Electro magnetic sensing keyboard springs for poly AT

[John Hendry]

Been thinking a lot about electro magnets as proper keyboard springs lately. You can check the compression on a car engine by monitoring the starter current so you just threw out the cost of a key switch being needed to give you movement and location information. After touch is provided by a simple cheap rubber density pad not reporting itself just creating a stop surface that allow a little more movement with added key pressure, poly AT comes free built in with the system by registering past stop key location. I have a tall short wish list I guess where synth control reaction is as important as everything else connected and that’s cheaper than using lasers and you can adjust the spring tension as needed…..wonder what it would make the action feel like done right. I don’t think most people would notice the difference till they adjusted it.

Seems to me you could program in hammer strike string reactions that match AT using proper pad density per string lower strings feeling loser and other effects of resistance related to real movement in time such as a throw that’s “spring” rate resistance doubles giving you two key board levels as close as you’ll ever get them. Second static uniform key layer easy to do. Could be time related in "existence" programming. Timing is everything and that would add a new dimension within it giving you two walls to bounce off with one wall disappearing in time to send you flying off into the back of the first wall and bounce off it…unless it disappears and that’s a good question where you bounce next but its timing and you’ll find out.

I gota lay of the coffee….sorry. But hay, it’s cheap in quantity to build as its just monitored voltage resistance controlled electromagnets that never touch each other! Just a wire wrapped around a light wood stick going through a loop of wire. That is the basic keyboard’s hardware less circuit boards you needed anyway replacing their parts with a clock driven circuit monitoring computer control system and I just threw away the expensive addition of needed polyphonic after touch hardware and the normal switches so this deserves some more coffee… back and see I forgot the metal springs….they cost more than the wire perhaps. Gotta name it, EM> Spring Action. That’s it. Don’t buy one unless I get paid for it;-) My idea….

Attractive magnets give an unstable static force, but the other side is ok. Energy is a constant to measure with, magnetism is a conservative force and gives little if any damping and that’s what we want from a spring to create feel. Key action (usually) controls dampening, hammer actions etc, however this is all programmable too if programmed to “feel” that way even though key C/G is fixed as key movement is recorded. All keys with real hammer action C/G is fixed too and non moving at point of rotation and feel of hammer C/G movement is sensed as variable in rotation toque and this could be programmed in. Weight and its moment arm in key could be adjustable very easy but at 2 minutes an adjustment that’s 3 hours and 6 will be more realistic on 88 keys so key oscillation speed preference setup only. This is just a spring that knows where its key is at to save money.

I don’t know…what do you think. Only got to make one key to start with and see what it feels like and imagine controlling it’s resistance curves and try it to see what it feels like controling voltage. Imagine 2nd "key plane of view level" using voltage increase at that key location. That would be cool….you hit the key 5 on a scale of 10 and it stops half way, then harder and go into second layer to base at 6 or 7. That for me would be same sound deepened say or opposite of that, anything really….a new way to do something and get poly AT free. If you are just measuring the resistance and can control it with EM……it makes sense;-)

Back to what you think till the coffee is gone. If I’m crazy just tell me….but its seems really simple using what we now have compared to other things we do with it and it eliminates all mechanical parts down to the key and pivot rotation point and one electromagnet that could have 2 windings and 4 wires for second level force. It has few mechanical parts to build and seems cost effective as the fancy stuff is just programming a computer to control voltage at specific key location points, so might as well add time for movement controlling walls that open up. You could walk right through them as they open up. You could keep it simple at first and just use it as a sensing spring that would be easy to adjust tension to start with and 2nd layer feel/resistance is just a second winding in the electromagnet controlled by key location switch in circut with no computer needed for static second layer, all AT is polyphonic as requires no additional key sensor, just compression pad.

John^^

[Sharp]

Seriously, you should patent that idea.
I can think of all sorts of benefits from doing this. Imagine having an area in GLOBAL mode on your KORG where you could adjust the current going to the magnets. You could control the feel of the keyboard.

lol... that's just fantastic.

Regards
Sharp.

[Timo]

Haven't a clue as to how real piano keys actually work in the real world, or even synth keybeds, so a lot of your description went straight over my head by some several goalposts, but in layman's terms am I getting the gist that you wanted to use electromagnets as a repelling force to push the keys upwards giving them (customisable) "resistance", instead of springs? Along with another electromagnetic circuit to detect how far close (ie. pushed down) a key is at any time, along with a foam pad to allow you to push down further for "aftertouch"?

