Ugly Parametric Magnetic Effector by rodsmith

Last crawled date: 3 years ago

Quick Start
This Thing is a highly parametric effector for use with magnetic diagonal rods on delta-style printers. It consists of three parts, each of which has 2 or 4 variants (in .stl form):

Effector-{hotend}-{probe|noprobe}.stl -- This is the effector itself. I provide versions for the E3D V6 (and clones) and for the Hexagon hotend. Both variants include versions with and without a mount for a bed probe, such as an inductive sensor.


Lock-{hotend}.stl -- This is a small triangular piece that locks the hot end in place.

FanDuct-{hotend}.stl -- This is an optional 180-degree fan duct for a 50mm centrifugal (aka "blower" or "radial") fan.
If you have an E3D V6, a clone of it, or a Hexagon hotend, you should be able to print one set of three parts (or two parts, if you don't want to use a part cooling fan) and go; however, you'll need to find a carriage with a matching 51mm inter-ball distance to use the provided .stl files. (My carriage, "Yet Another Kossel Magnetic Carriage," https://www.thingiverse.com/thing:2833905, comes with such a variant.) If you need to use the effector with a carriage that uses another inter-ball spacing, you'll need to customize this effector, as described later.
Non-Printed Parts
To use this effector, you'll need a number of non-printed parts, in addition to the hot end:

6x magball studs -- These are often sold with magnetic arms from vendors such as UltiBots and Filastruder. You'll normally use matching studs on the carriage.


7x M3 nuts -- You need six nuts for the magball studs and one for the fan's mounting screw.


2x - 4x M3x16 screws -- These are needed for the hot end fan. Optimum length depends on the fan's depth, but 16mm works for most fans.


2x M3x6 screws -- These attach the fan duct to the effector.


3x M3x10 screws -- These attach the locking nut to the effector. You may be able to get by with M3x8 screws, particularly on the Hexagon variant.


1x M3x25 screw -- This screw attaches the centrifugal fan to its fan shroud.

Fans -- The Hexagon version uses a 25mm hot end fan, and the E3D V6 version uses a 30mm fan. (These sizes match what comes with each hot end.) The part cooling duct is designed for a 50mm centrifugal fan.
Of course, if you don't use the part cooling fan duct, you won't need the three screws or the nut associated with it.
Features & Caveats
I designed this effector with several goals in mind:

Parametric, to enable easy adaptation to different hot ends and diagonal rod spacing.


Support for a part-cooling fan.


Support for an inductive bed sensor. (This turned out not to work too well; see below.)


Designed for use with magnetic diagonal rods.


Low attachment point for diagonal rods, to maximize print height; but not so low that the view of the hot end's nozzle is seriously obstructed or that the heat block would come into near-direct contact with the effector's plastic.

No need for supports or rafts when printing. (The effector includes two concentric hollow cylinders that function as built-in supports for the top piece. These are easily removed after printing.)
This effector does meet these goals; however, there are several caveats:

The centrifugal fan protrudes significantly from the effector, which robs about 10mm of print area on one side. This figure may be greater or less for you. Using a 40mm centrifugal fan should reduce this loss; but the 40mm fans I have are thicker than the 50mm fan I have, so this would require further tweaks to the design, likely including a wider inter-ball spacing to make room for the mounting screws.


In use on my printer, the bed sensor's trigger height varied significantly depending on the point on the bed being probed. This greatly reduced the bed sensor's usefulness, to the point that I switched to a Delta Smart Effector. Using Force-Sensitive Resistors (FSRs) under the print bed rather than an inductive sensor might have worked as well.

