Hall effect keyswitches

[philpirj]

http://www.meci.com/electronics/parts/switches/slide/hall-effect-keyboard-switch.html

Hall effect keyboard switch for $0.09 (0.61" square mount)
Not much info on seller site, but looks very similar to the one from Honeywell http://deskthority.net/wiki/Honeywell_Hall_Effect

They seem to have ~10k of those in stock.

Would be very appreciated if someone from CONUS could proxy some of those for me. Please PM. I would buy 250.
US: Ground $15.01 (cheapest option)
Russia: Priority Mail International $119.55 (the only option)

Canada shipping seems to be way too expensive, $54.50 for standard USPS.
Maybe it makes sense to do a group buy? Is there an option to order Space Cadet keycap clone for a 10k batch?

[Parak]

The PN doesn't match any known honeywell hall effects at all, although it does seem to resemble one. Some intrepid soul would need to buy a few to see what they are first, in case they are some superheavy latching variant or something (both of which exist).

[dorkvader]

The 4-pins do point towards hall-effect. I'll gladly mail forward you some. Depending if I can fit them in a small flat rate box, it's likely be $20-35 or so to russia. I would, of course, be interested in several, especially for $.10 each.

parak: the 540-0960 is listed as the SKU, it could be just that vendors internal sku for this product and not the actual model number of the switch itself. I'm contacting them for more info.

[jabar]

Please let us know, dorkvader. Also will need to know all the specs such as power requirements... I would love a custom hall effect keyboard.

[Halverson]

Hmmm may need to grab some of these myself!

[dorkvader]

Please let us know, dorkvader. Also will need to know all the specs such as power requirements... I would love a custom hall effect keyboard.

I can give you power requirements of "normal" micro switch hall effect keyswitches, I just have to desolder one.

Bythway, the honeywell micro switch ones are rated for 30 billion cycles.

[elttaboi]

Would it be possible to mount these on a regular cherry board? Very cool. I wonder what they feel like.

[The_Beast]

Would it be possible to mount these on a regular cherry board? Very cool. I wonder what they feel like.

Nope

[Halverson]

Would it be possible to mount these on a regular cherry board? Very cool. I wonder what they feel like.

Sadly no, and they also don't have many caps available.

And as for feel, they are linear.

[xmagusx]

According to this GH thread, the manufacturer is Micro and the part number is 1B3S. According to the Deskthority wiki page, this would indicate that it was:

• 1 - White stem
• B - Angled stem
• 3 - Normal (78cn)
• S - 4 pins, 2 [redundant] sense lines (Unknown difference from A)

This seems to be in line with the picture, which shows a 4 pin switch with an angled white stem.

[xmagusx]

Anyone played/built something with these yet?

[dorkvader]

Anyone played/built something with these yet?

Not these speciifcally, but I will be ordering some soon.

[xmagusx]

Have you worked with similar switches in that case? What did you build?

[dorkvader]

Have you worked with similar switches in that case? What did you build?

I "only" have two different types of microswitch hall effect keyswitches. The newer ones in my keypad from haata, and my older ones in a keyboard assembly from 1977 (with tripleshot keycaps) I've been planning how to actually mod it into a usable keyboard, and it'll be hard. The hardest part will be worrying about bottoming out (there is a soft landing pad under each keyswitch, between it and the PCB.) If you don't need to worry about that, then it should be pretty easy to take the mounting plates and take the PCB off and the solder it point-to-point. The next issue becomes linking up all the power and ground connections, which would be easier with the PCB.

Because of this, it would be much easier to mod this by keeping the original PCB, but it won't have NKRO unless I add diodes. So it's possible to keep the power and ground connections, and cut the traces to the rest. But it's a 2 layer PCB, so I'll have to desolder the switches from it to cut the traces on top of the pcb: big hassle. Maybe it's easier to just solder up 3 connections per switch, wiring point-to-point.

anyway, I'd be using this as a gaming keyboard, for obvious reasons, so layout isn't that important. I can get away with a lot. Right now I'm closely inspecting the switches I have and looking into mounting plate options. (recreating the original microswitch mounting plate will be a pain, but If I can make a cherry/alps style plate, that'd work well)

[HongKongFui]

What is a hall effect switch? What is it doing?

[ndlu2]

What is a hall effect switch? What is it doing?

Hall effect switches use magnetic fields and tend to last very long and have a linear feel.

[xmagusx]

Welp, mine have arrived. Wish me luck in projectland!

[dorkvader]

Welp, mine have arrived. Wish me luck in projectland!

I am very interested in any pictures you may have of the actual switches.

[xmagusx]

Ask and ye shall receive. All six angles. Final picture features three switches, the upper and lower one holding up the middle one.

Edit: Can't seem to figure out how to change attachments, and accidentally added one of the side views twice instead of the other side view. Last picture in next reply.

[xmagusx]

Goofed on previous.

[dorkvader]

Wow, thanks very much for the pictures! Now that I've acquired another set of keycaps, I think I'll get started in making a "modern" hall effect KB in earnest. I'll have to figure out the plate hole dimensions from my BUD keypad and decide on a layout. It'd be great to have a 3D printer handy so I can get some blank keycaps, though. I strongly suspect it'll be easier to make keycaps for these.

I notice the "old style" HE switches require a mounting PCB for the sensors. Mounting these is much easier with the spring included in the switch housing (old style are mounted on a bump on the plate) as well.