[John Hendry]

First, I make this this stuff up as I go. I found this looking at the math of soloniods AFTER I got the idea.

http://ocw.mit.edu/NR/rdonlyres/4775A30 ... lenoid.pdf

EXAMPLE: ELECTROMAGNETIC SOLENOID
“A common electromechanical actuator for linear (translational) motion is a solenoid. Current in the coil sets up a magnetic field that tends to center the movable Armature JFH^^ Key.^^

On the electrical side the device behaves like an inductor

—but the inductance depends on the position of the movable armature.
This “position-modulated inductor” is properly represented by a two-port
energy-storage element with an electrical port and a mechanical port.
On the mechanical side, a force is required to displace the armature from its center position

—the device looks like a spring.”

It is a spring. Monitor the EM key circuit and you have a free sensor to go with the spring giving you release velocity sensing as well as polyphonic after touch. This is keeping me up, not the coffee.

John^^

[John Hendry]

Release velocity like on the Roland JD800 is found in any EM system just as poly AT is and can be done very simply. When EM system is used as the key’s spring the system uses a lot more energy but there are ways to deal with it. I am a mechanic not a book keeper or one with money to pay attorney fees right now and have been here before so rather than another lost good idea, if it is one, I am putting my name in good faith on it and these are just preliminary spec's:

Class A Keyboard keys fall into key-bed when turned off and are 100% EM> sprung and program dampened and will use the most power with 88 x spring rate rest energy used when keys are up.

Class A/B is programmed or mechanically switched so 100% constant 88 x spring rate rest energy is not used at exact top key location creating an adjustable hard to silky smooth action saving energy. Like class A, keys fall into bed when turned off.

Class B uses a very light spring to hold the key up and has all A/B features. Keys stay up when turned off. The B system reduces energy use relative to the spring tension used, but is intended as key support system not energy saving solution and I suspect a downgrade in feel from class A.

Class B/B uses a simple weak air bladder of proper design instead of a spring as a key damper that may be wanted with EM springs at higher spring rates. Just an idea. Keys stay up when turned off.

Class C uses normal springs for majority spring rate to 99.99% and uses EM for key location information only, not as a spring resistance system leaving out the crazy good stuff but putting in the basics: all direction velocity response and polyphonic after touch and looking a lot like what we already use from the outside. .

Class C2 adds second switched or programmed key level view plane location invoking EM spring tension creating a grater abrupt resistance layer. This will show the difference in “feel” between a normal keyboard spring and EM spring to talk about. From here it is not such a drop from Class A performance saying EM spring interaction is even wanted. I think I want it. I think direct EM link will connect you better and feel really nice done right allowing the keyboard to evolve to literally programming layers in future time affected by past time layers, just like in real life. Too cool if you see what I am saying… the delays creating walls staying ahead of you to use.

Class C is low current design (smaller EMs) and only requires a class A EM spring to add second inertial C2 layer controlled with an on/off switch. This is basically simple caveman magnet technology and very practical with no touching EM spring parts. I would expect it to cost less to manufacture than current high tech micro key switch systems used that have always needed polyphonic after touch and this adds release velocity control as well without increasing costs to do this, an impossibility without some caveman tech like this applied. This takes costs the other way because you are using an inexpensive EM spring’s resistance to monitor the system.

Question is, does an EM spring as a “working sense-able spring” really work. Since complex “whole energy” engine inspection computers and hybrid electric cars using "whole energy" theory applications as well as EM suspension systems putting energy beck into electric motors is working way past our simple needs here it works and only way to see how it responds as a spring is try it and my money says if it works its huge because of the feeling of the response system it creates. I have created something very simple here for poly AT worth concideration I believe. Cost of EM key springs designed properly should be low getting a much needed job done exceptionaly well and inexpensively.

Its how much energy EM springs need that has my attention now but I see no problem with consumption kept down by using close tolerances. Biggest problem is Class A boards look funny turned off without a key-up parking device that adds weight and expense not needed. Class B may resolve this but I’m already used to look of keys falling down when off. For modern music I don’t know of anything outside of a Neurophone acting as a mic that it will get better than this just by adding a computer to control the key's spring resistance to create felt layers of key view plains created in time as well as monitor key location and direction. This technology opens new doors into music. Imagine a time in a pattern or song when the keys or group of keys stay down at 50% offering that level sound only letting you get there quietly. That's class A. The programming possibilities are endless.