There seems to be no standardization of inter-ball spacing for magnetic diagonal rods. I used 51mm spacing to match an earlier effector design I tried. Thus, customization of the OpenSCAD file and rebuilding the .stl file, as described in the next section, or doing the equivalent for the carriage, may be necessary for most people.
Customization
I designed this effector to be highly parametric, so it's easy to customize it. I used OpenSCAD (http://www.openscad.org), which is free software, so you can customize this Thing quite easily with a minimal understanding of OpenSCAD. The main file is Ugly-Parametric-Magnetic-Effector.scad; it contains designs for all three parts. Variables that you're most likely to want to change are on lines 6-48:

makeSensorMount -- Set this to true to build the effector with the sensor mount, or to false to omit the sensor mount.


makeWireTieDown -- Set this to true to add a half-torus above the hot end fan mount. I intended this for helping to secure wires, but it proved to get in the way more than help, at least for me. You might decide otherwise.


recycleNum -- Set this to a number greater than 0 to add a recycling symbol and number on the effector and fan duct. Note that this feature requires the recycling_symbol.scad file from the OpenSCAD Recycling Symbol Library (https://www.thingiverse.com/thing:216963) to be in the same directory as the other .scad files.


ballDistance -- Set this value to match the ball spacing on your carriage. The ball spacing on the effector and carriage MUST match! Greater ball distances can increase stability, but too great and the ball mounts will physically detach from the main base and/or the ball mounts from different sides will begin to overlap.


baseLength -- Setting this shorter will bring the mounting balls closer to the hot end without changing the distance between parallel balls. Shorter baseLength values increase stability, but too short and the effector becomes structurally unsound and/or the ball studs won't fit in place. Also, a baseLength much shorter than the default of 51 makes it impossible to mount the fan duct to the effector. (Note that this variable sets the size of the triangle that forms the basis of the effector's base.) The baseLength and ballDistance values are normally similar -- they're both 51mm in the default configuration and .stl files I provide.


makeEffector -- Set this to 1 to build the effector model with a Preview (F5) or Render (F6) operation.


makeLock -- Set this to 1 to build the lock model with a Preview (F5) or Render (F6) operation.


makeCentrifugalFanShroud -- Set this to 1 to build the fan duct model with a Preview (F5) or Render (F6) operation.


sensorDiameter -- This sets the inner diameter of the Z probe sensor mount.

sensorBrimDiameter -- This sets the outer diameter of the Z probe sensor mount. It must be greater than sensorDiameter.
In addition, lines 27-29 include the .scad files that hold measurements of the hot end; you must uncomment the line for the hot end you're using and comment out the rest. If you want to build a version of this effector for another hot end, you can create your own .scad file for it and include that file. Use one of the existing hot end files as a model. I provide three sample files, for the E3D V6, the Hexagon, and an effector that looked like a Hexagon clone on eBay but that turned out to be quite different.
Additional variables in lines 6-103 may interest you if you have exotic needs. The heFanSize variable, in the hot end .scad file, sets the size of the fan used by the effector.
Work in Progress
I designed this effector for a homebrew Kossel XL; however, I was unsatisifed with the performance of my inductive bed probe and so decided to buy a Delta Smart Effector. Thus, I've abandoned this effector, at least for the moment. I put a lot of work into it, though, and thought it might be of interest to others -- particularly those who want to build delta printers using Hexagon or more exotic hot ends. I have used this effector with both
an E3D V6 and the Hexagon not-really-a-clone. Although I've printed the version for the genuine Hexagon to see that it fits, I've not actually printed with it.
The main problem I had with the inductive bed probe is that, because of varying effector tilt at different points on the bed, its readings were not consistent, compared to actual nozzle height. This necessitated tedious measurements to make the RepRapFirmware's G32 command produce good results, and G29 never produced good results for me. You may have better luck with force-sensitive resistors (FSRs) under the print bed, or simply doing manual bed leveling. Perhaps another firmware, or a future version of RepRapFirmware, will work better even with an inductive sensor. It's also conceivable that other physical design aspects might improve the reliability of an inductive sensor.
Shortly before buying my Smart Effector, I began experimenting with a design to put a smaller 40mm box fan above the main body of the effector, so as to recover the print area lost to the centrifugal fan. You can see what this looks like by setting makeBoxFanShroud = 1 on line 19. I never printed (or even completed) this design, since it looked ridiculously top-heavy and would have blocked the hot end's filament entry hole. I've left the code in the .scad file in case somebody else might want to modify it into something useful.