I can understand your confusion.
The engine of the piano is about 90% of the value, and it's shot. So instead of trashing the body, why not drop in a hybrid and give it additional life? Although I already own a perfectly wonderful piano, the advantages to a digital piano are of great value to me (recording, composing, plugging headphones in when the family sleeps). So instead of having an ugly digital piano sitting in my living room, why not incorporate it into the beautiful Ferrari body that would otherwise end up in the scrap yard?
I just want to put lipstick on the pig, really. And I have never found a speaker system on a digital piano that sounded good. They are all very poor, so I want to do it better... and do it with some style/class.
Digital pianos aren't ugly. Form follows function. What you're talking about is creating a façade. Modern architecture has moved away from such.
A piano isn't just a harp. It's a soundboard. You could try one (or more) of those speaker drivers that are used to create hidden systems by exciting wall panels or pieces of furniture.
I'd just stick with the digital piano though, and maybe bolt on a really nice speaker, but good luck with the project, all the same.
w
Ok...here is what I have so far, keep all amplification out of the shell and ONLY in the bench seat. Dont ask me why, but that is what I'm going with right now.
What type of amplification are you interested in? Tubes, class D, class A? Heat may be an issue depending on your choice. Still, a low watt set amp with high efficiency drivers would compliment the digital Roland sound. Same for class A. Class D I would think would work good with a less efficient speaker choice. DOnt focus on speaker choices right now. Lets talk about amplification and how that would work with the Roland.
What type of amplification are you interested in? Tubes, class D, class A? Heat may be an issue depending on your choice. Still, a low watt set amp with high efficiency drivers would compliment the digital Roland sound. Same for class A. Class D I would think would work good with a less efficient speaker choice. DOnt focus on speaker choices right now. Lets talk about amplification and how that would work with the Roland.
OK, that makes more sense.
I'll be interested to see what you end up with.
Best wishes for your project.
Cheers,
Alex
I can imagine puzzling together rectangular ESL panels inside a grand piano. You could solicit old ML panels from owners who've swapped them for new (if flat to 20kHz is not needed), or you could contact f.ex. Justrealmusic.com (ESL 'kit' builder) to supply the right mix of panel sizes (in curved or flat) and bias supply. The total ESL square footage would be enormous compared to the typical ESL stereo loudspeaker, so it should get pretty dang loud. And if bass is still lacking, you could fit in some free-air cone woofer subs...
Can we see pics of the donor piano shell?
Can we see pics of the donor piano shell?
Last edited:
@tylersphile
Attached PDF is about simulation of the sound
radiation characterstics of a piano.
Maybe you can exploit that article using google
translate.
The author uses Manger transducers in his experiments.
But i think this is not the only way to go.
---
Your requirement is to have the speaker(s) builtin
to the piano body ?
Will the sound board still be in place ?
Kind Regards
Attached PDF is about simulation of the sound
radiation characterstics of a piano.
Maybe you can exploit that article using google
translate.
The author uses Manger transducers in his experiments.
But i think this is not the only way to go.
---
Your requirement is to have the speaker(s) builtin
to the piano body ?
Will the sound board still be in place ?
Kind Regards
Last edited:
Exiting the soundboard with the synthesizer's output
signal, will introduce all the modes (resonances, formants)
of the physical sound board.
But since a modern digital instrument already simulates
the soundboard modes/formants i doubt this leading
to good or "realistic" results.
It may be a "quick" and interesting option for the first.
Jer, I'm wondering if a smallish panel like you describe, used near-field, would be good as a recording monitor. As you know from our other discussion, I'm particularly interested in piano sound at this time, but I also will need a good monitor that reproduces other solo instruments as well. Apparently these things have dispersion issues, but in most cases, as a monitor they would be used from a fairly stationary position in front of a keyboard. Maybe a small woofer driver could be biamped and actively crossed over to get enough extension.. How low can a one-foot panel go anyway? If the physics is the same as open baffles, it wouldn't be nearly low enough by itself. Do you think a crossover at 300 Hz or so would damage the sound quality of the system for monitor use? Seems like that would be a bad place to have a cross, but maybe a high Q active cross would be transparent. What do you think? I'm still curious about what application you are using your single panel monaurally. HiFi listening or this kind of thing?
Would this method make an array of narrow ESL panels' radiation pattern more string-like, like a real piano (and with a nod to the Quad ESL-63 rings, but using lines instead: Install 5 (or 6, depending on the donor size) 9-inch wide ESL panels next to each other, perpendicularly to the keyboard, feed one signal to the center panel, then use a digital xo to send a delayed signal to the adjacent-to-center two panels, and a third signal further delayed to the outermost 2 (or 3) panels.
(Then crossed over at ~200Hz to one or two 12inch free-air woofers?)
(Then crossed over at ~200Hz to one or two 12inch free-air woofers?)
Last edited:
Oh very much so!
You are on the right track.
I plan on using them as nearfield recording monitors in front of my mixer.
