You are probably aware that each string is actually 3 wires spaced about 0.15 inches apart. Each hammer actually hits 3 wires simultaneously. So you'll need a field that is fairly uniform over a width of about 0.3 to 0.4 inches, for each note.

I can't say that I am familar with piano wire or piano wire installation.

A question for you though: Does the installation of piano wire allow you to connect a lead to each of the wires (one at each end) overwhich an electrical current could be introduced?

If this is true, then perhaps a magnetic field could be set up in individual wires to the strength necessary to quickly move it towards a small electro-magnet placed over the wires.

This could be an interesting project to pursue.

Not in a piano where it has a metal frame - like most of the modern ones do.

Perhaps it's a difference in terminology. In my piano a small felt-headed hammer strikes each set of strings. I can see this action when I lift the piano lid and press a key.

The dampers are harder to see, being typically below/behind the hammer mechanism. There are variations depending on grand/upright/spinet piano construction, in particular for the piano (soft) and sostenuto pedal actions. The damper pedal is easy; it just lifts them all.

Hello everyone - thank you for all for responding so quickly to my query. I have no knowledge in these areas (I'm a composer), but I can sort of follow along with what everyone is saying. If I can ask a few follow up questions... 1) How to I create/buy an excitation oscillator? 2) Where can I go to learn about controlling AC, in case I want to build something that matches the frequencies of the different notes? Piano frequencies are easy enough to calculate that even I can do it, and being able to control the current to each string would let me create uniform attack/sustain across the keyboard. While I'm sure even my questions are full of errors, I appreciate getting help from experts! I think what I really need is an eager grad student to give me hands on advice...

It might be valuable for you to hook up with a piano tuner/reconditioner, who can give pointers on how the actions work and where to have the electromagnets interact with the strings. Solar Eagle's intervening post (GA) looks worth following up, too.

You may want to start by getting an Arduino and an audio shield. A shield is another board that plugs on top of the Arduino.

The Arduino is easy easy easy to write code for, and it is very versitile. It will make what ever you do to drive coils over wires or motors or whatever easy to setup and test.

I am currently working on a "coil" project that will use an Arduino and Audio shield to drive the coil and the arduino has made the project more about building the coil than controlling it. I don't have to sweat the control part!

If not hitting the strings, would this be a single excitation or continuous? If continuous, would you not have to excite each one at it's natural frequency? Why not use a programmable electronic keyboard? Or am I missing something.

I salute your ambition, but wonder what you will gain if you are only able to excite 10 strings at a time.

The response you will attian with electromagnets will not be as crisp as a hammer strike and you will find it challenging to control the pitch/sustain electronically with the foot pedals.

Not impossible, but a challenge.

It seems to me you won't have a piano any more.

Wouldn't it be easier to borrow a Claviona <usual disclaimer>?

Maybe a look at this gentleman's work may help.

http://en.wikipedia.org/wiki/Robert_Moog

I haven't seen this done with a piano but i have with a violin. Rather than exciting the strings in the manner you are suggesting each of the violin strings had a small rotating wheel above it which would push down onto the string. Have a look at the Paul Corin museum site for details. Corin runs a mechanical music museum.

Someone has already pointed out that the piano strings are in groups of 1, 2 and 3 dependant on pitch - the groups of 3 being the higher frequency.

Is this what you are trying to do:

http://music.ece.drexel.edu/research/mrp

I got a picture of a system where in the anchors for the strings have an 'excitable' component. Imagine a solenoid attached at right angles to the point(s) where the strings are attached. The strings would not get 'touched' but the end(s) of the string would 'vibrate'. I don't think the 'solenoid' would have to be triggered but one time to create the desired vibration on the string. The strength of impulse (power) to the solenoid could be regulated according to a computerized program &/or stress sensors in the piano keys themselves. String dampers would operate as normally. This would sidestep the issue of creating a new technology specifically for affecting piano strings (in a manner commentary seems to be examining). Carlos, My Dad's son.

Again, thanks to everyone for posting. There are lots of great ideas. Both the links posted by SolarEagle and silverfox are close to what I have been imagining. It's very helpful, so thanks for sharing! I don't mind not being the first to have this idea :)

As for the doubters, of course similar sounds can be synthesized, but most pianists and composers will always want the real thing. Producing these beautiful sounds from an acoustic piano has a certain "something" that a traditional synth just won't be able to replicate. Check out the links and you'll see what I mean!

I'm meeting with something today who has some solutions as well. I will keep this thread posted with developments!

If you had a coil and an amplifier circuit for each set of strings (88 in all) , you might be able to set up positive feedback if the strings are steel. (I think they are, check to see with a magnet.) The circuit would favor whatever harmonics had a anti-node (maximum movement) at the position of the coil. I would assume you would put the coil at the hammer point, to get the same timbre as the piano sound. The attack portion of the sound envelope may be more gradual than you want, but if it works, it should build up to a sustained level.

It might be an interesting project to try with one string set before you make any big investment.

The piano strings don't just vibrate freely - you have a damping action on it. Pressing the right pedal moves the felt pads away and the left pedal pushes the felts on harder.

If you play a piano with the right pedal pressed you will just end up with a mush of sound.

Getting a string to sound is an aggressive action with a hammer (it is a percussive instrument) so you might find the gentle vibrating action will be a massive disappointment.

The preceding discussion was interesting, and covered a fair territory.

BUT

It left out two fundamentals:

1,. A tone in an instrument is defined as unique by its attack signature. The sustained part is not at all different. A hammer in a piano, a plucking in a clavichord or harp, a guitar, the attack of the bow on a violin is largely determining the perceived sound. An electromagnetic excitation provides none of it.

2,. The tone produced by a triplet of strings. They need to be off of each other to provide loud and quality sound from your piano. When I was unhappy with my tuner (a nice guy any other way), I learned tuning. Now, I can tune up a piano from honky tonk, to iazz, to Bach recital. Fair, not nearly professional quality, not one tenth their speed. But can do on a yeoman level.

2.1,. On a technical level, every one of a tightly coupled triplet needs to be excited at its own, by coupling constantly modulated frequency to sound even approximately right. That is a filter electronic subspecialty.

3,. You do not need to believe me. Research it, then build a single triplet. And let us know your results.

best regards

Make the strings the coils.

In my opinion a good piano is one that is on fire. Being dropped from a high place can also help it.

Time to come off the fence with this piano non sense and start looking for an old Hammond B3. Now you can really do stuff with an old Hammond.

No other make though - make sure it has tone wheels.

Pianos - pah!

This is a positive note for our enterprising experimenter.

I would build the triplet on the edge of a 2x8 or 2x10 for stiffness, attaching a thin plywood as a sound board. The piano parts do not cost much.

I would mount the electromagnets on a slide. You may excite the strings at their fundamentals, 3rd, 5th, 7th etc harmonics that way.

Why not even harmonics? You tell us!!

Why does a piano crosses sets of strings? (think of nodes of resonances).

And a few more, that does not come to mind right now.

You can cover a fair portion of classical physics with a few experiment that way. Go for it, as you cannot go wrong.!