I mean, hall effect switches are more than half a century old at this point, but I do agree that the rest of the concept is fairly innovative. I haven't been this excited about keyboard tech in a while.
@@DamnZodiak My dad said they had a few keyboards with magnet switches, with the old instalation and servers. When I got interested in keyboards, he looked at his work in storages to see if there were any left, or if someone kept a few, but unfortunatelly they were srapped. Really wanted to try them.
@@Riskable3DPrinting So in absence of springs for weighting, you can adjust the magnets in order to have desired weights/effects, yeah? I'm still trying to wrap my head around this, but it seems like you might have genuinely the greatest switch innovation of the last forty years. I hope a major manufacturer can work with you to get this standardized or something. I need this in my life.
@@malaythakkar8772 just need to buy a 3d printer, resin, and magnets. then learn the cad software... right? yeah, na. i think some people would rather just purchase them
When you make a new switch and you send it to one of the most brutal guys on RUclips, that guy haves some balls. Thanks for the video, there's some really interesting and skilled people out there and it will be cool if some get some recognition.
@Riskable 3D Printing Good for you for making a product that makes you confident enough to subject it to Thomas! From the looks of it it was a smash success. I hope you can commercialize some aspect of this.
@@gyozakeynsianism who pays $200 for only mx switches? The most expensive switches for the largest mainstream layout would end at about $120. The most popular switches cost significantly less. And gamers usually buy like $25 switches (gateron/cherry), not $200 switches.
@@user-kz9rz8ui1c Sure seems like $200 MX keyboards are still on the market (usually with lots of lights). And the custom market (which is more for typist snobs than gamers I think) is MX dominated for some reason, and those keyboards are even more expensive.
@@gyozakeynsianism oh you're talking about keyboards. Any mx style prebuilt that is over $100 is overpriced imo. The only prebuilt worth getting that costs over $150 is the apex pro, which isn't really mx style. Customs actually usually range much more expensive, but youre mostly paying for great build quality. You can pick up decent customs for less then $150, any more than that and you are paying for design, material, and weight.
The sheer level of and ease of modification to the user experience puts these leagues ahead of everything else on the market as far as fine tuning for the end user. I absolutely can not wait to see where this project goes!
He could even potentially design in adjustable magnet distances so you can adjust the tactility with something like a screw mechanism. That’s really the only improvement I can think of!
@@BlueTemplar15 basically, when you push down a key, at a certain point it becomes much easier to push. By adjusting where the magnet is, you could adjust at what point this happens, and how much of a difference in force there will be. The important part is that this could be adjusted by the user very easily, where normal moulded plastic key switches can’t
These are fascinating and clever as hell. I'd never heard of these before now, but they seem like a great idea. Just the adjustments in tactiliy alone blow me away...add to that the options in height of travel and this is some crazy brilliant stuff.
@@subterficial Even in a machine shop they'd be fine. The metal bits would just collect around the edges of the plastic and maybe a little bit on the parts of the stem that are exposed. They wouldn't impact the functionality at all. Not unless you start getting millimeters worth of metal piles under the plate :D
There are already many switches that use Hall effect technology which means they use magnets. They don't all gum up with metal so I don't think it would be a problem. In fact, hall effect switches are considered very reliable.
@@chubeviewer Well, it isn’t really “all the same”, otherwise why would people be interested in them. But I’m still very interested in this design. Much like the guy who built his own case out of stainless steel and frosted plastic (weighing the rather small keyboard in at over 2kg), I’m immensely proud of Riskable for trying to make this idea work. As a concept, it’s very cool, but it will really only stack up if it works in an actual keyboard.
The idolisation of cherry does tick me off, but one thing that I do appreciate and wish was an actual industry wide thing is how all the cherry clones have the same socket for keycaps
@@davidy22 I dont think cherry is that much idolized, its just the whole scene with hot swapabilitie and all kind of was sure to end in one switch design becoming the norm. Its not just they keycap mount, also the pins for the pcb. All these new switch designs really only make sense as full non customizable boards.
There's something just really excellent about the concept here. I hope a keyboard based on this switch is comes to market because the switch looks just absolutely brilliant. Bravo!
ive never been this excited about getting my hands on a switch since i first discovered salmon alps. This seems extremely promising, and I really hope this gets into production. Easily the coolest switch concept ive seen in a bit
This gives me so many ideas! Ive been wishing someone made an MX-compatible magnetic seperation switch for a while now, glad to see someone else is thinking down that track as well.
Well only the switch stem is MX-compatible. The switch itself needs a hall effect sensor or reed switch to actually function as a keyboard key (use hall effect though it's better and cheaper 👍). I only added support for MX stems to the switch but I have a placeholder in the code for adding Alps stem support too.
Agreed, if he has not filed a patent, or one has not already been filed for this design, then this guy needs to do so ASAP, and either partner with someone like Gateron, or start his own company.
@@CommodoreFan64 it is prob to close based on old patents to be protectable but as soon as something is public no one else can claim patent rights to the tech(in theory) meaning if someone could the switch would be safe
I had a similar idea but with coils or one coil and using an electric field to control the weight and tactility. You could create a dampened switch and control the sound by using a few tiny speakers. I don't really have the time to pursue, so I thought I might as well put it out here before some arsehat patents it. Oh, almost forgot, you could also use the coil to power the switch itself I suppose, for displaying options or the LED.
I already made *three* keyboards that use this design: A numpad, my Riskeyboard 70, and the world's first hand-wired analog keyboard. You can see all three in this gif I made yesterday (though the numpad doesn't have any RGB hehe): gfycat.com/fluidsevereairedale
I have several questions: 1. I've been wondering how smooth you could get a 3d printed switch, are these any good in that regard? 2. Can we get some force curves for these?
