The ultimate Grand Piano

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello Gentleman and Gentlewomen,

I am an audiophile and pianist, and am looking to build the ultimate digital piano in a grand piano shell. The source will be from high quality keyboard (Roland V Piano seamlessly integrated to appear as OEM). I am seeking advice and guidance on how to effectively representing an accurate reproduciton of a real piano. I do not wish to use conventional speakers, as the cross over point scares me with phasing issues, etc.

My budget is <$10,000, so ribbons are out... I do have access to a full shop and regard myself as talented with tools, so I’m up for just about anything.

I would like to hear your feedback on what type of speaker system you guys would recommend I construct to achieve maximum realism. On a side note, to fulfil my curiosity, I have a set of transducers on the way (Clark Synthesis Platinum) that I would like to attempt to mount to the existing soundboard. I would like to try this, as it almost makes sense to play a piano through a piano... you know? I have my reservations of this going well, but at least I will have a pair of transducers to play with if it doesn’t work out. Thoughts on this would be appreciated!

Would ESL’s work? I am not familiar with the level of tension of the diaphragm, and if a panel could successfully be laid flat without big complications arising from sagging. Also, would the volume levels of ESL’s high enough to match that of a real piano?

What other suggestions would you have? What would you guys do? I am hungry for your input!

Thanks in advance!

Robert
 
Wow Robert.....This is the craziest thing I have ever read. I dont really know where to begin, but I have a few questions.

1- Where will you be utilizing this instrument? Environment?
2 - How do you want the instrument to direct the sound? Omnidirectional, directional...?
3 - Is your goal to have a complete all in one unit?
4- Do you plan on having amplification within the unit?
5- Will there also be a kitchen sink installed?
6 - Will there be the typical grand piano top that you can open and close?

I'm also a pianist btw.

Spiro
 
A ten-grand-piano. Interesting concept.

To start you're going to need a really, really good keyboard controller if you want action even close to the real thing. Then there is the waveform synthesis.... that's gonna be another big chunk of change.

As to the speaker system. You have two issues that I see right away. First, the strings vibrate and bounce off the angled lid of the piano, but from different positions inside the shell. Second, the shell itself resonates from how the strings are mounted.

My first thought from all this is to mount a bunch of drivers directly to the baseboard of the shell using big screws and stand offs. This way the sound waves radiate both up and down, like they do from the strings. The drivers maybe should be different sizes just the way the strings are sized for each note and each powered one with it's own amplifier. And finally, an active multi-multi-way cross over between the amps and the synthesizer.

But that's just a guess. While Yamaha may have proprietary experience with this kind of thing, I really think you're in truly unexplored territory.

:)ensen.
 
" #3
user_offline.gif
aptquark
An externally hosted image should be here but it was not working when we last tested it.

diyAudio Member

http://www.diyaudio.com/forums/planars-exotics/members/aptquark.html

Join Date: Jun 2010
Location: MS
Blog Entries: 1


Wow Robert.....This is the craziest thing I have ever read. I dont really know where to begin, but I have a few questions.

1- Where will you be utilizing this instrument? Environment?
2 - How do you want the instrument to direct the sound? Omnidirectional, directional...?
3 - Is your goal to have a complete all in one unit?
4- Do you plan on having amplification within the unit?
5- Will there also be a kitchen sink installed?
6 - Will there be the typical grand piano top that you can open and close?

I'm also a pianist btw.

Spiro "





These are all very good questions that must be condisidered.

An ESL driver is without question the lowest distortion and uncolored driver in exsistence (IMO) as well as many magnetic ribbon and planar designs and only second to plasma technology as far as sound quality.

However they do have their own issues as far as size vs lowest reproduceable frequency and sound dispersion.

The drive electronics is a very big issue and although we here at DIY Audio have made some very large leaps and bounds and it is getting better and better and more less as expensive than it used to be.
However, It can be very costly and time consuming to get/find the right combinatoin of crossovers,amplifiers and especialy transformers to drive them properly.

All of these things must be considered along with, Is mounting the (any) driver in the piano case going to give you the kind of sound you are looking for?
As well as is might sound great in the nearfield,But how will it sound across to the other side of the room?

ESl's are well noted for there incredible detail and they don't fair well in an enclosure or having anything (even like the soft side of fiber glass insulation) near the proximity of the diagphram.
I have done some investigations on this matter.

