I actually hadn’t heard of that; something like this?
I kind of want to try it for clicks and giggles, but yeah there’s more than one reason most controllers don’t use microswitches for everything. Didn’t realize you could just solder a two-pin microswitch onto a dome pad and have it work, though it makes sense in retrospect.
I switched my Goto gamepad to an EasySMX X10 for hall effect and actually Microswitches, I really like that way more. Did not know about that dome mod either, very good to know.
I wanted to play around with the Raed HE switches, so got a Geon Venom60 HE board and matching plate. I expected a quick build, but it wound up being a bit more frustrating and not entirely happy with the end result.
Geon Venom60 HE PCB
Geon Venom60 plate
Aeboards x Tecsee Raed HE switches
Invokeys Reserve stabilizers
Keychron Hacker Mint keycaps
ye olde generic low profile GH60 case
The crux of the problem is that the Raed (or Naevy, for that matter) switches aren’t physically anchored to the PCB itself. Additionally, the bottom being a silicone dome makes it much more flexible than the typical harder plastic used for switch housings.
I discovered this partway through the build when I found that split backspace was NOT gonna work – if you put almost any non-perpendicular force on them, they popped out of the board. This seemed due to the split backspace having more open space around them on the plate (to accommodate stabilizers). Ok, fine, I went back to a unified backspace. All should be golden, right?
Nope. When I finally got everything together, the left shift kept sticking. It wasn’t doing it until I got the plate attached, which seemed strange. I originally figured it was the plastic case I was using not being perfectly straight, so tried an aluminum low profile case. But that also seemed to have the same issue.
It turns out it’s an alignment issue: if I moved the switch slightly “south” (towards the Ctrl key), it would stick instantly. Pushing it slightly “north” (towards Caps Lock) unbound it until it got too far and then it started sticking again.
I eventually got the alignment working, but it was finicky as hell and took several tries. But it finally came together. So why am I still unhappy with it?
Well, remember that problem with the keys popping out? Yeah, didn’t exactly go away.
Worse yet, it impacts keys with stabilizers.
Backspace working:
…vs. backspace out of alignment:
Spacebar impacted, as well (working vs. broken):
It doesn’t take much force at all to do this, either. I accidentally popped off a switch just handling it to show my roommate, while I clonked the spacebar carrying it around. It’s easily fixable, but obnoxious how easily the switches pop out.
Higher profile case would help, but not sure if there’s much to be done here – this seems like a design issue with the switches themselves.
I have the Cipulot EC60X PCB and Naevies for a build and was waiting on the CK plate, but now I’m wondering if it’s going to have the same problem. The springs might help the wobble problem some.
Really appreciate you reporting your findings on trying out the switch with Geon’s Venom parts, I am guessing that maybe part of the issue is there isn’t enough standoff positions/screws between the plate and PCB to hold the assembly perfectly still. Idk obviously how much that would help with the switch slipping out but maybe it could help. I only say that because I believe I remember hearing & seeing that the potential Cannonkey’s Plate is gonna have alot of standoff screw positions.
There are 8 screws on the plate, and they mostly seem to do the trick of holding the plate still. Now, part of it might be that they are not quite in alignment with the board, since there is a little bit of play in how the standoffs are soldered in (not exactly centered in the holes). That’s probably what caused most of my issues with the left shift sticking.
But, I think the larger problem is that the switches housings themselves are also flexible, due to the silicone dome bottom.
Any non-normal force applied to the keycap + switch combo is going to create torque on the switch housing, with the fulcrum being where the housing meets the plate. Normally, this doesn’t actually do anything, b/c the switch housing is hard plastic in contact with the PCB – trying to rotate the switch body that way would push it into the PCB, so it doesn’t budge.
However, the silicone bottom of the Raeds is made to flex, and so enough force eventually collapses it, levering the switch out of the plate.
It’s easiest to demonstrate by pushing sideways on the switch. On a normal board, there’s some wobble from the stem in the housing, but that’s it. With the Raeds, they pop right out of the plate.
I should probably draw a diagram of this.
That is what I had imagined was the main issue since the bottom of the switch is effectively “open”, and yea you are most likely right in the sense that regardless of standoffs it will still probably slip out.
Haven’t seen that before, but what I was referring to was how some (crazy) people take the analogue sliders in GameCube controllers and replace them with (I think) low profile switches instead.
I swear I have seen people do it. I am not crazy myself. I think…
I am sorry to hear about your difficulties with the switches.
This person installed Raed HE domes on a Tiger Lite, and it seems to have worked out:
It was an experiment, since I’ve always been Topre-curious, but buying “ready-to-use” boards is pretty much the opposite of what I do. So been looking on/off for a project board.
Actually going to give the Tiger Lite/Venom HE TKL combo a try. Don’t think it will fix all the issues, but the case being able to flex more might actually help the issue some.
Also, looking closer at the Raeds, the plastic housing goes down further than I first thought…but only on two sides, where it has a small “peg” that anchors the silicone dome. Since the plastic goes all the way down and is anchored with the peg, pushing at an angle doesn’t deform things much.
On the other two sides, the plastic stops short and the silcone dome “floats” between it and the PCB. This is where the buckling is happening. And, unfortunately, there’s no way to orient the switch to keep it from happening.
I’m no mechanical engineer, but it seems like if they did the pegs on all four corners, it’d be much more stable.
The EC Naevies might not be as vulnerable to this issue, as the spring is going to provide some stiffness. Still need to get a plate cut for the EC60X.
Anyone know of a TKL EC PCB/mod out there?
I’m using raeds in a wooting HE60. I honestly think they’re some of the best switches I’ve ever used. It seems like all of the issues you experienced may have come from the plate not being cut to the required tolerances. Perhaps these switches require much more precise plates to work correctly?
I think it’s more likely the soldered on nuts that connect it to the plate. I spotted a few that didn’t look fully centered. The sticking would actually change depending on how much I’d tighten certain screws, as well.
The Tiger Lite also uses suspension mount, making the plate/PCB assembly float, which will help with flex problems (the force going to bending the suspension rather than deforming the switch). I was using tray-mount, which has two different surfaces to attach to, and has a lot less flex to it. Maybe o-ring mount might have done better?
But you say the problem is that the switches aren’t anchored into the PCB. So they seem to easily pull out and have movement. They should snap into the plate and not move at all. They don’t move on my wooting. I have to use a switch puller to get them out. So it seems that the holes on your plate may be cut too large for these switches.
Giving Pantheon’s “This Switch Is Called” linears a try today in my AL71. Oh man these are nice. Buttery, creamy, smooth - pleasant feel and delightful sound. Recording in the next day or so when the house is quiet.