I always saw midi time control as a valuable tool playing with the Elka MK76 that’s poly after touch and cool midi features made manufacturing too expensive. The need for poly AT creates a need for a low cost and better solution and this might just be found in class C control.

John^^

[dfahrner]

I'd been thinking about this too - to a lot of us (older) gigging keyboardists, a lightweight keybed is almost as important as the action. The problem is that the weight added under each key will be multiplied by 61 or 76 or 88, so even very small weights really add up. Even small air-core coils could amount to a large total weight. Has anybody done any experimental work to see just how much wire would be required in the coil(s) to provide the necessary force?

You would think that this idea would have occurred to Yamaha, Roland, Korg, and the other manufacturers, too - I wonder why we've never seen it implemented?

df

[Timo]

Guess it also depends on the new, additional constant drain of electrical current required (multiplied throughout 61, 76, 88 keys), as well as added heat and/or fire risk in the event of spillage, rain, etc..

[X-Trade]

Biggest problem is Class A boards look funny turned off without a key-up parking device that adds weight and expense not needed. Class B may resolve this but I’m already used to look of keys falling down when off. For modern music I don’t know of anything outside of a Neurophone acting as a mic that it will get better than this just by adding a computer to control the key's spring resistance to create felt layers of key view plains created in time as well as monitor key location and direction. This technology opens new doors into music. Imagine a time in a pattern or song when the keys or group of keys stay down at 50% offering that level sound only letting you get there quietly. That's class A. The programming possibilities are endless.

I don't think i'd have a problem if the keys fell down when it was off - at least you'd have a definate way to tell that it is off

additionally, I could see this in a WS or controller keyboard accepting MIDI for the current channel so that you can see which notes/keys the sequencer is already playing, or for learning/ease of use purposes 'disabling' the keys which are off-key (e.g. kaosillator scales).

I lve this idea, I used to come up with all kinds of stuff like this...
you should build it!

I can imagine that high current spillage/shock danger could actually be reduced over normal keyboards because like pickups, magnets can be enclosed without affecting their forces - you don't have to expose the delicate or dangerous electronics...


keep your mobile phone away though.... and your digital watch. pesonally I get paranoid about magnets getting near any of my personal electronic devices e.g. pda etc...

[RiotNrrd]

See, this is MUCH better than the DNA Spiral Time\Space Relativity Thesis stuff.

MUCH better because it's a really cool idea (and music related). Quite novel.

For a studio keyboard, weight wouldn't matter much, and I think that having the ability to easily customize the keyboards action, on a note by note basis, if desired, is a fantastic feature to have. Sometimes you want mush-boards, and sometimes you want a very crisp feel, and all points in between.

I do wonder, though - part of the feel of a keystrike is in the strike. When the key becomes fully depressed. I personally like a definite, "solid" bottom to the keys, and I wonder if a magnetically generated resistance could produce that satisfactorily?

[John Hendry]

RiotNrrd said: “

"I do wonder, though - part of the feel of a keystrike is in the strike. When the key becomes fully depressed. I personally like a definite, "solid" bottom to the keys, and I wonder if a magnetically generated resistance could produce that satisfactorily?"

Well, it all boils down to how much energy you use. Using high current EM springs you could probably break a finger on it them like any hard surface, and if a down key sprang up un-dampened at full force catching a finger it would really hurt. But its like an escalator and you’d have to try to hurt yourself but come to think of it could be user programmed any way so a mass return speed over ride protection program on such a high current design would be a smart idea.

I was thinking more of an arrival feel on second layer but come to think of it velocity or set state layers could increase layer voltage at that location and give you a hard stab feeling, or crisp feeling layer like breaking through thin rice paper held tight if you tap or push too hard. That could sound just heart breaking, or give you a rush.