The switches are quite smooth because the print orientation results in layer lines that are like this: || || (they run parallel to each other in the axis of motion). For *extra* smoothness/less friction you can print the stem at a slightly different layer height (e.g. print the stem at 0.12mm and the sheath at 0.16mm). I never bother though because they're pretty damned smooth right off the printer. Regardless of how terrible your printer is you can always *make* the switches ultra smooth by just sanding them! Also, if you sand the bottom (flat) corner edges of the stem with something like 400 grit sandpaper that makes it extra smooth as well because it gets rid of the little crinkle that's possible from first-layer-squish (aka elephant's foot). It would wear off after a few hundred keypresses anyway though.
Here's a force curve I measured of a 0.0mm MAGNET_GAP Void Switch with 4x1.7mm N35 magnets by hand (lowering my 3D printer in 0.1mm increments on to a switch that was sitting on a digital scale): i.imgur.com/FNoTSpL.png (I think it was a switch with 3.5mm of travel)
How neat. With the analog HAL sensor you can measure the speed or force applied to the switch. Finally a possibility to measure anger level directly! Maybe even auto-shutdown to force you to go out if you're too heated?
So cool! I hope he gets in touch with an enthusiastic company that will help him commercialize this as a product. I'd love to install these switches on a nice keeb and give them a go.
Jesus, this is a switch family which can be either linear or have varying tactility while not only lacking friction on contacts but also the friction of the tactile element/clicker, AND IS ALSO ANALOG CAPABLE. It is the first of its kind.
hey this is kinda how the old school Tube Mod Vaporizers work. With two magnets of opposite polarity the button is open pushing on the button wont allow the magnets to touch so it forces the battery up into the contact. pretty nifty
@@KayvonJavid Yes it would! Absolutely! Haha. It's not as big a deal as you think though: My firmware can read 96 analog hall effect sensors in under 200 microseconds which leaves 800 or so to send a strong electromagnetic tactile jolt to stay within that 1ms polling window 👍
@@NickShabazz probably it also has analog capability too. But man the life time of this switch design could be well beyond billions because there really isn’t anything to wear out other than the electronics
I am 100% going to pick one of these up when or if they become available: Even if it turns out I have to buy a set of keycaps with them too. Brilliantly designed!
I see little reason for this design to not go with MX style mounting. I feel like the mounting system is one of the key reasons MX style has become so standard. And its not like its a bad system or anything.
This is extremely interesting and exciting. This is real innovation, and I can't wait for more work to be done on them and for them to be implemented fully. We've just witnessed key switch history.
That's _really_ interesting. that one design makes a switch on any point of the spectrum of slightly to very tactile, or from low to mega-mega-high profile, is awesome Also, have you seen his clicky keyboard? It uses _relays._
My Riskeyboard 70 is actually a microcontroller board in its own right... So hooking up relays was a cinch since all those extra pins are exposed at the back of the keyboard for hooking up things like that. I also added a QWIIC connector but have yet to try it out with anything yet. Was thinking about hooking up a color sensor board so you could hold up a color swatch (like you'd get at the hardware/paint store) and it'd automatically type out the HTML hex code for that color. Would be a cool demo of the tech 👍
Ya know, I've been doing testing with long-ish (6mm) travel switches and low-profile, short (2mm) travel switches lately and it's amazing how much of a difference that makes in the sound. Using different materials also has a huge impact. For example, if you print the switch in PLA that makes a higher pitch sound (because it's a harder material than PETG). It also sounds different if you print say, the stem/slider in PETG but the switch body/sheath in PETG. The keycaps also make a HUGE difference in the sound of the keyboard. Thick PETG keycaps sound and feel the best IMHO. They have a softness to them that just feels nice and they generate some of the lowest, "thockiest" (hehe) sounds. If you print everything in PLA you get that high-pitched "clack" sound that a lot of people prefer though. Of course, lube (and different types) also modify the sound profile quite a bit.
That's brilliant! If the magnets were slimmed down (and some parts being made from, say, an CNCed aluminum bracket) the design could be slimmed down enough to hold some gorgeous flat keys for immensely satisfying highspeed typing. I don't think we'll ever see anything close to the absurdly thin mechanical thing from apple but given the high failure rate of those keys perhaps it's just not meant to happen, but this looks like a great potential leap forward :)
Yeah but it sucks, haha. It's a gimmick. Linear motion with only ~4mm of travel doesn't give you much in the way of fine-grained control. Human fingers need a bit more travel than that for it to be a decent analog control. Normal game controller thumbsticks give you about 10mm of motion (in a circle). However, that's with a thumb-based control that uses muscles for lateral movements... Which human thumbs are good at (for fine muscle control motion anyway). Your middle finger pressing down on W is going to drive you nuts if you try to use it to get the same level of control. Having said that, there *are* cool things you can do with the analog detection other than just picking your point of actuation/release. For example, you can detect *velocity*! So if SOMEONE IS WRONG ON THE INTERNET you can detect that in the firmware and send those keystrokes along with a shift key modifier 😁👍
Very interesting! But since it's 3D printed, how comparatively smooth is it compared to injection molded plastic? Are there plans to have it be mass produced by a switch manufacturer using injection molding?
Getting moulds made costs a fortune that I don't have. It would also mean creating moulds for each strength of switch (0.0mm through 1.9mm or so) which is a lot of combinations. As far as smoothness goes, they're super smooth right off the printer because I designed them so that the orientation of the parts has the layer lines running parallel to each other in the axis of motion (|| against ||). PETG has a very low coefficient of friction--especially rubbing against itself (it sheds hydrogen atoms and has self-lubricating properties). POM and Nylon would be smoother but not smooth enough to make a huge difference in feel IMHO. Besides, you can always just spend some time sanding your 3D printed stem to make it as smooth as glass if you so desire.
