toolhead switching with an electropermanent magnet


We’ve been doing some really cool research into toolhead switching options in the big-thompson research project ( The concept is that a printer could have a rack of toolheads (different colors, flexy’s, MOARstruders, etc) and could actively set down and pick up different toolheads during the print similar to CNC tool changing.

We considered using an electromagnet to pick up the toolheads, but this would become an issue if power failed and you dropped your fancy toolhead. Instead, we’ve been playing with the ultra cool OpenGrab board ( It’s an electro-permanent magnet that can be “switched” on or off, and requires no power input to maintain a surprisingly strong holding force.

We’ve got a quick and dirty prototype set up that combines the opengrab with a maxwell style kinematic coupling that should make the pickup sequence super repeatable. The holding force is more than enough to tightly secure our heaviest toolhead (the dual extruder) well enough for reasonably fast printing.

Pretty stoked about this :slight_smile:


COOL! May be there will be a small demo at the Open House? :wink:

Huh, neat! so the board powers on to repel the permanent magnet when you want to switch, but holds when power is off? I like it. What temperature does the magnet start to loose attraction at? Any issues with it being in a heated enclosure?

It’ll definitely be demo’d at the open house, along with absolutely everything else we’ve been working on. Open source is cool like that :slight_smile:

Right? It actually can switch from being magnetic or non magnetic using a small amount of power (5v and ~3A for a second or two) and requires no input power to stay in either state. My understanding of how it works is that it contains a permanent magnet and a material that can be easily magnetized to either cancel out or reinforce the permanent magnet’s effects. You can read more about it here: I was incredibly skeptical of the tech until I saw it working, but it’s one of the coolest things I’ve ever seen. Shout out to Andreas Jochum and Nicadrone as well as Zubax robotics for fighting the good fight and making this technology available under the GPL.

I’m not sure on what temperature it starts to loose magnetism at, that’s definitely something we need to test. I know that permanent magnets will start to demagnetize at around 80C, but the low coercivity material might act differently.

Oh wow, that is cool. I can see a couple applications for that. A removable bed plate comes to mind too.

Do you have a gauss meter on site to see how far the magnetic pulse goes when switching it on or off?

This gives me a couple ideas.

The coupling method is very clever, absolutely.
But I’m not too optimistic about the real world application of an exchangeable extruder. If you have a multimaterial or multicolor part that has both materials or colors in nearly every layer, the head change will increase print time dramaticaly especialy on not too big layers.
Do you plan to implement electrical connections in the park positions so the hotend can keep it’s working temperature?

To throw in another idea: What’s about a dual nozzle, multi material switching extruder? Two nozzles could print at two temperatures, each could be feed by one stepper with a switching servo that engages / disengages 2 or even more different filaments per nozzle… :sunglasses: The idea is there (Dondolo, only to name one), it “only” needs someone that has money and time to develop a stable mass-compatible solution :wink:

Edit: Another idea for the toolhead switching (I know this is not trivial or cheap, but who cares atm…). Make the parking position capable to do an automatic filament change with some kind of slots for 2 or more filaments. This way we could get a fast color changing system: While one tool head is printing with color one, the parked toolhead is loaded with the next color needed. In this case, we could easily print with a lot of different colors and the tool head changing time is also no problem anymore because we save the time needed for a prime tower compared to usual dual color-single nozzle systems. That would be awsome!