BMG + Mosquito + Tool Changer

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

  1. Greg_The_Maker

    Greg_The_Maker Administrator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    1,035
    Likes Received:
    633
    it's 465g with the toolchanger parts, ie the hermes tool is 465g.
     
    Ntesla66, mhe and Amr like this.
  2. Ntesla66

    Ntesla66 Active Member

    Joined:
    Sep 9, 2019
    Messages:
    34
    Likes Received:
    32
    Just finished up the V6 +BMG , here's its weight and how it's mounted which isn't any different from the Mosquito. I think if I made an actual tool plate as opposed to an adapter to tool plate I could cut a fair amount of weight out of this arrangement. @Greg Holloway , are you going to sell the TC-Ramp as an item for those of us who can mill aluminum? scaled9-min.jpg scaled10-min.jpg
     
    Diyar Tatar and Greg_The_Maker like this.
  3. Greg_The_Maker

    Greg_The_Maker Administrator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    1,035
    Likes Received:
    633
    Good question. I shall see what can be done.
     
    Jai Stanley, Ntesla66 and mhe like this.
  4. Winand

    Winand Member

    Joined:
    Jan 6, 2016
    Messages:
    12
    Likes Received:
    0
    @Ntesla66 any chance you want to share your files?
     
  5. Ntesla66

    Ntesla66 Active Member

    Joined:
    Sep 9, 2019
    Messages:
    34
    Likes Received:
    32
    Yes, I will share what I've got so far. Build, use and break at your own risk! ;) I'll get around to uploading the stl and step files to T-verse tomorrow ... Well... it'll Benchy . Not a great benchy, but way better than I've gotten before. The yellow is with the V6+BMG and the grey is with the Mosq+BMG. Printed at 60mm/sec. I would really like to have better control over how much/many far/times the bed drops between tool changes. benchy1_scaled-min.jpg benchy2_scaled-min.jpg
     
  6. Winand

    Winand Member

    Joined:
    Jan 6, 2016
    Messages:
    12
    Likes Received:
    0
    By introducing the Aero direct tools I also had some compatibility issues with the different docking depths. I printed a 6mm spacer that fits between the toolplate and the PTFE part on the V6 bowden tools to fix the problem.

    By introducing Brendon's Aero direct tools I also had some compatibility issues with the different docking depths. I printed a 6mm spacer that fits between the toolplate and the PTFE part on the V6 bowden tools to fix the problem. Maybe you can do something similar for your bowden tools?
     

    Attached Files:

  7. Ntesla66

    Ntesla66 Active Member

    Joined:
    Sep 9, 2019
    Messages:
    34
    Likes Received:
    32
  8. Cam_Motion

    Cam_Motion Active Member

    Joined:
    Jan 6, 2019
    Messages:
    30
    Likes Received:
    4
    It still looks like it's the Z axis that's the problem, I'm hoping to see the setup you use with your dynamic signal analyzer.

    Has anybody tried printing the yellow areas in non planar g code?
     
  9. Ntesla66

    Ntesla66 Active Member

    Joined:
    Sep 9, 2019
    Messages:
    34
    Likes Received:
    32
    I'll certainly show you when I get it set up but until I'm satisfied that the modifications I make to the platform are beneficial; I'll not be setting it up. Good news is that I'm getting pretty close to satisfied. This was printed last night with no purge nor wipe movements although it would have benefited from some form of wipe. I slowed it down to 35mm/sec. I'm still not happy with the PCF. If I can get accurate multi-extrusion parts that take a little longer, I'll take that over inaccurate parts printed at screaming speed. After the motion system is dialed in it's really rock solid on repeatability and the tool-plate/tool-changer mechanism is just a damned fine piece of engineering work! The second pic is after I cleaned it up a little. scaled_lattice1-min.jpg scaled_lattice2-min.jpg
     
  10. Cam_Motion

    Cam_Motion Active Member

    Joined:
    Jan 6, 2019
    Messages:
    30
    Likes Received:
    4
    It looks very good, you have the multi tool calibration dialed in anyway. I'd be interested in your thoughts on what can be improved.
     
  11. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    Subscribed! I have BMG/Mosquito parts and I want to CAD up a more compact tool lead. Hopefully get to work on it next week.
     
  12. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    So this is the arrangement I imagine we can use. The nozzle ends up being in about the same spot as the V6 tool. The dock can be re-used without changes. The filament path is very short. I'm not sure yet of the right approach to design the plastic parts that go around this to hold everything. There is room to put a part cooling fan on pretty much any side of this.

    I'm still trying to come to grips with joints and such in Fusion 360.
    Mosquito-Tool-Parts-Fit.png
    Mpsquito-Tool-Parts-Fit-2.png
     
    Spoon Unit and Ntesla66 like this.
  13. Ntesla66

    Ntesla66 Active Member

    Joined:
    Sep 9, 2019
    Messages:
    34
    Likes Received:
    32
    Keep going!
     