@@Riskable3DPrinting You really thought of everything! I was concerned that 3D printing leaves layer lines, and it didn't occur to me that you could print them that way where all the lines run parallel to each other up and down. Can't wait to see the finished product!
I've been thinking about completely magnetic switches like this recently, the only thing that could make these better is if there was somehow a magnet so strong it would repel the slider in all directions in such a way that there would be no friction at all but the switch would still be just as stable. Although I'm not an engineer by any means so I'm probably missing something important with that idea.
I like the project, especially how you can control the tactility because it is a 3D printed switch. Though how do you attach the switch to the keyboard? Is it the same way like with a solder and hotswap or is that not a thing yet?
I thought of this sort of idea a while ago after seeing the At&t switches. Really glad to see that it's possible and so elegantly done. Although I was wondering if you could even remove all friction by putting a magnetic slider in the center of a magnetic ring or something, so the slider would never need to rub against anything at all inside the housing. The solution used here is a lot simpler though, and tons easier to make, I'd think.
Why was the magnet on the keycap not integrated inside the housing? Im assuming having it under the housing would have similar repelling effect to push it back up.
It just saves vertical space. There's no reason it couldn't have been shoved inside the body of the switch. It would've just added an additional 4mm (or more if you generated it with extra travel) to the overall height. By shoving it on the stem and hiding it under the keycap it makes everything more compact 👍 Also note that the switches I sent to Chyrosran22 had 4x2mm magnets in the levitator. There's no reason why you can't use 4x1mm magnets and a slightly thinner levitator if you want to save even more space.
Cool, want to try to build these, thanks for sharing. I would like to see how you are connecting them to the board/circuit, ie. how you interface with the contacts. Do the magnets sit on a pre-wired contacts, and are they used as the key trigger contacts themselves? KISS? Also, is there an idea to deal with things like debouncing, or maybe that should be dealt with by default in the firmware for this design?
This is something I'm working on, actually. I designed a collection of PCBs for making your own hand-wired analog keyboards in any configuration you desire. I just built the world's first hand-wired analog keyboard (a 6x2 macro pad) the other day as a proof-of-concept and I'm currently putting together a guide and figuring out how to get the PCBs into the hands of end users 👍
That looks so very cool. Why aren't leaf springs more of a thing in modern switches? Or finger springs, wave springs? It seems like a lot of effort is spent on compensating for the shortcomings of compression springs.
Metal leaf springs aren't easy to prototype with. You need to be able to make them with very precise tolerances in order for them to work well in any given switch design. They're also *tiny* and tend to go SPROING!!! off into the distance of your shop while you're trying to put switches together. Ask me how I know this! :D
I've been thinking about something like this for a while. I'd love to see if the tactile bump could be moved closer to the middle of the travel. I've been trying to model something with few moving parts, but can't get it to return without a spring.
7.9ish billion people in the world. It's pretty likely other people have the same idea as you. I also thought of custom-made or 3d printed switches some time ago, but I don't have the stuff needed to make them. Really amazing to see them actually made now.
This is actually possible! With a very expensive PCB though. You can put several layers (like an 8-layer board) with PCB tracks in a rings (one ring per layer) and place them under each switch. Then when you detect a keypress you can fire up an H-bridge to give each switch a reasonably strong "push" upwards to provide force feedback. It won't be as strong as an old-school linear actuator (e.g. old IBM keyboards) but you'd definitely be able to feel it. Might even be able to get rid of the levitator magnet since you can get that "bottom out repulsion" effect with the PCB coils. I keep meaning to try it but... Funds for experiments are low at the moment (and I really need to spend more time on firmware!).
If they ever make a group buy or Kickstarter to get a full line of switches and keyboards kits... I'm interested. The custom mech world is missing out with how much it's focused on MX style. I want to try totally different concepts like this.
At first the only thing I'll be able to offer is some PCBs with the instructions: "Print your own damned switches!" Would you still be interested? :D I'd make the files to generate the switches publicly available of course 👍
@@Riskable3DPrinting Lol! If I could afford a 3D printer or knew someone who had one, absolutely, but I don’t have access to that kind of thing. It’s awesome that you’re looking to make the files publicly available. I’ll keep an eye on your channel for any updates you might post about these switches. I hope they catch on!
Any keyboard that needs more than universal HID software to work is a no go in my book, as I don't need a keyboard to say be locked down to Windows to fully work, simply because the maker can't be assed to port the software to Linux, BSD, Mac, or any other OS that may come in the future, thus rendering it a brick.
@@CommodoreFan64 You can be rest assured that any keyboard I make is going to have full Linux support. Why? I've been using Linux as my full-time desktop since 1999! There's no Windows in this house (except for testing VMs hehe).
@@Riskable3DPrinting Awesome, the only Windows I currently run is an old P4 retro build with 98SE/XP(most of the hardware was free) that never goes online so I can play my collection of old physical PC games. otherwise it's a Windows free house for me as well.
Joined the custom keyboard community recently and was thinking about potential alternative designs for switches that allow for minimal contact to reduce unwanted friction noises. Right after thinking about using magnets, this video popped up. Does anyone know if the maker of this switch actually plans on turning it into a product?
Neodymium magnets "hitting stuff" doesn't demagnetize them (hehe). What *does* demagnetize them is heat so keep them away from the oven/stove. I've run a few switches through the dishwasher and the magnets were unaffected (the switch smoothness was though! I've since learned that it's because dishwashing detergent has mild abraisives and that's why the switches became a little scratchy for a while after... They became smooth again though after much fidgeting thanks to PETG's self-lubricating properties).
Yoo i wanna build these but the documentation doesnt have much in the way of how theyre connected to the board. the void switches look like they dont have pins and these obviously have the hall sensor in each one but the hall sensors are out of stock. To my underatanding the void switches are easier to build?