In theory using EM base with the rubber AT pad part of its stop key mass you could just about match the “hard feeling” using stiff parts and enough energy but I see using a reverse EM spring to hold keys tight to replace key stops that have keys tensioned for this type of level feel. The more I look at this the more the simplicity of a normal sensor solenoid having two EMs makes complete sense, no pun intended. This is not a mechanical marvel outside of class C use going unnoticed, it’s a programming marvel where the manufacture will be overloaded with requests of “this is how it should be done” from one guy and “man its gotta look like this when I get there” from another. Its far less programming and more straight forward however than the sound within the walls and setting it up will be easy if Korg builds it, not too bad if Roland builds it, and very difficult to use if Yamaha builds it but built well;-)

John^^

[Sharp]

At this stage I hope you have your name securely attached to this idea by whatever means are necessary so it's not stolen on you.

You should consider contacting the guys at Fatar who make keybeds for KORG and many others ?. http://www.fatar.com/

This is a great idea and you clearly have the knowledge to explain this idea to them. You never know where it will end up going.

Regards.
Sharp.

[ausser]

Well... it wouldn't be a WS seller for me.
I'm a 'Live with the Action' kinda guy - if its anyway workable atall.

It certainly would be interesting to see such a keyboard developed though - and to try it or get reports on how it 'feels' from some of the Classical Circuit guys....

However, I can't see any of the big manufacturers interested in it because of the way Keyboard action has developed - ie. its hammer action now, which doesn't use 'Ye Ole Weighted' action set-up - so the weight is massively reduced to start with - weight being a big factor generally, but specifically for 'giging' musos - lessen your big enough to have a few roadies.
The new 'Hammer Action' is very very good - as it is, so I see no real need for something else.

Also, I imagine the weight due to the magnets - sheilding - or the EM Coils would be substancial - maybe even more than the ole weighted action - unless you've got some other idea aside from magnets - I'm no EME -
But really, a test key system being built is the only way really to find out.

LOL - it really would be some kinda keyboard wouldn't it.
Set the action on each key individually - LOL!

Peace
Ausser.

PS:
I wonder would there be a way to set up some kind of renewable energy source from such an electromagnetical configuration - ie generating electricity someway by playing the keyboard....
Of Course, you'd get more milage out of some pieces than others I expect.
A sort of 'Pay as you Play' arrangement....

NAAM 2010 " The GREENEST KEYBAORD ON EARTH HAS ARRIVED!!!"

Now THAT!!! certainly would sell!!!

I'd want to buy it - just to test the power output of certain classical pieces.

[John Hendry]

Dfahrner said: “I wonder why we've never seen it implemented?”

That is exactly what the owner of Hood River Hobbies said when I showed him how I used simple electronics to eliminate adverse toque reaction in a single or double rotor system system on a helicopter. He had an experience when he saw the reality of observer time in E=mc2 made simple and applied in understanding to reveal the symmetry of the simple solution that obviously required w from the equation applied in t and I promises this will make you laugh when you see it and remember me saying this. Nothing is more obvious.

This is blindness you are observing. Complete and utter blindness caused by symmetry and precession. Apply a force to a spinning mass and it moves at a 90 degree angle. In a world of symmetry only time can tell you which way is left and right. Time has had science confused for some space…sounds funny doesn’t it, but it’s says the same thing….and making it “relative” only helped build bombs quicker, not measure their source size. If you want to know why space is relative to time spend 20 years in college and write a 50 page essay that ends with a question explaining it, but don’t stick you finger in clay at a steady rate and count because the source of the hole made is too confusing to most experts. We can count something’s parts with the observer included and this gives us center eyesight and you may be taking for granted that you see this solution so easily as it requires two separate energy compartments, two separate inertial frames that add up to one sum, just like the equation. After a year of rambling you might have picked up a little rumbling. I am sure many will not see this simple solution so quickly as you have. There was absolutely no reason to have not applied its class C application to lower costs and resolve the Polyphonic after touch issue 2 decades ago. I can and do say the same exact thing about eliminating the tail rotor on a helicopter. There was absolutely no reason to have not applied class C outside of over looking it.

John^^

[John Hendry]

Timo said:

"I getting the gist that you wanted to use electromagnets as a repelling force to push the keys upwards giving them (customisable) "resistance", instead of springs? Along with another electromagnetic circuit to detect how far close (ie. pushed down) a key is at any time, along with a foam pad to allow you to push down further for "aftertouch"?"

