JAW Delta Printer Kossel Mini Remix by JAW

Last crawled date: 3 years ago

11/20/18: The Delta Printer is up and running, and it works quite well. I'll upload videos and a BOM for anyone who wants to construct their own very shortly.
This is running list of the ongoing delta parts I am creating/remixing for a custom 2020 Mini Kossel build running on the Duet Wifi and using 2020 profile extrusions from Robot Digg. This thing is a work in progress, and will be updated as I add parts and pictures to my build.
CURRENT LIST OF PARTS:
Vertices: These are remixed from Sneq. They are unique in that they better seat into standard v-slot profile to aid in creating a stiffer frame. They also feature cutouts that ease warp that happens when you print larger parts in ABS and PETG. Printed (2mm layer height, .4mm nozzle, no supports, PETG)
Upgraded Max Z Effector: This is a JAW original design. The effector was conceived with several goals in mind. Mainly, I wanted to create an effector for the popular E3D V6 hotend that would raise most of the part above the effector plane. This is desirable to two reasons. First, having the hotend closer to the effector plane increases print accuracy. This is because the hotend is like a pendulum hanging below the effector plane. Any movement or slop from the ball ends, arms, and rails of a delta printer is magnified the further it is from the effector plane (just like the pendulum of a clock swings a further distance the farther it is from its pivot point). By moving the hotend closer to the effector the print becomes less affected by small movements in the ball ends, rods, linear rails, or even vibration in the frame. Secondly, raising the hotend gives more print height. However, there is a tradeoff associated with raising the hotend above the effector; you have to increase the size of the effector to accommodate cooling fans, sensors, and the coldend while maintaining clearance for the parallel rods to move the effector to all areas of the printbed. If the effector becomes too large, you may end up losing x/y print area (negating any gains in z print area). This effector simultaneously raises the hotend while keeping the effector relatively small to maintain x and y print area (hence the need for 30mm fans). I strongly recomend printing in PETG or ABS or high temp PLA as the hotend cartridge will be in close proximity to the part (you may want to insulate the bottom of the effector as well). Printed (2mm layer height, .4mm nozzle, no supports, PETG)
Carriage: Remixed from Boksbox, this carriage incorporates GT2 belt clips that lock the belt to the carriage rather than to itself. These clips have more plastic support than most of the carriages where the belt doubles over to lock into itself, making them less likely to crack the plastic under tension. The carriage also has an adjustable M3 screw at the top that triggers the x,y,z tower endstops. This is an easy way to adjust the printer to make sure it homes the effector in the center of the frame (it saves you from moving your endstops or adjusting the tower to accomplish the same result). Printed (2mm layer height, .4mm nozzle, no supports, PETG)
Endstops: An original JAW part, there is nothing really special here (at the moment anyways). I am not sure there is much to be innovated on with the endstops; however, I have seen a few people incorporate endstop holders with frame braces, and that is pretty cool! So. . . that might be an update coming very soon! Printed (2mm layer height, .4mm nozzle, no supports, PETG)
V-Slot Insert/T-Nut: This is another JAW original part. These are nothing too special, but I wanted to print these in groups of 10, and on end so no support is needed. I do plan on adding a drop-in nut holder soon, so stay tuned for that! Printed (Slow 25mm/sec speed, 2mm layer height, .4mm nozzle, no supports, PETG)
GT2 16T Idler Pulley: This is yet another JAW original. The part was created to work with 2 683zz bearings per idler pulley (6 in total). However, 16T metal idlers are a part that can be purchased, and buying them is honestly not much more than 6 bearings ($2 more on Amazon, plus they will be stronger). So I would recommend purchasing idler pulleys unless you have some 683zz bearings sitting on the shelf (which I do!). Printed (Slow 20mm/sec speed, 2mm layer height, .4mm nozzle, zig zag supports, PETG, print on end!)
PARTS COMING SOON!!!
-Extruder: I am working on a remix of the popular B-Struder. It will incorporate a bowden tube that runs all the way through the arm (however it will be cut where the bearing and MK8 hobbed gear meet). This will keep the filament from wearing away at the printed parts and it will keep still keep the filament path very tight to help reduce slope from the bowden setup. Almost all credit to tempo502, as the B-struder is already a great design. In addition to the through-bowden tube, I am adding some lever arms/grips that can be pinched to make filament loading and unloading easier. I am also considering tweaking the design to allow better access to the hobbed gear (just incase I need to clean up ground filament) and evaluating if I need to give the setup some type of mechanical advantage with the stepper motor I plan on using.
Frame braces: I am considering making frame braces, but am not sure they are necessary right now. Let me start by saying that the vertices I have used make my frame very stiff. I cannot detect any movement in the frame at all. However, all printers will have a frequency at which they resonate, and when the effector is making very small movements at this frequency this could become an issue that shows up in a print. I will also say that adding braces certainly wouldn't hurt anything, if anything they may just be a way to make the frame overbuilt (or better prepared for expansion). So I suspect that these will be coming soon!
Enclosure: I have a huge piece of acrylic glass sitting around and I am very interested in enclosing my mini kossel to improve print quality and reduce warping of certain thermoplastics like ABS. I have looked at a number of enclosures, and I think my inspiration may be taken from the one constructed by physics anonymous (go to YouTube and search 3D Printer Upgrade (Part 3)). Enclosing the printer will hopefully reduce warp as the heated bed heats the air in the chamber and reduces the effect of draft on the print. An enclosure would also keep dust off the linear bearings, and my cats away from a print in progress.
Dust covers for the bottom motors: These will help enclose the print area.
A power supply cover for my Meanwell powersupply to prevent shock.
UPDATES ON PROGRESS:
Update #1: I've been assembling my frame, and putting the vertices together with 10mm M3 screws is possible but very difficult with the current version of the vertices. I am updating the part to add a 1mm deep dimple around all the M3 frame holes to allow the bolt to reach deeper into the extrusion. If you have longer bolts (say 12mm M3's), then you would just add a washer or use the old files. I'll leave the old vertice files up, and label the new files with "new".
Update #2: I've been working on the effector recently. I updated all the.stl files associated with the effector (effector7, hotend clip, cooling duct). The design now includes ports for cord management for the fans, upgraded points for the parallel arm connections. I also added a few extra holes/hardpoints at the top of the effector to add a pcb board with connectors to easy wire fatigue and make swapping parts easier. Unfortunately my Printrbot is down due to wire fatigue issues with the z-probe, so I am not able to print until replacement parts arrive and are installed. In the meantime, I've added pictures of my build progress. The images show the vertices, endstops, and the effector. You can see in the image that the effector doesn't reduce the print area at all! Enjoy, and leave a comment if you need anything or have any questions.
Update #3: I have designed a 16 tooth GT2 idler pulley that uses x2 683zz bearings per pulley. All of the walls are at least .45mm, so I expect that it should print with a .4mm nozzle; however, my printer is still down, so be aware that this file may change.
Update #4: Printer is back up (I think). I have updated the effector with a better z-sensor holder. I settled on using the following inductive sensor: No. LJ12A3-4-Z/BY, PNP, 4mm, 6-36vDC.