That's because Void Switches aren't *meant* to "connect to the board". They're meant to just sort of "hover above the PCB" (where the hall effect sensors are placed). I started a tutorial talking about how to make a PCB like that: ruclips.net/video/TfKz_FbZWLQ/видео.html
Theoretically since theres no mechanical wear on the sensor detecting the "key press" it should last many many times longer than any other key design. No mechanical electrical contacts to wear out. so really no wear on the switch.
Question: What the lifespan and durability of the magnets Plus how would usage effect it? (Unless the magnets are electrically powered probably it would eventually become weaker and weaker with usage) (I am also no scientist but I do fidget with magnets every so often)
The lifespan of the magnets is thus: They lose about 5% of their strength every 100 years. So after 300-500 years it may be time to consider replacing them (unless you don't mind the slightly weaker feel hehe).
You could! This would definitely work! Only problem: The electromagnets would have to be like 20mm long and placed under each key. So you'd have a very tall keyboard (that spits out EMI like crazy, haha).
@@lucidnonsense942 I was pointing out that if the term sticks, it could be confusing having the same names in two pc parts that could even someday come from the same brand. Just that.
Hence why I decided to call them "Void Switches": Because the strength of the switch is controlled via the "void" between the magnets (aka the distance between them which is filled with plastic).
Finally something truly new in key switches.
I mean, hall effect switches are more than half a century old at this point, but I do agree that the rest of the concept is fairly innovative. I haven't been this excited about keyboard tech in a while.
@@DamnZodiak The point is that It took a while so anyone made a New design inspired by something Else than Cherry MX or Buckling Springs
@@DamnZodiak My dad said they had a few keyboards with magnet switches, with the old instalation and servers. When I got interested in keyboards, he looked at his work in storages to see if there were any left, or if someone kept a few, but unfortunatelly they were srapped.
Really wanted to try them.
This is absolutely bonkers, and pretty rad.
Bonk
-tf2 scout
Can I get this comment in imperial units? My smooth American brain can’t realllt understand it.
@@shakie6074 "this right here is by God the most craziest fuckin shit I tell you what, Boot scootin boogie"
@@xFreSh999_ Thank you friend much appreciated 😂❤️
apt description
I'm really interested in how these switches turn out in a keyboard... couple of challenges could include the sensing assembly and stabilizers, tho
@@Riskable3DPrinting Are you planning to open source your design?
@@Riskable3DPrinting So in absence of springs for weighting, you can adjust the magnets in order to have desired weights/effects, yeah? I'm still trying to wrap my head around this, but it seems like you might have genuinely the greatest switch innovation of the last forty years. I hope a major manufacturer can work with you to get this standardized or something. I need this in my life.
@@harjoat Yeah, eventually 👍
@@Riskable3DPrinting glad to see your switches are gaining popularity
@@Riskable3DPrinting Got any typing sound samples?
Dude needs to kickstart this. This is amazing.
Amen. I'd buy it. I want this tactile contactless analog gold!
These are amazing, I can just see them being _really_ expensive.
@@myrealusername2193 you can just 3d print them and materials are really cheap
@@malaythakkar8772 I was talking more about the magnets
@@malaythakkar8772 just need to buy a 3d printer, resin, and magnets. then learn the cad software... right?
yeah, na. i think some people would rather just purchase them
When you make a new switch and you send it to one of the most brutal guys on RUclips, that guy haves some balls.
Thanks for the video, there's some really interesting and skilled people out there and it will be cool if some get some recognition.
Honestly, that's a big reason why I sent them to Chryos. I knew he'd give honest feedback 👍
@Riskable 3D Printing Good for you for making a product that makes you confident enough to subject it to Thomas! From the looks of it it was a smash success. I hope you can commercialize some aspect of this.
@@Riskable3DPrinting beware, a lot of people now will try to steal the design and mass produce it (I hope you already patented it).
An age has ended and a new era has begun.
Phineas: An age has ended and a new era has begun.
Gamers: $200 for MX clones? Take my money!!!
@@gyozakeynsianism who pays $200 for only mx switches? The most expensive switches for the largest mainstream layout would end at about $120. The most popular switches cost significantly less. And gamers usually buy like $25 switches (gateron/cherry), not $200 switches.
@@user-kz9rz8ui1c Sure seems like $200 MX keyboards are still on the market (usually with lots of lights). And the custom market (which is more for typist snobs than gamers I think) is MX dominated for some reason, and those keyboards are even more expensive.
@@gyozakeynsianism oh you're talking about keyboards. Any mx style prebuilt that is over $100 is overpriced imo. The only prebuilt worth getting that costs over $150 is the apex pro, which isn't really mx style. Customs actually usually range much more expensive, but youre mostly paying for great build quality. You can pick up decent customs for less then $150, any more than that and you are paying for design, material, and weight.
@@user-kz9rz8ui1c Yeah I wasn't clear. I meant keyboards with MX or MX clones.
The sheer level of and ease of modification to the user experience puts these leagues ahead of everything else on the market as far as fine tuning for the end user. I absolutely can not wait to see where this project goes!
He could even potentially design in adjustable magnet distances so you can adjust the tactility with something like a screw mechanism. That’s really the only improvement I can think of!
@@myrealusername2193 So, what does he mean by «tactility» ?
@@BlueTemplar15 basically, when you push down a key, at a certain point it becomes much easier to push. By adjusting where the magnet is, you could adjust at what point this happens, and how much of a difference in force there will be. The important part is that this could be adjusted by the user very easily, where normal moulded plastic key switches can’t
When production starts I will buy a set! The dampener was next level quite and function.