No way, this is the wonderful magic key here....no second circuit or switch is needed to detect key location because the system reads itself giving key location, speed, and direction in real time without a second information circuit. It uses the EM spring circuit itself to tell key location and direction in time. So very cost effective especially when you take much needed poly AT into account as well as release velocity thrown in for free in a class C system that may cost less to build than non-poly AT system now used even with processor as this is caveman simple data to process and program compared to what we already do. The more I learn about our current keyboard trigger tech the more I see the problem creating “tail rotors” in every application to avoid a simpler solution.

We may be very surprised by the difference in control an EM system offers. I will wait to learn more before commenting on current system, but it looks sad from here….poly AT is not the only thing missing from it.

John^^

[John Hendry]

Using today’s applied technology the main disadvantage of solenoids, especially linear solenoids, is that they are inductive devices which will convert some of the electrical current used into heat. The longer the power is applied to a solenoid coil, the hotter the coil will become. As the coil heats up, its electrical resistance changes. With a continuous voltage input applied to the coil the solenoid’s coil does not have the opportunity to cool down as on a car starter because the input current is always flowing. In general practice to reduce this self generated heat it is necessary to reduce either the amount of time the coil is energized or reduce the amount of current flowing through it. In a class C system heat sinks allowing air to cool small EMs, perhaps foil made ones known to be efficient, will resolve this as the EMs are only being used to register key location and movement and can be made as small as necessary as they have no real work load.

The second issue of concern with inductive devices as some of you also pointed out is some form of electrical shield is required to prevent high back emf voltages from affecting the other emf sensitive components. Again in class C application this is not a problem with small EMs using today’s applied technology. I am a long ways from an electrical engineer and humbled by their working knowledge researching this. Most of the issues involved however go back to the old days and are well known and looking at them briefly with courteous but curious layman eyes open for change class C looks to me like it could be done immediately and should be to reduce costs and add poly AT and release velocity while creating a better system.

As some of you may know Open Sound Control (OSC) is a proposed replacement for MIDI, and is designed for online networking. Unlike MIDI, OSC allows thousands of synthesizers or computers to share music performance data over the Internet in real time. This is the time to implement such changes: http://opensoundcontrol.org/introduction-osc
“There are dozens of implementations of OSC, including real-time sound and media processing environments, web interactivity tools, software synthesizers, a large variety programming languages, and hardware devices for sensor measurement”.

Because the time signature of electromagnetism and gravity within the weak force mass oscillation cycles are not separated yet in electronics science has had its hands tied behind its back observing Nature. It is my hope this idea will open the door to let my simple math conversations and equations that support a Great mans work and natural insight be understood within the math of electronics and applied out in the open without politics and religion standing in the way. This “invention” may very well allow this to happen almost by accident. If you look at the EM> solenoid device, it includes the observer by function and looks a lot like an equation I know of;-).

As far as class A, A/B systems go, it too can be done but using current awareness of weak force to apply IE: N/A, it will require a cooling system and I see hollow alum tubes with cold air or fluid cooling using a radiator the solution. Air may work fine from brief study. This will add weight but worth every pound and key weight with hammers, real wood, etc, can be cut down as system will create this sensation and control key oscillation speed and response. Very stiff light weight keys will be wanted in class A systems, not hammer actions than can be sampled and copied. With advances in EM> tech can see cost very low in the future.

I would build it using current cave tech in 3 pieces to resolve back emf and weight by separating keyboard from circuitry module and radiator that could be attached from below. Estimated weight: 12lbs circuit module, 58 lbs 76 keyboard, weight of alum radiator depends on fluid capacity but just air cooled can be quite reasonable and connection would be easy if designed right. Not just for studio use, 5 min setup max. It will not produce heat like a car engine, but look how light alum/plastic radiators are and even a light weight steel radiator is light enough to use as fluid is main weight. I would use a thermo sensing fan for voltage stability. If you have ever water cooled your computer’s cpu you know this system can be done practically using current tech. I see no reason the EMs cannot be made for $25 a piece in quantity looking at computer CPU water cooling part prices like $35 XIGMATEK AIO-S80DP All-in-one liquid cooling system for cpus. I see a possible $4,500 retail cost on this, maybe even for 88 keys as alum tubing is not that expensive and EMs are part of key system. I would think a hammer action key to cost to be the same so I know little of real world manufacturing costs but with no other key switches needed for Poly AT and RV I think class A is very possible to do right now and once felt and tried out no going back ever. Its getting better all the time….

http://www.newegg.com/Product/Product.a ... 6835233011

John^^