  14. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    I switched over to OnShape. I can't seem to play nice with Fusion 360. The project is here.

    So I was thinking that a back-plate and a front-plate, like the BMG design, would be the easy way to go. But the back-plate I designed is really unprintable without a lot of support material/bridging. So I'm having to re-think how best design this for print-ability. I really want that stepper close to the middle of the Tool Plate to support the majority of the mass but that put it in the way of a lot of things. the filament path will collide with the pins on the dock, the stepper hits the dock etc. I'll keep plugging away at it.

    Screen Shot 2019-11-17 at 6.08.13 PM.png Screen Shot 2019-11-17 at 6.08.32 PM.png Screen Shot 2019-11-17 at 6.08.45 PM.png
     
    Ntesla66 likes this.
  15. Jai Stanley

    Jai Stanley Well-Known Member

    Joined:
    Aug 28, 2019
    Messages:
    115
    Likes Received:
    60
    +1 vote for this Greg
     
  16. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    Printed my first test parts for the design this evening. I didn't expect these to work, just wanted a gut check on tolerances, print-ability, supports etc. The biggest thing I learned is that the NEMA 17 frame can't fit between the back wall of the enclosure and the stock TC plate location. The TC plate has to be moved forward by about 8mm from stock. This is why printing test parts is so great, its so easy to be mislead by the computer.

    So I'm maybe 60% the way there now. Another 10 hours of CAD time and 5 test prints or so to get it dialed in.

    Also I picked up a second Mosquito from Slice for Black Friday, so I'm pretty committed to this project now :D

    IMG_4876.jpg
     
  17. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    10 hours, LOLZ :eek:. Happy New Year everyone, just wanted to post some progress.

    I spent the 6 days I was on vacation glued to a screen doing CAD. I'm printing prototypes this week. Still dialing in the diameters of all the holes and adjusting things for print-ability, assembly etc. But here it is mounted on the tool changer showing that it does actually fit. And it looks like the nozzle is about 2mm ABOVE the stock one! I can use those 2 mm for some longer threaded inserts so I should be able to get it to match the stock one nearly exactly. Its got an upturned cooling fan vent, much better tension arm design and external cable management zip tie loops.

    I will update later this week when I have it successfully printing and put the SLT files up on Git Hub.

    IMG_4930.jpg IMG_4931.jpg Internals-front frame hidden.jpg
     
    mhe and Greg_The_Maker like this.
  18. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    Ok, here is it! STL files and all the info to make your own are up on GitHub: https://github.com/garethky/Slice-Mosquito-Direct-Tool-Head

    IMG_4947.jpeg IMG_4949.jpeg IMG_4948.jpeg
    There is a standard sized stepper attached because Duet3D says my pancake has a short :(. Pancake testing will have to wait for later. But weight with the pancake was 375g.

    It printed a fine enough Benchy using the gcode included with the printer firmware. No issues with filament drive and this hasn't been optimized for the direct drive setup yet. Very minimal ringing in X, there is a small amount of flex with the Mosquito mount that I could not eliminate and that might be the root cause. But keep in mind this was printed in ~40 minutes which is shockingly fast to me, with a slower speed on the outer perimeters it would probably go away.
    Benchy.jpg
    It does show obvious signs of insufficient part cooling. Maybe the fan shroud can be improved to fix this, or maybe we need a bigger blower. What do you all want?

    I printed a lot of prototypes to try and find all of the bugs for you:
    IMG_4941.jpg
    If you make one I would love to hear your feedback.
     
  19. Jai Stanley

    Jai Stanley Well-Known Member

    Joined:
    Aug 28, 2019
    Messages:
    115
    Likes Received:
    60
    I might have to get a mosquito and nema 14... The extruders I've been prototyping work; but after moves with linear advance I get the occasional missed steps and extrusion gaps. Not sure whats going on; but I think the motor is too tight on the gears (printing one with clearance atm). If this doesn't work; I'm trying yours!!

    Great work!

    Jai
     
    garethky likes this.
  20. garethky

    garethky Well-Known Member

    Joined:
    Oct 8, 2019
    Messages:
    87
    Likes Received:
    40
    Thanks Jai!

    I pulled the depthing of the gear train from the BMG and got 17mm center-to-center from the stepper to the reduction gear. It does have a small amount of backlash. Getting the tension right is also pretty critical. I kinda like Pursa's spring better because its longer and softer than the stock BMG one for a wider sweet spot. But I couldn't find a source for them so, literally the last change I made was to use the stock BMG spring and thumb screw.
     

Share This Page