That’s gotta be one of the coolest switches I’ve seen before. Awesome to see new switch designs
Whoa. This really feels like a game changing concept I've ever seen in key switches, from top to bottom just a really intriguing concept.
this was something i thought about when i got into mech keyboards but didn't have the resources
it's nice to see it's becoming a reality
Same, thank goodness Riskable has the drive to see ideas like this into reality!
yeah ive seen maglev mounts and this seemed like the next logical step
Always promising when you can SEE the tactility in a video.
Definitely on my radar, gonna be interesting to see where this is going.
I made some magnetic springless keyboard switches over 30 years ago, that also used reed switches for my ZX Spectrum.
These are fascinating and clever as hell. I'd never heard of these before now, but they seem like a great idea. Just the adjustments in tactiliy alone blow me away...add to that the options in height of travel and this is some crazy brilliant stuff.
They do look great - but could the magnets attract small iron particles over time gum up?
Maybe not machine shop safe? But unless you live on a beach I can't picture this being much of an issue.
@@subterficial Even in a machine shop they'd be fine. The metal bits would just collect around the edges of the plastic and maybe a little bit on the parts of the stem that are exposed. They wouldn't impact the functionality at all. Not unless you start getting millimeters worth of metal piles under the plate :D
There are already many switches that use Hall effect technology which means they use magnets. They don't all gum up with metal so I don't think it would be a problem. In fact, hall effect switches are considered very reliable.
Considering how tiny the magnets are, it wouldn't take you very long to find a big enough magnet to clean them off if that ever did become a problem.
This is the type of innovation the keyboard technology needs to get out of the this swamp of Cherry clones.
So true. the "keyboard enthusiasts" seem to be fascinated by all the cherry mx clones with ridiculous names but its all pretty much the same isnt it
@@chubeviewer Well, it isn’t really “all the same”, otherwise why would people be interested in them. But I’m still very interested in this design. Much like the guy who built his own case out of stainless steel and frosted plastic (weighing the rather small keyboard in at over 2kg), I’m immensely proud of Riskable for trying to make this idea work. As a concept, it’s very cool, but it will really only stack up if it works in an actual keyboard.
@@offbrandbiscuit His channel ruclips.net/channel/UCSsm4h2ccJMdR6bsAFyIu6Avideos. He has an actual 3d printed keyboard.
The idolisation of cherry does tick me off, but one thing that I do appreciate and wish was an actual industry wide thing is how all the cherry clones have the same socket for keycaps
@@davidy22 I dont think cherry is that much idolized, its just the whole scene with hot swapabilitie and all kind of was sure to end in one switch design becoming the norm. Its not just they keycap mount, also the pins for the pcb. All these new switch designs really only make sense as full non customizable boards.
riskable is a madman, so glad you got to take a look at these
Finally something new, seems actually viable for mass production.
I was fallowing this project a while ago.
I'm so happy you made a video about it!
There's something just really excellent about the concept here. I hope a keyboard based on this switch is comes to market because the switch looks just absolutely brilliant. Bravo!
OMG and the dampened one!!
ive never been this excited about getting my hands on a switch since i first discovered salmon alps. This seems extremely promising, and I really hope this gets into production. Easily the coolest switch concept ive seen in a bit
Amazing and fresh idea. Definitely want a whole board of these.
Riskable dose awesome work with these and is constantly improving. A really awesome project :) And a great teardown on your side :D
This gives me so many ideas! Ive been wishing someone made an MX-compatible magnetic seperation switch for a while now, glad to see someone else is thinking down that track as well.
Well only the switch stem is MX-compatible. The switch itself needs a hall effect sensor or reed switch to actually function as a keyboard key (use hall effect though it's better and cheaper 👍). I only added support for MX stems to the switch but I have a placeholder in the code for adding Alps stem support too.
@@Riskable3DPrinting true, didn't think of that. Hall effect is better anyway!
Ok I'm sold, I need to do something with these. Definitely can't wait to see what this turns into once proper manufacturing happens
Your voice sounds really nice, and so do the switches
Gateron give this man some money
Agreed, if he has not filed a patent, or one has not already been filed for this design, then this guy needs to do so ASAP, and either partner with someone like Gateron, or start his own company.
@@CommodoreFan64 it is prob to close based on old patents to be protectable but as soon as something is public no one else can claim patent rights to the tech(in theory) meaning if someone could the switch would be safe
@@CommodoreFan64 He's open-sourcing it.
@@WhenDoesTheVideoActuallyStart then that's a good thing.
Dude designed a whole new switch while I tried to take a sip of my water bottle, totally forgetting the cap is still on
I need a much more in-depth review of these. And I need to get myself some. Wow these look so good.
I had a similar idea but with coils or one coil and using an electric field to control the weight and tactility. You could create a dampened switch and control the sound by using a few tiny speakers. I don't really have the time to pursue, so I thought I might as well put it out here before some arsehat patents it. Oh, almost forgot, you could also use the coil to power the switch itself I suppose, for displaying options or the LED.
This is probably the coolest modern switch i've ever seen.
Those are game changing switches, can't wait to buy a set.
Brilliant design, hopefully someday we can see a full board with them
The maker has a channel, and has built a full board. His channel name is 'riskable 3d printing '
I already made *three* keyboards that use this design: A numpad, my Riskeyboard 70, and the world's first hand-wired analog keyboard. You can see all three in this gif I made yesterday (though the numpad doesn't have any RGB hehe): gfycat.com/fluidsevereairedale
Thanks for the reply, definitely checking that out!
I have several questions:
1. I've been wondering how smooth you could get a 3d printed switch, are these any good in that regard?
2. Can we get some force curves for these?
The switches are quite smooth because the print orientation results in layer lines that are like this: || || (they run parallel to each other in the axis of motion). For *extra* smoothness/less friction you can print the stem at a slightly different layer height (e.g. print the stem at 0.12mm and the sheath at 0.16mm). I never bother though because they're pretty damned smooth right off the printer.
