geekhack

geekhack Projects => Making Stuff Together! => DIY Discussions ARCHIVE => Topic started by: talis on Thu, 09 July 2009, 13:36:52

Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Thu, 09 July 2009, 13:36:52
I'm actually starting to think that a good portion of the key sound comes from the spring buckling against the little angled ridge in the key cap itself.  That would also give it the most tactility when it buckles.

(http://geekhack.org/attachment.php?attachmentid=3096&stc=1&d=1247164541)

I pulled a spring out of one of my boards last night, and could actually get a reasonably loud click just buckling the spring against that alone.  The spring being bound in the barrel probably contributes as well, as does the hammer.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: keyb_gr on Thu, 09 July 2009, 14:19:41
Speaking of buckling spring mechanisms, today I found this Alps patent (http://www.google.com/patents/about?id=J_kcAAAAEBAJ) describing one that apparently is supposed to improve upon IBM's. Now I'm no ALPSpert but I had absolutely no clue that they also looked into these. I wonder whether this ever made it into a real keyboard?
Title: Buckling Spring Model Understanding the IBM Patent
Post by: lowpoly on Thu, 09 July 2009, 14:52:29
Excellent work, ripster.

As for the sound source, I got the A01 from lam. It came with some spare switches. The spare switches, even with keycap, are a lot quieter than those in the keyboard. So, at least with the A01 much of the noise is generated by the hammer hitting the rubber sheet below. Might not be 100% transferable to regular Ms.

And that animated gif reminds me of a musician tapping with his foot. :smile:
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Thu, 09 July 2009, 15:41:25
Its not the barrel, its the wedge in the key cap, the spring hits it fairly hard when it buckles over and makes contact along a fairly large area.

(http://geekhack.org/attachment.php?attachmentid=3097&stc=1&d=124717200 8)

Watch how the spring whacks against the cap as it tips over (the top of the spring no longer makes contact with the key cap, it ends up at about 75deg angle.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: lowpoly on Thu, 09 July 2009, 16:03:10
Quote from: talis;101905
Its not the barrel, its the wedge in the key cap, the spring hits it fairly hard when it buckles over and makes contact along a fairly large area.

Watch how the spring whacks against the cap as it tips over (the top of the spring no longer makes contact with the key cap, it ends up at about 75deg angle.
That's the A01 removable switch. If the hammer doesn't hit anything the switch is fairly quiet. This perfectly fits with ripster's discovery about the hurting foot.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: lowpoly on Thu, 09 July 2009, 16:20:51
The spring does contribute but the 'clack' seems to come from the hammer. I'll check for details tomorrow, don't have the 'board here right now.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: timw4mail on Thu, 09 July 2009, 17:00:18
If the noise comes mostly from the hammer and the grease helps silence it, maybe the grease acts as soft of a shock-absorber?
Title: Buckling Spring Model Understanding the IBM Patent
Post by: lowpoly on Thu, 09 July 2009, 17:09:41
Maybe the grease makes the spring slower and thus indirectly affecting the hammer noise.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Thu, 09 July 2009, 17:55:07
I think the best way to test is to pull a board apart and hold the hammer stationary while pressing the key, then repeating with the hammer allowed to move freely.

I've been wanting to replace the rivets on my 401 for a while now, this may be a good excuse to pull it apart.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: ch_123 on Thu, 09 July 2009, 18:14:01
Quote from: webwit;101922
Noise comes from the spring and hammer, and by greasing it more of only the hammer click remains.

+1

Remember having this discussion recently enough here. (http://geekhack.org/showthread.php?t=5265&page=2)
Title: Buckling Spring Model Understanding the IBM Patent
Post by: ch_123 on Thu, 09 July 2009, 18:34:38
I don't think so, I'd read the posts in the thread I linked above... It definitely still "clicks" even when the hammer doesnt hit anything. The hammer is responsible for the metallic ringing - if you rest the assembly on a cardboard box, you get a hollow sound and a click. A wooden desk gives a hard thud and a click.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Thu, 09 July 2009, 23:48:24
Ok, as promised I tore my 401 apart just now.

