How do the keys trigger sound?

[Robbins]

Hi. I am planning on creating a modified keyboard from an ordinary keyboard. The modification would be done using 3d printed keys. I would create new keys, but would not modify the body of the keyboard.
For this modified keyboard to work optimally I might need to drastically reduce the distance the keys are able to travel from their resting position to fully pressed position. That is, the key depth (key dip) in this keyboard should be very small.

To create this shallow dip I thought of extending the bits of the keys which presses the buttons that triggers the production of sound. The buttons which I mention are the little circles in the grey stripes on the attached picture.

If I extended the button pressing bits of the keys, the sound would be triggered even when the key is only slightly depressed. I also thought I would have to extend the part that ultimately stops the key from dipping any further. This way the extended button pressing bits wouldn’t crush the buttons beneath them when the keys are fully pressed.

Seemed like a reasonable plan, but then I realized that the keys in a normal keyboard trigger sound not in the bottom of the dip, but in mid dip. This would imply that the keys in a normal keyboard already kind of smash the buttons. It would also imply that the buttons are depressed themselves by the keys, which press and displace them. In principle my plan should work even if that is the case, but this seemed odd to me, and I wonder if I don’t have a completely wrong idea of how the mechanism actually works. If it works in a very different way as the one I imagine, I may make a mistake in the keys alteration that may damage the keyboard (and make me sad).
Can anyone explain how this mechanism works? Are the buttons really depressed with the keys pressing them?

Thanks in advance.

 

[happyrat1]

There are two switches to every key in a velocity sensitive keyboard.

The first one activates at the top of the stroke. The second one activates at the bottom of the stroke.

The time interval between these two events is calculated by the CPU of the keyboard to determine the velocity (ergo the force) with which the key is struck.

The shorter the interval, the greater the force, ergo the louder the note.

That in a nutshell, is how these things work.

If you are designing a custom keyboard I would suggest building some sort of adjustable test jig with a single key so you can measure the dimensions necessary to achieve a desired pulse width or interval.

Bear in mind you are redesigning the wheel which countless hundreds of electrical engineers have perfected over the past 40 years since these boards were first invented.

If you're game for a better project, might I suggest ditching the silicone membrane switches in favor of optronic switches which would never wear out or foul up with oxides and grit as most commercial keyboards eventually do.

I designed such a switch once but never got around to building it.

I posted a thread on EEVBlog a few years back.

What I'd like to see on my next Musical Keyboard - Page 1

What I'd like to see on my next Musical Keyboard - Page 1

www.eevblog.com

The schematics have expired from the image host I used back then.

I will search my machine to see if I can recover the image.

Gary

 

[happyrat1]

I've found the drawings.

The first design uses a single optronic switch and calculates the velocity by the period of the pulse as the beam is interrupted or transmitted depending on whether you use positive or negative logic.

The second design adds a second switch to detect aftertouch if you wish to add that feature.

Use it in good health.

Gary

 

[Robbins]

There are two switches to every key in a velocity sensitive keyboard.

The first one activates at the top of the stroke. The second one activates at the bottom of the stroke.

The time interval between these two events is calculated by the CPU of the keyboard to determine the velocity (ergo the force) with which the key is struck.

The shorter the interval, the greater the force, ergo the louder the note.

That in a nutshell, is how these things work.

If you are designing a custom keyboard I would suggest building some sort of adjustable test jig with a single key so you can measure the dimensions necessary to achieve a desired pulse width or interval.

Bear in mind you are redesigning the wheel which countless hundreds of electrical engineers have perfected over the past 40 years since these boards were first invented.

If you're game for a better project, might I suggest ditching the silicone membrane switches in favor of optronic switches which would never wear out or foul up with oxides and grit as most commercial keyboards eventually do.

I designed such a switch once but never got around to building it.

I posted a thread on EEVBlog a few years back.

What I'd like to see on my next Musical Keyboard - Page 1

What I'd like to see on my next Musical Keyboard - Page 1

www.eevblog.com

The schematics have expired from the image host I used back then.

I will search my machine to see if I can recover the image.

Gary

Hi, Garry.
Thanks for the reply.
I already read similar explanations about how the velocity system works.
My doubt is about what happens after these buttons are pushed and the keys keep pressing them down. Because, as I said, the sound is triggered midway, not near maximum depth. If I do the alterations I mentioned should I be worried I might damage the buttons/keyboard?

I'm quite ignorant about engineering. My aim was to copy the key exactly and modify only the part that interacts with the musician. But, the depth of the normal keyboard is too big for what I intend to do (well, it is ok, but not optimal).
I am building a symmetrical keyboard, with no differentiation of black or white keys, all keys will be the same. The keys would also be all in the same horizontal plain, in opposition to the traditional keyboard in which black keys are higher. In a sense it would be as if the keys are closer together, with the black keys extending all the way to the same length as the white ones and also lowered to be in the same plain as the white ones.
With this design, if your finger plays a note at an angle and the key dips too deep, it is easy to hit a wrong adjacent note. That is the only reason why I intend to try to alter the side that interacts with the body of the keyboard. I would prefer to not change anything at all on that side.
I appreciate the optronic sugestion, but it's just way beyond me.