Regardless of how terrible your printer is you can always *make* the switches ultra smooth by just sanding them! Also, if you sand the bottom (flat) corner edges of the stem with something like 400 grit sandpaper that makes it extra smooth as well because it gets rid of the little crinkle that's possible from first-layer-squish (aka elephant's foot). It would wear off after a few hundred keypresses anyway though.
Here's a force curve I measured of a 0.0mm MAGNET_GAP Void Switch with 4x1.7mm N35 magnets by hand (lowering my 3D printer in 0.1mm increments on to a switch that was sitting on a digital scale): i.imgur.com/FNoTSpL.png (I think it was a switch with 3.5mm of travel)
Love it. Need those super long travel switches.
These are very cool.
Do you know if this is the first instance of a contactless return force as well?
How neat. With the analog HAL sensor you can measure the speed or force applied to the switch.
Finally a possibility to measure anger level directly! Maybe even auto-shutdown to force you to go out if you're too heated?
Forget all that: Finally we can have proper slam-head-into-keyboard detection! Unless the velocity is over a certain threshold the macro won't fire 👍
@@Riskable3DPrinting Hahah :D
And if it detects a forehead sized pattern with too much velocity it calls an ambulance
@@juweinert Nooo! It deletes your browser history!
Those are just... better magnetic seperation switches. Insane work.
Also incredible video as always, Thomas!
So cool! I hope he gets in touch with an enthusiastic company that will help him commercialize this as a product. I'd love to install these switches on a nice keeb and give them a go.
"Maniacal" indeed, wtf
his voice is just very soothing
Innovative switches are back baby
This is a very cool design, I'd love to help out if needed. Either way, I'm excited to see what comes of it
I remember chatting to riskable about these. Such a cool design.
Jesus, this is a switch family which can be either linear or have varying tactility while not only lacking friction on contacts but also the friction of the tactile element/clicker, AND IS ALSO ANALOG CAPABLE. It is the first of its kind.
hey this is kinda how the old school Tube Mod Vaporizers work. With two magnets of opposite polarity the button is open pushing on the button wont allow the magnets to touch so it forces the battery up into the contact. pretty nifty
Just use an electromagnet on the switch housing and PWM it to the amount of tactility!
I actually thought about doing this! I got close recently when I hooked up a relay board to my keyboard: ruclips.net/video/6hMOGKTudcg/видео.html
@@Riskable3DPrinting that is funny! Do you think it would interfere with the Hall effect sensing?
@@KayvonJavid Yes it would! Absolutely! Haha. It's not as big a deal as you think though: My firmware can read 96 analog hall effect sensors in under 200 microseconds which leaves 800 or so to send a strong electromagnetic tactile jolt to stay within that 1ms polling window 👍
Maybe moving to optical switching could remove any penalty from this?
@@NickShabazz probably it also has analog capability too. But man the life time of this switch design could be well beyond billions because there really isn’t anything to wear out other than the electronics
I am 100% going to pick one of these up when or if they become available: Even if it turns out I have to buy a set of keycaps with them too. Brilliantly designed!
I see little reason for this design to not go with MX style mounting. I feel like the mounting system is one of the key reasons MX style has become so standard. And its not like its a bad system or anything.
@@prokopf-9332 I think the magnet in keycap might be a problem though
his voice is so deep that it is deeper than my grades
This is extremely interesting and exciting.
This is real innovation, and I can't wait for more work to be done on them and for them to be implemented fully.
We've just witnessed key switch history.
Nice project!
I would recommend SLA (resin) printing for such small parts...
That's _really_ interesting. that one design makes a switch on any point of the spectrum of slightly to very tactile, or from low to mega-mega-high profile, is awesome
Also, have you seen his clicky keyboard? It uses _relays._
My Riskeyboard 70 is actually a microcontroller board in its own right... So hooking up relays was a cinch since all those extra pins are exposed at the back of the keyboard for hooking up things like that. I also added a QWIIC connector but have yet to try it out with anything yet. Was thinking about hooking up a color sensor board so you could hold up a color swatch (like you'd get at the hardware/paint store) and it'd automatically type out the HTML hex code for that color. Would be a cool demo of the tech 👍
really interesting and unique cant wait to see where this goes
It would be insane to have a keyboard with a single screw that adjusts the tactility of the entire board. That would truly be something else.
Electromagnet in every switch 😂
That's pretty interesting.
Would love seeing them in a keyboard for sound, comfort and durability testing.
Ya know, I've been doing testing with long-ish (6mm) travel switches and low-profile, short (2mm) travel switches lately and it's amazing how much of a difference that makes in the sound. Using different materials also has a huge impact. For example, if you print the switch in PLA that makes a higher pitch sound (because it's a harder material than PETG). It also sounds different if you print say, the stem/slider in PETG but the switch body/sheath in PETG.
The keycaps also make a HUGE difference in the sound of the keyboard. Thick PETG keycaps sound and feel the best IMHO. They have a softness to them that just feels nice and they generate some of the lowest, "thockiest" (hehe) sounds. If you print everything in PLA you get that high-pitched "clack" sound that a lot of people prefer though.
Of course, lube (and different types) also modify the sound profile quite a bit.
As a thicc click lover I now just have to try these at some point, even If I have to buy a 3d printer just for it.
That's brilliant! If the magnets were slimmed down (and some parts being made from, say, an CNCed aluminum bracket) the design could be slimmed down enough to hold some gorgeous flat keys for immensely satisfying highspeed typing. I don't think we'll ever see anything close to the absurdly thin mechanical thing from apple but given the high failure rate of those keys perhaps it's just not meant to happen, but this looks like a great potential leap forward :)
With analog input using the hall effect sensor you could emulate analogue game pad inputs and have throttle, as an example, on your W key. Very cool.