The tools of the trade:
(http://geekhack.org/attachment.php?attachmentid=3101&stc=1&d=1247201072)

Poor thing never saw it coming:
(http://geekhack.org/attachment.php?attachmentid=3102&stc=1&d=1247201078)

One lone key left standing:
(http://geekhack.org/attachment.php?attachmentid=3103&stc=1&d=1247201086)

Setup:
(http://geekhack.org/attachment.php?attachmentid=3104&stc=1&d=1247201092)

I held the hammer stationary while I actuated the key with my other hand.  The click was almost exactly the same as when the board was together.  When I put the board down on my desk, and held the hammer in place with a screwdriver it sounded exactly like it does when fully assembled.

So my conclusions are :

1) The sound comes almost entirely from the spring, key stem, and key cap.  The foot slapping may add some small amount to the sound, but its not the primary source.
2) The back plate must resonate a bit and add to the sound.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Thu, 09 July 2009, 23:56:25
And while I have the board open.

Close up of the spring assembly from the back of the board:
(http://geekhack.org/attachment.php?attachmentid=3105&stc=1&d=1247201699)

Spring assembly out of the board:
(http://geekhack.org/attachment.php?attachmentid=3106&stc=1&d=1247201747)

It'll probably be a few days before I put the board back together, so if there's any more pictures anyone wants me to take, just let me know.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Fri, 10 July 2009, 01:31:27
Sorry the quality isn't super great.  The lens on my camera casts a shadow when using the flash in macro mode.  

(http://geekhack.org/attachment.php?attachmentid=3107&stc=1&d=1247207385)

(http://geekhack.org/attachment.php?attachmentid=3108&stc=1&d=1247207397)

(http://geekhack.org/attachment.php?attachmentid=3109&stc=1&d=1247207405)

(http://geekhack.org/attachment.php?attachmentid=3110&stc=1&d=1247207415)
Title: Buckling Spring Model Understanding the IBM Patent
Post by: talis on Fri, 10 July 2009, 01:33:49
Quote
And since you have it open do you have any opinion on what causes the metallic sound AFTER the key is pressed and then released?

I have a feeling its related to the spring top rocking back into position on the key cap as it un-buckles.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: The Solutor on Tue, 21 June 2011, 03:13:58
Quote
They are all quite different. I would NOT be surprised if a lot of the difference in key feel is due to the foot design, NOT because one is a [strike]capacitance switch[/strike] and the other is a membrane switch.


Not because one is a a switch and the other is not.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: greyhounds on Tue, 21 June 2011, 07:58:00
It looks to me like IBM added material to the hammer on either side of the spring connection point. This material isn't shown in the 4528431 patent sketches. There's a much more "U" shaped opening between the spring post and teh pivot arms on teh foot shown in the patent sketches.

I'm not sure if it was added for strength or maybe to give the ability for the key to work even if the spring was broken? I don't think that would work though as the downward froce from the key is in the wrong spot to rotate the hammer.

You can see that the earlier Model F hammer doesn't have that extra material.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: sealcouch on Wed, 22 June 2011, 20:10:45
This thing is awesome. Really smart idea.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: Bruce on Tue, 06 March 2012, 11:25:08
Ripster, this is genius. You are the man!
Title: Buckling Spring Model Understanding the IBM Patent
Post by: pol123456 on Mon, 02 April 2012, 14:08:23
hello,

we are French students.
As part of our studies, we make a model of a buckling spring keyboard key. We have seen the work of ripster and wanted to know if he could send us the characteristic of your models, particularly your spring.

we are makers of all councils, thanks.

Sincerely.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: alaricljs on Wed, 04 April 2012, 08:15:22
It's a spring, 4-5" long, ~.75" in diameter, ~5 turns per inch.  And none of that matters if all you're doing is making a model.  Any spring that isn't too stiff or too soft and has reasonable dimensions will do.  

PS - work on the english translation because this: 'we are makers of all councils' doesn't mean anything intelligible.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: pol123456 on Wed, 04 April 2012, 09:21:30
thanks, we will try with this sizes.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: alaricljs on Wed, 04 April 2012, 14:19:33
They used english, they can use SAE.  Insulting me doesn't help any.
Title: Buckling Spring Model Understanding the IBM Patent
Post by: alaricljs on Wed, 04 April 2012, 14:27:55
And where is this insult? and why didn't you report it?
Title: Buckling Spring Model Understanding the IBM Patent
Post by: pol123456 on Wed, 02 May 2012, 10:05:23
thanks you everyone for every things, we came back from vacancy. We will start with a wood model and we will do a new model in two weeks with the 3D-print.

we will post it soon as possible