Will it click?

[nubbinator]

So, I saw the MX compatible buckling spring thread and I was intrigued.  OP really seemed stuck on the design where the spring was part of the switch mechanism and the stabilizer was the other part, making it MX mount, but requiring a specialized PCB.  These two parts really didn't seem to be the optimal design, so I decided to play around and work on my own version.

I figured why my with what works and piggback on the original design.  I have no idea if it even works yet as it's in the drafting stage and I am anything but an engineer, but I'm having fun working on it regardless of whether or not it works.  If anyone sees any obvious flaws, please chip in before I get too far along since it's a laborious process for someone blundering his way with Autocad's 123D.

Without further ado, pics of what I have so far:








As you can see, the plan is to stick with the hammer design.  I still have to figure out the contacts, but once I get it clicking, I can work that out pretty easily.  The unfortunate part of the hammer design is that it adds about 3mm to the height, but some of that is mitigated with the slider and planned switch top design.  The spring sits pretty much flush with the top of the switch and may actually extend a little past it.

I'm thinking of making the slider mechanism's slider legs much longer.  I went with four legs since it should make it more stable and making the legs longer would provide a twofold purpose, they would further stabilize it and they would also serve as stops for bottoming out so you can get the caps on them without a problem.  I'm not sure if four legs is the best or if I should try and switch to closer together legs or two legs.

If you look at the housings, you can also see I'm looking at two different designs, one that would require LEDs to be placed at the top or diodes inside at the bottom (which might or might not interfere with operation.  The other is a solid wall with a cutout for a 2x3x4mm LED that would let an LED be housed almost entirely in the switch housing and just a tiny nib sitting in the switch top to stabilize it in place.

And, yes, unless a shorter weaker spring can work, it will have to be that tall.

Any and all input for improvement or fixes would be greatly appreciated.  Suggestions for a switch top securing system would also be nice.  I want to try something a little different so you don't have to worry about whether there are cutouts for switch top tabs or not.

[rowdy]

Would this be a drop-in replacement for Cherry MX or compatible switches?

Apart from the extra height, of course.

Or is it a switch that would need a new keyboard design?

[nubbinator]

I'm trying to make it an exact drop in replacement.  It will be taller, but would fit a standard PCB and plate.

[ghostjuggernaut]

I think you may want to add some fillets to the stabilizer legs on the top piece. I'm not sure if it will help much, but it may help prevent any binding that could occur from an off center press.

[nubbinator]

There's some minor fillets there.  You don't really notice because of the distance.  I'm hoping having longer legs and the fillets will prevent binding.

[Zekromtor]

Possible contact method: have the hammer engage two separate contacts like so:


The shape is a little more complex, but it's still just a single 90 degree bend for each contact. To ensure the top piece is not normally in contact with the bottom, add some height to the bottom of the switch here:


Obviously you'd want the hammer designed specifically to engage at only the two contact points, so you'd want to add or remove material to facilitate that.

Having two contact points doesn't make any different under ideal conditions, but when foreign material creeps its way into your switch, it can be the difference between the switch remaining fully functional and requiring repeated cleaning.

[Melvang]

Possible contact method: have the hammer engage two separate contacts like so:

Show Image

The shape is a little more complex, but it's still just a single 90 degree bend for each contact. To ensure the top piece is not normally in contact with the bottom, add some height to the bottom of the switch here:

Show Image

Obviously you'd want the hammer designed specifically to engage at only the two contact points, so you'd want to add or remove material to facilitate that.

Having two contact points doesn't make any different under ideal conditions, but when foreign material creeps its way into your switch, it can be the difference between the switch remaining fully functional and requiring repeated cleaning.

I would stick with a single contact point, and cross style similar to Cherry.  The cross point style is actually pretty efficient at self cleaning.  And of course, I wouldn't go with anything less than gold plated, for both corrosion and electrical arcing problems down the road.  Granted the current and voltage levels don't generally warrent these measures, but with potential switch life in your tens of millions of cycles, it can make or break.

