To put together the little circuit board you will need to make it first. Start by printing the strain relief so you have the shape to trace onto a PCB prototyping board (you can just print the first 10 layers if you want to save time), you can pick up a board from your local electrical parts supply for a few dollaroos. Just make sure you get the PCB board where every through hole in the board has its own sperate copper pad (so no copper connection between multiple holes), we are just using the board to hold the connectors and make it easier to do the solder work later. Originally, I was going to design a double sided PCB circuit board to etch out, but in the end I determined it would take less time to make the 4 boards I needed by hand than create a circuit diagram and etch each board (and a lot cheaper).
Using a standard pencil, trace the shape of the strain relief onto the PCB, try to get it lined up so all the connectors can go on easily, this may take a couple of goes hence use the Pencil to begin with to make it easy to remove (fine felt tip pent and some alcohol to remove also works well). Once you are happy with the layout (remembering the side of the board with the copper will be facing down), use a fine tip felt pen to trace the shape you need and cut it out.
Once you are marked up and everything double checked, trim the bulk of the unwanted PCB off, I found a good set of side cutters will get you pretty close then follow up with a disc sander (Dremel) to sand down the last of the PCB and get nice round corners and a perfect fit. Once you are happy make sure you clean up the board well with some alcohol.
Next thing to do is mount all the connectors, the main one to get right is the larger 10 pin connector across the back, make sure you will be able to get a good solder connection on the back for the pins. Go through and fiddle with placement of the connectors till you are happy they are all where you want them remembering they need to be inserted into the PCB board from the side without the copper surface. Pop all the connectors in then flip the board and place t down on some foam wrapper or a towel to help take up the variance in connector height.
Solder each connector into place ensuring it is firmly pushed up against the PCB, you only need a small amount of solder so don’t go crazy. If you are not experience in soldering, make sure you use a nice fine tip soldering iron and plug in the connectors matching plug when soldering the pins to prevent any warping from you accidently adding too much heat.
Next is to start wiring up the connections, I used gauge 24AWG silicon coated Turnigy brand wire from hobbyking, this stuff is amazingly good and also aluminium so it won’t break easily from flexing, for this as only the heaterblock will be drawing any current and that will be handled by the bridging traces on the PCB directly so the 50W will not be an issue((I used 16AWG of the same wire for the effector umbilical cord to prevent wires breaking due to the flexing of cable on delta setup). Start from the centre of the 10pin connector and work your way towards the edge. If you have placed your connectors well then you will find the 5V and Hotend power lines can be supplied by bridging some of the PCB points together saving you some additional space to make fitting the wires into the strain relief later much easier. If you decide to bridge the PCB for powering the heaterblock and 5V supply, you will want to do this before soldering any other wires. This is not really necessary, but it does neaten everything up (and I like neat and tidy 😊). Using a multimeter, check each connection as you solder it to ensure it is working as well as making sure you check/ensure the connection you just soldered has not short-circuited to any other pins.
Once everything is soldered up, flip the board back over and add some glue around the edge of each connector to glue it to the top of the PCB, I personally used UV sensitive glue but I also recommend a good epoxy. Do not use CA as it is not tolerant of the heat and will likely just crack away after a couple of uses. We are adding this glue as the PH2 connectors have a bad habit of coming off with the plug leaving just the 2 pins behind.
Once the glue is dry you can test the board to make sure it is working then add some temp plugs into the connectors and give it a quick spray of paint to match the colour of your strain relief. While you wait for the paint is dry, add the strain relief to the effector (suggest you add the fan duct while you are there) and if not already done, add all the other electronics with the cables running free so you can cut them to length and add the plugs. Once the everything is mounted up in the effector add your new board you and secure it in place with some glue (use the same stuff you used before), it is critical you get some glue front and back of the main 10pin connector and on the front edges of the PCB (where the semicircle ends). You don’t need much, it only needs to be on the edge of the PCB and strain relief, it only has to be able to handle the force of removing a plug which is very small to begin with.
With the glue dried you can make up you plugs and route the wiring, this of course is much easier to do with the effector removed from the printer. Once the wiring is done, insert the pneumatic coupled (must be a pneumatic coupler that allows the PTFE tube to pass all the way through it) into the strain relief, with the last step being to add the clamp to hold the umbilical cord in place.