[xmagusx]

My goal with these is to first build a keypad (probably a 5x5) as a way to learn how to DIY a full keyboard, so if you stumble across a way to procure inexpensive keycaps for these guys, please do let me know.

[TheFlyingRaccoon]

Just ordered 50 of these. Plan to make a gamepad like this:

I have access to a 3D printer so the keycaps and case shouldn't be a problem. This is gonna be a fun project. 

[dorkvader]

Just ordered 50 of these. Plan to make a gamepad like this:

Show Image

I have access to a 3D printer so the keycaps and case shouldn't be a problem. This is gonna be a fun project.  :)

Sweet! This is so exciting!

I'll be hopefully in there with you as well, making a small keyboard (for gaming). I'd be very interested in keycaps as well. The hardest part will likely be stabilizing the spacebar. I'm studying up on these, so hopefully we can help eachother out some.

Honeywell has a PDF of their hall effect sensors. It's a lot of more "general" hall effect stuff, but there's some specifics on the keyswitches. This is where I get my 3*10^10 actuations number. Lots of good information in there!
sensing.honeywell.com/index.php?ci_id=47847

If their 100 KHz scan time is any sort of guide, I would expect the debounce time to be quite short, but this number could be inherently meaningless. It won't matter if you scan the matrix 100,000 times per second, if it takes a while to debounce. Viewing the waveform on my oscilloscope (with a camera) shows pretty quick ramp up from ~0V output to ~5V output (the switch outputs 5V directly when it detects a keypress) so I would expect this would make it useful for high speed applications (gaming) but we'll see how quickly it can actually actuate and reset in the real world, and how much hysterisis there is (there in some information about adding hysterisis to hall effect sensors. No idea if it's included here or not)

[xmagusx]

I'll be hopefully in there with you as well, making a small keyboard (for gaming). I'd be very interested in keycaps as well. The hardest part will likely be stabilizing the spacebar. I'm studying up on these, so hopefully we can help eachother out some.

If the goal is a gamepad, I'd suggest going with a couple of smaller centrally balanced buttons that don't require stabilization rather than one giant one, unless part of the project is specifically to learn spacebar stabilization.
Example with 21 1x buttons and 2 2x buttons (that still allows for both three finger salutes):

Code: [Select]

|esc|| 2 || 3 || 4 || 5 |
| 1 || q || w || e || r |
|ctl|| a || s || d || f |
|shf|| z || x || c || v |
|del||  alt   ||  space |


[TheFlyingRaccoon]

Just ordered 50 of these. Plan to make a gamepad like this:

Show Image

I have access to a 3D printer so the keycaps and case shouldn't be a problem. This is gonna be a fun project.  :)

Sweet! This is so exciting!

I'll be hopefully in there with you as well, making a small keyboard (for gaming). I'd be very interested in keycaps as well. The hardest part will likely be stabilizing the spacebar. I'm studying up on these, so hopefully we can help eachother out some.

Honeywell has a PDF of their hall effect sensors. It's a lot of more "general" hall effect stuff, but there's some specifics on the keyswitches. This is where I get my 3*10^10 actuations number. Lots of good information in there!
sensing.honeywell.com/index.php?ci_id=47847

If their 100 KHz scan time is any sort of guide, I would expect the debounce time to be quite short, but this number could be inherently meaningless. It won't matter if you scan the matrix 100,000 times per second, if it takes a while to debounce. Viewing the waveform on my oscilloscope (with a camera) shows pretty quick ramp up from ~0V output to ~5V output (the switch outputs 5V directly when it detects a keypress) so I would expect this would make it useful for high speed applications (gaming) but we'll see how quickly it can actually actuate and reset in the real world, and how much hysterisis there is (there in some information about adding hysterisis to hall effect sensors. No idea if it's included here or not)

Thanks for that pdf. And your findings are pretty cool. You're right about the reset point, I think some extensive experimentation with actuation and debounce time would be needed before these were actually implemented in a keyboard.

I'll be hopefully in there with you as well, making a small keyboard (for gaming). I'd be very interested in keycaps as well. The hardest part will likely be stabilizing the spacebar. I'm studying up on these, so hopefully we can help eachother out some.

If the goal is a gamepad, I'd suggest going with a couple of smaller centrally balanced buttons that don't require stabilization rather than one giant one, unless part of the project is specifically to learn spacebar stabilization.
Example with 21 1x buttons and 2 2x buttons (that still allows for both three finger salutes):

Code: [Select]

|esc|| 2 || 3 || 4 || 5 |
| 1 || q || w || e || r |
|ctl|| a || s || d || f |
|shf|| z || x || c || v |
|del||  alt   ||  space |


Yes, I like your design much better. A 2x key for the spacebar should be long enough for comfortable use, but not long enough to require stabilization. Although the alt key should probably be split into two 1x keys.

[xmagusx]

Yes, I like your design much better. A 2x key for the spacebar should be long enough for comfortable use, but not long enough to require stabilization. Although the alt key should probably be split into two 1x keys.

Wasn't sure how much use the left of two 1x keys in that space would use, given that it would be an awkward reach for any of the fingers or thumb. Might make a good spot for a tab key though, since that frequently just brings up a scoreboard and other useful-but-infrequent game functions.

Code: [Select]

|esc|| 2 || 3 || 4 || 5 |
| 1 || q || w || e || r |
|ctl|| a || s || d || f |
|shf|| z || x || c || v |
|del||tab||alt||  space |

This would also allow for one more button in a 5x5 grid, which could take the form of a side-mounted 1x button in the bottom row if you can get the frame and housing to cooperate.