PP Keyboard - Top Row Actuation
#1
I'm late to the party on the PinePhone keyboard's top row (number row) actuation issue, but I found 2 solutions to this problem that are simpler than making tiny washer-shaped spacers.



For anyone who has been living under a rock, here are the symptoms:

When pressing a key on the top row of the PinePhone keyboard, the key stroke does not register at the tactile bump. One has to keep pressing the key beyond the tactile bump for the key to register. That last bit of travel is very squishy, and this is the observation that eventually allowed me to understand the root cause of the issue.


The solutions:
  1. Add a tiny strip or 2 of tape on the inner face of the key cap on the hinge side.
    (no need for a shim going all the way around).

    With a pair of good scissors and some tweezers, this is a very easy and fast solution. On my keyboard, a single strip of masking tape made a huge difference, and adding a second one seemed to make it feel a little better.
    In my mind, the cons are durability and adhesive compatibility. I imagine the strips eventually shifting around or falling out of place.
    [Image: 8O72z01.jpeg]

     --- or ---

  2. Remove some material from the hinge-side rim of the keycap.

    This solution was the most effective and resolves the root cause of the problem with no compromises.
    [Image: f3Hmj2v.jpeg]
    [Image: asU2iGP.jpeg]
    [Image: t73nWer.jpeg]


But why tho?
Here's how I found these solutions.

I started by validating the original solution of shimming the inside depth of the keycap (vertical distance between inner face and outer rim). I measured a keycap from the top row with high precision vernier calipers and compared this measurement with the the lower row keycaps.

I found no difference.

To rule out the membrane switches on the top row, I moved a number row keycap to one of the letter rows and the problem followed the keycap. I also noticed that all of the membrane switches registered key presses perfectly without a keycap present. After doing a lot more measuring and playing with the membrane key behavior, I realized that when pressing a keyswitch down, the surrounding membrane is displaced and needs somewhere to go.

The top row keycap doesn't have enough space under the hinge side of the cap. The displaced membrane binds against the rim, preventing it from traveling all the way down. Because of this, the conductive ring under the membrane can't make it all the way down to the contacts unless you press harder to flatten the displaced rubber.
[Image: uatYtXS.jpeg]

To illustrate how it's supposed to work, here's what a lower row keycap looks like when depressed.
[Image: t2IHh5a.jpeg]

Here is how my solutions work.
  1. Adding a shim so the hinge side of the keycap makes that side not travel as far, creating room for the displaced membrane under the rim. It has a slightly squishy feel, but the key actuates reliably at the tactile bump.[Image: mNp8aq4.jpeg]

  2. Removing the portion of keycap that interferes with the displaced membrane allows the keycap to travel all the way down as intended.
    I say this solution is no compromises because the modification is completely hidden from view, it solves the root cause of the problem, and they key does not feel squishy at all. It feels absolutely great.
    [Image: owVTtiN.jpeg]
    (Pictured is the proof of concept test. The final modification removed a bit more material than you see here.)


I hope someone finds this info useful.

Thanks!
  Reply
#2
Thanks, nice detailed post. I would hit the Rate button, but I guess it doesn't appear until you've made a few more posts.

It looks like they were using 3d printing for rapid pattern making, so might be able to revise things for the next production run.
  Reply
#3
(03-14-2022, 06:26 AM)wibble Wrote: Thanks, nice detailed post. I would hit the Rate button, but I guess it doesn't appear until you've made a few more posts.

It looks like they were using 3d printing for rapid pattern making, so might be able to revise things for the next production run.

I'm glad it's appreciated. I was a bit worried about posting on a topic that was already stickied and closed.

I hope they are able to apply a fix at the factory. From a DIY fix perspective, adding tape or removing material are pretty quick and easy fixes. Solution 1 should only take a few minutes, and solution 2 took me around 30 minutes. Doing these things in a factory environment might be more complicated because they require a human touch and extra QA.

I would be surprised if they were using 3D printing for a large production run like this. My guess is the marks under the key cap are tool marks from when the molds were machined. Removing material from the keycap would require adding material to the mold, which I think means having to mill a new section of mold. I think it's probably less costly to source some adhesive mylar spacers to be added at the factory during assembly. Then again, I don't have experience with injection molding in a factory environment, and the change might be easier to make than I think.

Here's an additional thought to adding spacers: potential interference with the phone screen. My PP 1.2 has the glass screen protector and it already touches the keycaps when the keyboard is closed. The PPP is slightly thicker, so with the screen protector, the situation may be even worse. Adding shims under the keycaps makes them stick up higher and could have some long-term negative consequences for both screen and keyboard membrane. Maybe this concern is completely overblown, I don't know.
  Reply
#4
(03-14-2022, 09:47 AM)johnnycontrario Wrote:
(03-14-2022, 06:26 AM)wibble Wrote: Thanks, nice detailed post. I would hit the Rate button, but I guess it doesn't appear until you've made a few more posts.

It looks like they were using 3d printing for rapid pattern making, so might be able to revise things for the next production run.

I'm glad it's appreciated. I was a bit worried about posting on a topic that was already stickied and closed.