[njbair]

Does the hammer have to be the means of making electrical contact? What if you kept the hammer in there for tactility, but used a separate mechanism for the electrical operation? Like how Alps does it with the click leaf. That way, you can fine-tune the feel of the switch without worrying about affecting its operation.

[Zekromtor]

Possible contact method: have the hammer engage two separate contacts like so:

Show Image

The shape is a little more complex, but it's still just a single 90 degree bend for each contact. To ensure the top piece is not normally in contact with the bottom, add some height to the bottom of the switch here:

Show Image

Obviously you'd want the hammer designed specifically to engage at only the two contact points, so you'd want to add or remove material to facilitate that.

Having two contact points doesn't make any different under ideal conditions, but when foreign material creeps its way into your switch, it can be the difference between the switch remaining fully functional and requiring repeated cleaning.

I would stick with a single contact point, and cross style similar to Cherry.  The cross point style is actually pretty efficient at self cleaning.  And of course, I wouldn't go with anything less than gold plated, for both corrosion and electrical arcing problems down the road.  Granted the current and voltage levels don't generally warrent these measures, but with potential switch life in your tens of millions of cycles, it can make or break.

I agree self-cleaning is preferred over dual contacts (both would be optimal) but isn't the self-cleaning nature of the cherry mechanism nearly all due to the contacts being vertical? Seems like gravity is the real clean there.

With the hammer as the actuator, he's limited to horizontal contacts I believe. How are you envisioning him using the cherry contacts with the hammer?

[Melvang]

Possible contact method: have the hammer engage two separate contacts like so:

Show Image

The shape is a little more complex, but it's still just a single 90 degree bend for each contact. To ensure the top piece is not normally in contact with the bottom, add some height to the bottom of the switch here:

Show Image

Obviously you'd want the hammer designed specifically to engage at only the two contact points, so you'd want to add or remove material to facilitate that.

Having two contact points doesn't make any different under ideal conditions, but when foreign material creeps its way into your switch, it can be the difference between the switch remaining fully functional and requiring repeated cleaning.

I would stick with a single contact point, and cross style similar to Cherry.  The cross point style is actually pretty efficient at self cleaning.  And of course, I wouldn't go with anything less than gold plated, for both corrosion and electrical arcing problems down the road.  Granted the current and voltage levels don't generally warrent these measures, but with potential switch life in your tens of millions of cycles, it can make or break.

I agree self-cleaning is preferred over dual contacts (both would be optimal) but isn't the self-cleaning nature of the cherry mechanism nearly all due to the contacts being vertical? Seems like gravity is the real clean there.

With the hammer as the actuator, he's limited to horizontal contacts I believe. How are you envisioning him using the cherry contacts with the hammer?

The contacts being vertical does help, but the big thing is the contacts are at 90° to each other and make contact at a single point.  Instead of to contacts that essentially look like an = when viewed from the side or l from the top. 

Plus the contacts are shaped so that a corner comes in contact with another corner, not a flat on a flat.  Think of it as two knife edges coming together.  Contact is guaranteed.  But you can set two 4'x8' sheets of steel on top of each other and not have physical steel to steel contact from relatively little sand between the two.

[Zekromtor]

Ah, yes. Point taken. Flat on flat is no good. The fix is rather simple though, just to bend one or both contact points so that the contact is made at the crease similar to cherries. That would require more headroom in my design, but he's got plenty of that.

[nubbinator]

Now you guys have me wondering, should I chop the front off of the hammer and modify the planned slider legs to bump the contact like with a linear MX switch?  That way it could just be an MX leaf drop in instead of designing a new leaf.  The leaf may interfere with the spring though.

[Zekromtor]

I vote no. The idea of separating the feedback and actuation is not a good one IMO. You want to ensure that they are one in the same so that you never have one firing without the other. I can't think of anything less satisfying than a beautiful buckling spring followed by no actuation, or vice versa.