What happens is that any reflections (no mater how small) that can be redirect back through the diagphram can cause cancelations at some frequency's, therefore, reducing the the high end detail that they are so noted for.

However this does not mean that a good compromise can not be had.
BUT, that is what it would be "a compromise" between the best and second best.
I know this as I have experimented with different ways attemping to turn the dipole esl driver into a monople driver.
It is not an easy tasc and all my attemps have turned out as "Oh it's okay" and not "Ahh that is just awesome".

Just as a suggestion, As I always try to inspire individuals to try building their own ESL's, is to maybe take a look at the magnetic planar technology.

You will still have the same issues as far as sound qaulity is concerned as listed above, But the drive requirements will be much easier and simpler than what an ESL requires.
Although good neo magnets are not cheap and are getting more expensive by the day.

I hope this gives you a little more insight as to what may be involved.
And I hope that any of this information does not discourage you in any way.
But too incourage you to see your project as a reality, as you never know, you might just come up with something totaly great as it happens quite a bit here in the DIY audio forums.

Good Luck! jer

P.S The very first time I ever heard a true sounding piano was through the Amazing Carver ribbons and then the Apogee Duetta's (which I still have) and now my very own DIY ESL's top the both of them,Cheers!
 
Last edited:
As to answer your other two questions sagging would not be an issue with an esl.

And with a panel area of at least 1 square foot or more the loudness would be more than enough when properly driven.

The problem is the beaming and the unequal dispersion of different frequency's are the issues that go along with most all planar drivers. jer
 
Hi Robert,

whichever transducer you use, you'll be faced with the problem that the keyboard controller is designed to reproduce the sound of the piano as a whole; if you put out this sound through transducers inside a piano shell, the shell will influence the sound "a second time", so to speak. Therefore I don't know if the end result will be what you'd expect.

Having said that, my feeling is that the best way to do this is to have many different transducers, with the bass notes being reproduced at the left side and the trebles at the right side, just as in the real thing. Using ESLs this is easily realized, especially if you would use wire ESLs with electrical sectioning.

Finally, depending on what your goals are, I'm not sure whether an ESL would produce sufficient sound pressure in the very low register...

Kenneth
 
Creating a foolable, exact reproduction is probably impossible, but I hope to get wowingly close...

The good news, is the controller used is a good one. The Roland V Piano is one of the best today. I paid just under $7000 for it, and it is specifically a piano (no other sounds). They action of the keys are very, very real, as it uses real keys and action geometry. It feels par to my Steinway.

As for how a piano works in its creation and dispersion of sound... The strings are tensioned by a "harp" and underneath is a large soundboard with a bridge, bridging the strings to the soundboard. The fundamental frequencies are most efficient in transferring to the soundboard, as they effectively at the terminated ends of the string. The harmonics are largely reproduced by the string, and fundamentals by the soundboard. There is no "bottom" cover on the piano, leaving the curved frame only. Sound radiates from the soundboard down and reflects off the floor, where the same occurs above with the added harmonics of the strings. The lid is typically open when played (at 45 degrees) and it will be used in a home environment (room = 28 x 22 x 10').

The challenge is the fundamental frequencies range between 27.5 Hz (A0) to 4186 Hz (C8). The harmonics of the top note (4186 Hz) extend to 18 kHz.

If I do have to add a crossover (& low end), I would like it to be as low as possible (obviously) as to maximize phasing alignment.

As for size and configuration, the grand piano shell is very heavy and sturdy. The soundboard is attached to it, contributing the vibration of the shell to the color of the instrument. I foresee these being natural piano resonances and hope that it does not negatively effect the tone. The surface area is rather large that I have to work with, and that is to my advantage.

I will address more of the issues addressed in your comments shortly...

Thanks for the response to this!
 
Hi Robert,

whichever transducer you use, you'll be faced with the problem that the keyboard controller is designed to reproduce the sound of the piano as a whole; if you put out this sound through transducers inside a piano shell, the shell will influence the sound "a second time", so to speak. Therefore I don't know if the end result will be what you'd expect.

Having said that, my feeling is that the best way to do this is to have many different transducers, with the bass notes being reproduced at the left side and the trebles at the right side, just as in the real thing. Using ESLs this is easily realized, especially if you would use wire ESLs with electrical sectioning.

Finally, depending on what your goals are, I'm not sure whether an ESL would produce sufficient sound pressure in the very low register...