I hope they are able to apply a fix at the factory. From a DIY fix perspective, adding tape or removing material are pretty quick and easy fixes. Solution 1 should only take a few minutes, and solution 2 took me around 30 minutes. Doing these things in a factory environment might be more complicated because they require a human touch and extra QA.

I would be surprised if they were using 3D printing for a large production run like this. My guess is the marks under the key cap are tool marks from when the molds were machined. Removing material from the keycap would require adding material to the mold, which I think means having to mill a new section of mold. I think it's probably less costly to source some adhesive mylar spacers to be added at the factory during assembly. Then again, I don't have experience with injection molding in a factory environment, and the change might be easier to make than I think.

Here's an additional thought to adding spacers: potential interference with the phone screen. My PP 1.2 has the glass screen protector and it already touches the keycaps when the keyboard is closed. The PPP is slightly thicker, so with the screen protector, the situation may be even worse. Adding shims under the keycaps makes them stick up higher and could have some long-term negative consequences for both screen and keyboard membrane. Maybe this concern is completely overblown, I don't know.

Thanks on the detail analysis and already pass your finding to keyboard vendor. Thumbs up!
  Reply
#5
I can confirm the issue here - after a bit of Dremel reworking, the top row is definitively working better! I have no macro lens, otherwise I could have sent a photo showing just the inner part of the key towards the hinge 'dremeled out'
Congrats for spotting the issue!
  Reply
#6
(03-14-2022, 09:47 AM)johnnycontrario Wrote: I would be surprised if they were using 3D printing for a large production run like this. My guess is the marks under the key cap are tool marks from when the molds were machined. Removing material from the keycap would require adding material to the mold, which I think means having to mill a new section of mold. I think it's probably less costly to source some adhesive mylar spacers to be added at the factory during assembly. Then again, I don't have experience with injection molding in a factory environment, and the change might be easier to make than I think.

I would have said the same until someone posted video in another thread detailing the use of 3d printing to reduce the cost of tooling for short production runs. The tooling apparently won't last long, but is cheap to make and to iterate improvements with. I'm just guessing though.
  Reply
#7
(03-17-2022, 03:53 AM)wibble Wrote:
(03-14-2022, 09:47 AM)johnnycontrario Wrote: I would be surprised if they were using 3D printing for a large production run like this. My guess is the marks under the key cap are tool marks from when the molds were machined. Removing material from the keycap would require adding material to the mold, which I think means having to mill a new section of mold. I think it's probably less costly to source some adhesive mylar spacers to be added at the factory during assembly. Then again, I don't have experience with injection molding in a factory environment, and the change might be easier to make than I think.

I would have said the same until someone posted video in another thread detailing the use of 3d printing to reduce the cost of tooling for short production runs. The tooling apparently won't last long, but is cheap to make and to iterate improvements with. I'm just guessing though.

That's pretty cool. I didn't know something like that was possible. I'll have to check that out.
  Reply
#8
It occurred to me to point out that I modified my key caps by patiently scraping away material with a hobby knife. I would hate for someone to ruin or lose a keycap in an enthusiastic first attempt with a high speed rotary tool. Obviously, follow your own best judgement and use the tools you are most comfortable with.
  Reply
#9
I am dying to know if any of this have been incorporated into the new production run!
Cheers,
TRS-80

What is Free Software and why is it so important for society?

Protocols, not Platforms

For the most Linux-y experience on your Linux phone, try SXMO!

I am (nominally) the Armbian Maintainer for PineBook Pro (although severely lacking in time these days).
  Reply
#10
(04-30-2022, 04:58 PM)TRS-80 Wrote: I am dying to know if any of this have been incorporated into the new production run!

yes
  Reply


Possibly Related Threads…
Thread Author Replies Views Last Post
  keyboard case charging but not typing new2pinephone 18 5,210 02-19-2024, 07:30 AM
Last Post: theResonant
  Keyboard doesn't charge my phone no matter what, so can I charge my phone directly ragreenburg 3 5,509 01-14-2024, 03:34 PM
Last Post: acid andy
  I've added an I2C connector to the PP keyboard eugenr 0 359 11-26-2023, 07:24 AM
Last Post: eugenr
  Update the PinePhone keyboard driver to support missing characters? Peter Gamma 8 7,902 10-22-2023, 08:43 PM
Last Post: Peter Gamma
  key repeat with pinephone keyboard rotwang 0 392 10-22-2023, 03:39 PM
Last Post: rotwang
Exclamation USB mouse isn't working when a PP keyboard is attached? mikeb 3 1,111 10-21-2023, 08:21 PM
Last Post: Peter Gamma
  Pinephone with mini keyboard, trackpad and large power bank? Peter Gamma 0 447 10-21-2023, 07:50 AM
Last Post: Peter Gamma
  Which mini keyboard and mouse to choose for the PP? Peter Gamma 0 562 10-21-2023, 02:55 AM
Last Post: Peter Gamma
  Use the Linux Console with the PinePhone Keyboard mburns 2 903 10-20-2023, 04:11 PM
Last Post: mburns
  How to disassemble Keyboard top and replace cables? elof 2 717 10-11-2023, 09:03 AM
Last Post: elof

Forum Jump:


Users browsing this thread: 1 Guest(s)