Once you have mounted the effector back into the printer cut your PTFE tube off perfectly square (super important if using the standard hotend) and insert the PTFE tube through both couplers. You want to get the lower coupling sorted out first as this is the one you can’t afford to move on a standard hotend and still causes issue on an upgraded hotend if not fully constrained. Make sure you pull up on the pneumatic coupler collar on the bottom coupler when inserting the PTFE tube, I find a set of tweezers works best for this (don’t rely on the little clips to do this part, they do not provide enough force pulling up on the coupler). Once the lower coupling is nice and secure, add a coupler clip to hold the tube in place (suggest you print one of the larger versions of the pneumatic couple clips to ensure maximum grip of PTFE tube). Once you are happy with the bottom repeat with the top couple, make sure you use plenty of downward pressure on the PTFE tube as technically you want some slack between top and bottom couple (you don’t want the tube to bend or warp, but you also don’t want any tension between top and bottom coupler either). When you are happy add another couple clip to the top pneumatic coupler.
This should then allow you to change out the hotend without taking the effector off or having to rewire anything. It is generally much easier to change an entire hotend than trying to change a nozzle (I have several different size nozzles setup in identical hotends, I then swap the entire hotend out to swap nozzle sizes, 2 bolts and 2 plugs and you are done, takes like 1 minute to change a nozzle/hotend). It also allows you to change over effectors super quick as there is 2 bolts and one large plug to release the umbilical cord from the effector and another 6 bolts to release the effector (I also have a number of different effector setups and this system means it is 6 bolts and one connector to change the entire effector, less than 5mins to make and entire effector change over). This resolved my issue with using a crappy excuse for effector connection (mag balls) and the need to use a so called “smart” effector, not sure if the names is meant to be a cynical play on words as there is nothing “smart” about them.
On a side note:
When it comes to easy effector changes there are a number of vendors shamelessly pushing their mag ball solutions on all the forums as a fast way of changing out the effector and they are generally pushed to be paired with a Duet “smart” effector or some other variant. The not so “smart” effector has so many design floors, ranging from having to flex in order to work which causes issues with retractions (bashes the nozzle into your print on retractions), Unable to make hot nozzle calibrations, the fact it is 10 times the size it needs to be which cuts down you print area substantial due to it increasing the overall size of the effector (going to smash into the towers before reaching full print diameter and print diameter is at a premium on a delta), it is made out of PCB material so it has very little stiffness allowing the whole effector to twist under load or fast direction changes (same reason you don’t use 3D printed effectors or 3D print structural or mechanical parts for your 3D printer), the arm connection points are over the top of the effector which reduces the constraint and puts the effector on and extra leverage point adding more load to the point which increases the warping/flexing of the effector while printing (especially with mag balls) and also makes it harder to get the connection points square/even as you are trying to aligning 3 axis instead of just 2 which is made worse again by the PCB material as it compresses under load (good luck getting all the connection points at the same height) and we won’t get into the ludicrous lack of thought to cooling fans.
I think a lot of these problems are attempted to be hidden by using mag balls as they won’t bind if things are not square (which is bad). The fact of the matter is 99.99% of mag balls are simply rubbish, the 0.01% that are not complete rubbish only semi-constrain less than 20% of the joint (less than 5% of the actual motion), they move around in their constraint/socket on every single motion change despite the yelling down of anyone that suggest this is the case by their vendors. They are after all just relying on a varying magnetic field to hold the joint in place which is NOT a solid connection (can’t beat physics, you have magnetism holding a joint together, duh, of course it moves), even the slightest bit of additional pressure on the joint (away from the attractive force of the magnets attraction) causes some movement apart.
The problems with mag balls are easily evident by getting 2 identical effector setups side by side, one with mag balls and one with rod ends, powering up the motors to actively hold the effectors position at “home” and then grabbing the effector and giving it a wiggle, the difference is night and day between the 2 setups with the rod ends allowing you to move the printer around without the effector having any wiggle/movement.
The rod end design on a delta is the only way to go, it gives a significant improvement to overall dimensional accuracy, improved print quality and the ability to run much fast print speeds and accelerating before showing artifacts in the print compared to mag balls (and less weight). This is because it completely constrains the joint, there is simple nowhere for it to move, this will also show you if you are out of square as the joints will bind/not line up if anything is out of whack. Just to be clear, we are talking about proper rode ends/heim joints not crappy traxxis joints. If people spent a third of the cost on rod ends as they do on one of these “pushed” mag ball kits you will get something 20 times as smooth and 1000 times more constrained than the best mag ball setup which will result in a significant bump in every way to your print quality.
I guess in some ways I should not be so harsh on their creators as most of them seem to only have every owned one or 2 printers and only a single mid-sized delta so they don’t have the experience or access to actually do some proper testing, but is that really an excuse considering the money they are charging people………