Kenneth

I also don't think ESL has enough dynamics for the whole 7 octaves. Standard drivers would be able to reproduce the really big crescendos. One issue is what to do with the harmonics. When striking a bottom G#, you get harmonics all the way up. Should there then be tweeters interspersed with the woofers at the left side or is it enough to utilize the drivers as in any multi-way speaker?

As to synthesis, while it might be best to produce the waveforms from scratch, it might also be sufficient for testing to output plain old plucked strings (with some added sustain). The sonic effects of the pedals is a whole other problem...

:)ensen.
 
Wow Robert.....This is the craziest thing I have ever read. I dont really know where to begin, but I have a few questions.

1- Where will you be utilizing this instrument? Environment? Home (Large front Room)
2 - How do you want the instrument to direct the sound? Omnidirectional, directional...? Much like a real grand piano (with compromise) - out the bottom and top of the cabinet. Dipole Omni directional could work
3 - Is your goal to have a complete all in one unit? Absolutely! :)
4- Do you plan on having amplification within the unit? Yes
5- Will there also be a kitchen sink installed? If it adds positively to the tonal coloring...
6 - Will there be the typical grand piano top that you can open and close? Yes. The top lid will be open when played, always


Here is an article by a gentleman that has performed the Clark transducer/soundboard assembly: The Hybrid Piano – Part 1 - NCF Music! I have spoken with him directly and he uses the instrument in his church every sunday. So it can't be that bad, right? But I have a feeling it could be done better...

Here is a youtube vid of a gentleman that has done a keyboard integration into an upright piano. I would perform this without all the "bling" and an improved sound system: YouTube - My Extreme Piano

I do have access to signal processing... EQ, Active X-Overs, time alignment and notch filtering. I will use it sparingly and only if needed, but it's there.

The piano itself is a Boston measuring in at just under 7 feet. This leaves about 6' x 4' with the curvature of the shell cutting into the square footage. I estimate I could construct a panel(s) with a total of 18 square feet (or even 36 if doubled up. This helps when attempting to get to the lowest note effectively. The frequency response of the fundamentals are 27.5 Hz (A0) to 4186 Hz (C8). The harmonics of the top note reach upwards of 18kHz. Hmmm

If I do have to utilize a crossover (and it looks like I would have to), I would like that point to be as low as possible (200Hz or below would be ideal). Hmmm...
 
...As to synthesis, while it might be best to produce the waveforms from scratch, it might also be sufficient for testing to output plain old plucked strings (with some added sustain). The sonic effects of the pedals is a whole other problem...

:)ensen.

The digital piano allows for half-pedaling and incorporates the sonic effects of the pedal quite well.

There is a sampling company by the name of PianoTeq that has some amazing samples available. There is a possibility of reproducing the strings and modeling the soundboard separately, but this starts to get pretty complicated. How many engineers can I hire for my $10,000... (mind you, with this economy, probably a few). I will look at going this route if I don't get satisfactory results from a single sample source.
 
I agree, As I have stated many times in other threads that I am quite content and happy with the sound of my little ESL panel (as I listen to it nearfield).

And I have been listening to it as a mono source since the day I got it working again for the very reason that "Does it sound the same as (this or that) instrument in a natural form and enviornment" and I have yet to hook the dang thing up in a stereo format.

Yes, it sounds natural and very natural at that.

But, It is not enough to be sounding natural enough when far away as a real instrument would sound.
And when it does it takes alot of power (alot of power)!
I mean electrical power.

Due to the surface area and the shape of the surface area radiating the sound into the listening space (free space).

And that is why a planar or any other type of physical driver from a single point or even a line source persay where its directionality is in one direction, can not create this kind of natural ambience.

I have pulled all kind's of tricks in make believing that the sound is over here or over there using delay's and reverb's with different special eq setupup's in speakers and in headphones.

This is nothing special in the recording industery.
Although it seems quite amazing.


jer
 
A digital piano mounted inside an actual full size grand piano with the guts ripped out... :confused:

Try as I might, I just don't get it.
Why, when you have the space and a real Steinway grand piano?
As someone who plays piano I don't see the point of all this.

Makes as much sense to me as pulling the engine and drive train out of a Ferrari, replacing them with all non-Ferrari parts, then trying to get it to perform and sound as close to the real thing as possible.

Each to their own I suppose...:rolleyes:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.