Yeah but it sucks, haha. It's a gimmick. Linear motion with only ~4mm of travel doesn't give you much in the way of fine-grained control. Human fingers need a bit more travel than that for it to be a decent analog control. Normal game controller thumbsticks give you about 10mm of motion (in a circle). However, that's with a thumb-based control that uses muscles for lateral movements... Which human thumbs are good at (for fine muscle control motion anyway).
Your middle finger pressing down on W is going to drive you nuts if you try to use it to get the same level of control. Having said that, there *are* cool things you can do with the analog detection other than just picking your point of actuation/release. For example, you can detect *velocity*! So if SOMEONE IS WRONG ON THE INTERNET you can detect that in the firmware and send those keystrokes along with a shift key modifier 😁👍
Very interesting! I've always wanted to try the mag sep and now mag lev switchs. Interesting times we live in. Thanks Tom, good to see you as always.
This looks fantastic
I hope this project goes mainstream.
This is just straight-up wacky town banana pants. Really cool stuff.
Very interesting! But since it's 3D printed, how comparatively smooth is it compared to injection molded plastic? Are there plans to have it be mass produced by a switch manufacturer using injection molding?
Getting moulds made costs a fortune that I don't have. It would also mean creating moulds for each strength of switch (0.0mm through 1.9mm or so) which is a lot of combinations. As far as smoothness goes, they're super smooth right off the printer because I designed them so that the orientation of the parts has the layer lines running parallel to each other in the axis of motion (|| against ||). PETG has a very low coefficient of friction--especially rubbing against itself (it sheds hydrogen atoms and has self-lubricating properties). POM and Nylon would be smoother but not smooth enough to make a huge difference in feel IMHO. Besides, you can always just spend some time sanding your 3D printed stem to make it as smooth as glass if you so desire.
@@Riskable3DPrinting You really thought of everything! I was concerned that 3D printing leaves layer lines, and it didn't occur to me that you could print them that way where all the lines run parallel to each other up and down. Can't wait to see the finished product!
I've been thinking about completely magnetic switches like this recently, the only thing that could make these better is if there was somehow a magnet so strong it would repel the slider in all directions in such a way that there would be no friction at all but the switch would still be just as stable. Although I'm not an engineer by any means so I'm probably missing something important with that idea.
That would be very very difficult and it would cost a ton 😂
@@imender9219 yeah lol, although that's assuming that it's even possible to make work.
OK that's bad ass, damn I need a 3-D printer.
this is absolutely ridiculous and i love it
I like the project, especially how you can control the tactility because it is a 3D printed switch. Though how do you attach the switch to the keyboard? Is it the same way like with a solder and hotswap or is that not a thing yet?
I thought of this sort of idea a while ago after seeing the At&t switches. Really glad to see that it's possible and so elegantly done. Although I was wondering if you could even remove all friction by putting a magnetic slider in the center of a magnetic ring or something, so the slider would never need to rub against anything at all inside the housing. The solution used here is a lot simpler though, and tons easier to make, I'd think.
Imagine using adjustable magnets to change the weight and tactility in every switch to make it perfect for the user
Why was the magnet on the keycap not integrated inside the housing? Im assuming having it under the housing would have similar repelling effect to push it back up.
It just saves vertical space. There's no reason it couldn't have been shoved inside the body of the switch. It would've just added an additional 4mm (or more if you generated it with extra travel) to the overall height. By shoving it on the stem and hiding it under the keycap it makes everything more compact 👍
Also note that the switches I sent to Chyrosran22 had 4x2mm magnets in the levitator. There's no reason why you can't use 4x1mm magnets and a slightly thinner levitator if you want to save even more space.
Cool, want to try to build these, thanks for sharing. I would like to see how you are connecting them to the board/circuit, ie. how you interface with the contacts. Do the magnets sit on a pre-wired contacts, and are they used as the key trigger contacts themselves? KISS?
Also, is there an idea to deal with things like debouncing, or maybe that should be dealt with by default in the firmware for this design?
love the new intro!
I need that waning keycap lol
These are pretty interesting switches.
Ahaha I was wondering if someone would notice it! I actually made a whole bunch of them and have some here on my desk. Maybe I should sell them? 😁
would love a tutorial tbh. Also, on how to do your own keywoard
This is something I'm working on, actually. I designed a collection of PCBs for making your own hand-wired analog keyboards in any configuration you desire. I just built the world's first hand-wired analog keyboard (a 6x2 macro pad) the other day as a proof-of-concept and I'm currently putting together a guide and figuring out how to get the PCBs into the hands of end users 👍
@@Riskable3DPrinting Amazing work, been following your channel for some time now, can't wait to get a 3D printer and try out some ideas myself.
That looks so very cool.
Why aren't leaf springs more of a thing in modern switches? Or finger springs, wave springs? It seems like a lot of effort is spent on compensating for the shortcomings of compression springs.
Metal leaf springs aren't easy to prototype with. You need to be able to make them with very precise tolerances in order for them to work well in any given switch design. They're also *tiny* and tend to go SPROING!!! off into the distance of your shop while you're trying to put switches together. Ask me how I know this! :D
I've been thinking about something like this for a while.
I'd love to see if the tactile bump could be moved closer to the middle of the travel.
I've been trying to model something with few moving parts, but can't get it to return without a spring.
is there a keyboard with those ?
So far it looks like it's just key switches that are still in development stages.
is this the voice from ghost in the shell the puppet master voice?
Is it possible to use these without a custom PCB with hall effect sensors and whatnot?
This is genius. Link to this person’s work, if it’s public/available?
