. This seems to be a property of the Mosquito melt zone. What does work is slowing down the retraction speed over a shorter distance. I'm using these settings:
    PETG: 0.5mm @ 25mm/sec
    PC: 0.35mm @ 20mm/sec
    PLA: 0.2mm @ 40mm/sec

    Pressure advance is set to 0.03s for all filaments. This gives the best top layers with no noticeable over or under extrusion.

    For PETG I dropped the max fan speed to 40% but left it at 100% for PLA and 20% for PC.

    The 2 blobs on the stringing test are where I turned off retractions. In the bridging test you can see deformation of the 45 degree tower sections due to uneven cooling.


    Nozzle Flow Visualization.jpeg PETG Bridging.jpeg PLA Stringing test.jpeg PLA Bridging Test.jpeg New Nozzle and fan shroud in PC.jpeg
     
    #77 garethky, Jan 13, 2021
    Last edited: Jan 14, 2021
    Joe Pomo likes this.
  18. Wiley

    Wiley Member

    Joined:
    Dec 28, 2016
    Messages:
    2
    Likes Received:
    0
    Hey, just wanted to say I'm finally printing what you have in the development branch on Github to try this out. Thanks!

    One quick question - I'm still on a Duet 2, which would mean I need to find a different 40mm 24v fan, right? It looks like the Duet 3 added a 12v rail for fans?
     
  19. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Yes, Duet 3 has jumpers so you can select 12V or 24V and even mix and match on 1 tool. You can still use the stock mosquito fan.

    I really need to update whats in that dev branch! (I've been busy working on bed probing for magnetic beds.)
     
  20. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    39
    Update the dev branch with the lastest versions of the parts. This is the last update before I do some more major modifications. Upcoming dev branch experiments include:
    * Change the 5015 fan mount to use the silicone fan mounts.
    * Do a real part cooling fan. I would love to hear some suggestions of ones that work really well!
    * Beef up that 40mm fan shroud.

    I also have some glass filled nylon on the way to experiment with printing the hot parts in that material to see if it has better dimensional stability.
     

Share This Page