I have my own channel where I post updates from time to time: ruclips.net/channel/UCSsm4h2ccJMdR6bsAFyIu6A
@@Riskable3DPrinting Some keyboard company should hire you. This shit is so cool
This is sick.
i was at work today and i thought of this idea. i went to youtube to see if anybody had already thought of this idea and yep here it is
SOoo, any chance of using a small coil to add adjustable force?
I wanna print them
same
This is unbelievably insane.
I thought 3d printed magnetic switches were my own idea well, apparently not
7.9ish billion people in the world. It's pretty likely other people have the same idea as you. I also thought of custom-made or 3d printed switches some time ago, but I don't have the stuff needed to make them. Really amazing to see them actually made now.
id like to see one of these with adressable electromagnets so that you can change the tactility through software
This is actually possible! With a very expensive PCB though. You can put several layers (like an 8-layer board) with PCB tracks in a rings (one ring per layer) and place them under each switch. Then when you detect a keypress you can fire up an H-bridge to give each switch a reasonably strong "push" upwards to provide force feedback. It won't be as strong as an old-school linear actuator (e.g. old IBM keyboards) but you'd definitely be able to feel it. Might even be able to get rid of the levitator magnet since you can get that "bottom out repulsion" effect with the PCB coils.
I keep meaning to try it but... Funds for experiments are low at the moment (and I really need to spend more time on firmware!).
Extremely cool idea! When does the group buy start?
How much would you spend? :D
I think as much as I did on the New Model F!
If they ever make a group buy or Kickstarter to get a full line of switches and keyboards kits... I'm interested. The custom mech world is missing out with how much it's focused on MX style. I want to try totally different concepts like this.
At first the only thing I'll be able to offer is some PCBs with the instructions: "Print your own damned switches!" Would you still be interested? :D I'd make the files to generate the switches publicly available of course 👍
@@Riskable3DPrinting Lol! If I could afford a 3D printer or knew someone who had one, absolutely, but I don’t have access to that kind of thing. It’s awesome that you’re looking to make the files publicly available. I’ll keep an eye on your channel for any updates you might post about these switches. I hope they catch on!
damn, this switch looks near perfect. The only thing I wish it had was a way to change the tactility on the fly or with software.
Any keyboard that needs more than universal HID software to work is a no go in my book, as I don't need a keyboard to say be locked down to Windows to fully work, simply because the maker can't be assed to port the software to Linux, BSD, Mac, or any other OS that may come in the future, thus rendering it a brick.
@@CommodoreFan64 You can be rest assured that any keyboard I make is going to have full Linux support. Why? I've been using Linux as my full-time desktop since 1999! There's no Windows in this house (except for testing VMs hehe).
@@Riskable3DPrinting Awesome, the only Windows I currently run is an old P4 retro build with 98SE/XP(most of the hardware was free) that never goes online so I can play my collection of old physical PC games. otherwise it's a Windows free house for me as well.
Changing tactility on the fly could have some serious video game applications. I think fighting game developers would love something like that.
Joined the custom keyboard community recently and was thinking about potential alternative designs for switches that allow for minimal contact to reduce unwanted friction noises. Right after thinking about using magnets, this video popped up. Does anyone know if the maker of this switch actually plans on turning it into a product?
I'll probably be selling PCBs very soon and maybe full kits (though not likely fully assembled) eventually 👍
Upgrades old laptop keyboard. Wipes hard drive. Cool design!
Now this is a genius design!
But is there no risk of accerated demagnitizaton over long term use from the magnets hitting stuff?
Neodymium magnets "hitting stuff" doesn't demagnetize them (hehe). What *does* demagnetize them is heat so keep them away from the oven/stove. I've run a few switches through the dishwasher and the magnets were unaffected (the switch smoothness was though! I've since learned that it's because dishwashing detergent has mild abraisives and that's why the switches became a little scratchy for a while after... They became smooth again though after much fidgeting thanks to PETG's self-lubricating properties).
Yoo i wanna build these but the documentation doesnt have much in the way of how theyre connected to the board. the void switches look like they dont have pins and these obviously have the hall sensor in each one but the hall sensors are out of stock. To my underatanding the void switches are easier to build?
That's because Void Switches aren't *meant* to "connect to the board". They're meant to just sort of "hover above the PCB" (where the hall effect sensors are placed). I started a tutorial talking about how to make a PCB like that: ruclips.net/video/TfKz_FbZWLQ/видео.html
Theoretically since theres no mechanical wear on the sensor detecting the "key press" it should last many many times longer than any other key design.
No mechanical electrical contacts to wear out. so really no wear on the switch.
This is amazing
Question: What the lifespan and durability of the magnets
Plus how would usage effect it?
(Unless the magnets are electrically powered probably it would eventually become weaker and weaker with usage)
(I am also no scientist but I do fidget with magnets every so often)
The lifespan of the magnets is thus: They lose about 5% of their strength every 100 years. So after 300-500 years it may be time to consider replacing them (unless you don't mind the slightly weaker feel hehe).
I wonder if you used electro magnets if you could make variable resistance switches :0
You could! This would definitely work! Only problem: The electromagnets would have to be like 20mm long and placed under each key. So you'd have a very tall keyboard (that spits out EMI like crazy, haha).
Curious hoe they would be inside an actual keeb and how they compare vs the mx style switch 🤔
The term maglev can be a bit confusing, as it is also used in some corsair fans
And trains and anything that uses magnetic repulsion, Corsair didn't invent the term, it's a generic one for what the switches do.
@@lucidnonsense942 I was pointing out that if the term sticks, it could be confusing having the same names in two pc parts that could even someday come from the same brand. Just that.
Hence why I decided to call them "Void Switches": Because the strength of the switch is controlled via the "void" between the magnets (aka the distance between them which is filled with plastic).
This is cool af