As the whole idea is to have a speaker that has absolutley no coloration.
If one where to ask me what do they sound like I would have to say they don't sound like anything at all.
That is just the nature of the beast.
My little panels should be crossed over about 800hz to 1200hz technicaly speaking.
But at a low level and only 18" away I run them fullrange and get a very reasonable output so that any more than that ear fatiuge sets any very quickley.
Only because of my many years of high volume rock-n-roll giutar playing.
But my goal is to cross them over in the 200hz to 300hz region as to retain the body of the low registers of the vocals.
That 8" sub that you saw in the picture is good till 1200hz and the crossover for that setup was some where in between 350hz and 700hz by my measurements.
And believe it or not they melded together very nicely.
I can't tell you exactly because it was cobbled together and I eventualy biamped the thing and used the eq's on my mixer as a crosssover so I haven't built a proper crossover yet.
I basicaly just just took out everything below 80hz that I could and added a little bit of around 300hz to 500hz to make up the difference too bring out the low end due to the back wave cancelations and ran the rest flat.
As far as bass is concerned in nearfield they certainley hold there own on stuff like the spin doctors and that snare OMG!
Even with no eq.
As well as that awesome vocal from spandel balette's "true" and then on to the delicate sound of crowed house's " don't dream its over" after the intro the at bass run just punches right through and stands on its own with it own independence without getting muddied and buried in the mix.
And the same with david bowe's "golden years" you can literaly feel the depth of the reverb, even in mono, the detail is just that great!
The moody blues,boston and as I already mentioned Kansas and the list goes on.
It is those impressions as to why I have become lazy to hook up the other channel. Besides the fact that I was looking to hear any flaws or descrepincy's from the driver itself.
And I didn't want to become distracted from having the other channel that might have been mismatched from the one that I was trying to judge.
I have found that both of the panels are an exact match to each other as well as the other one that uses just spray paint as the stator coating.
Only that it doesn't with stand as high of a bias voltage as the powder coated ones do.
But this is an extreme limit.
And is almost irrelevent as they do have there limits aswell and this is because of the voltage breakdown of the air itself.
I did lose one of those painted ones some where while I was out one night showing it to some friend and forgot where I had left it (like a dumbass).
Passive cross overs are a tremendous PIA to figure out for an esl due to it being completely reactive and the combined reactance of the transformer and transformation ratio however it can be done.
As the frequency raise's the impedence lowers.
The THD of a an esl is generaly that of the electronics driving them.
And usally < .1% thd and typicaly < 0.01% or less across the entire audio bandwidth.
I would like to hear common box speaker do that with no resonance's typicaly found in boxed systems. jer
You are on the right track.
I plan on using them as nearfield recording monitors in front of my mixer.
As the whole idea is to have a speaker that has absolutley no coloration.
If one where to ask me what do they sound like I would have to say they don't sound like anything at all.
That is just the nature of the beast.
My little panels should be crossed over about 800hz to 1200hz technicaly speaking.
But at a low level and only 18" away I run them fullrange and get a very reasonable output so that any more than that ear fatiuge sets any very quickley.
Only because of my many years of high volume rock-n-roll giutar playing.
But my goal is to cross them over in the 200hz to 300hz region as to retain the body of the low registers of the vocals.
That 8" sub that you saw in the picture is good till 1200hz and the crossover for that setup was some where in between 350hz and 700hz by my measurements.
And believe it or not they melded together very nicely.
I can't tell you exactly because it was cobbled together and I eventualy biamped the thing and used the eq's on my mixer as a crosssover so I haven't built a proper crossover yet.
I basicaly just just took out everything below 80hz that I could and added a little bit of around 300hz to 500hz to make up the difference too bring out the low end due to the back wave cancelations and ran the rest flat.
As far as bass is concerned in nearfield they certainley hold there own on stuff like the spin doctors and that snare OMG!
Even with no eq.
As well as that awesome vocal from spandel balette's "true" and then on to the delicate sound of crowed house's " don't dream its over" after the intro the at bass run just punches right through and stands on its own with it own independence without getting muddied and buried in the mix.
And the same with david bowe's "golden years" you can literaly feel the depth of the reverb, even in mono, the detail is just that great!
The moody blues,boston and as I already mentioned Kansas and the list goes on.
It is those impressions as to why I have become lazy to hook up the other channel. Besides the fact that I was looking to hear any flaws or descrepincy's from the driver itself.
And I didn't want to become distracted from having the other channel that might have been mismatched from the one that I was trying to judge.
I have found that both of the panels are an exact match to each other as well as the other one that uses just spray paint as the stator coating.
Only that it doesn't with stand as high of a bias voltage as the powder coated ones do.
But this is an extreme limit.
And is almost irrelevent as they do have there limits aswell and this is because of the voltage breakdown of the air itself.
I did lose one of those painted ones some where while I was out one night showing it to some friend and forgot where I had left it (like a dumbass).
Passive cross overs are a tremendous PIA to figure out for an esl due to it being completely reactive and the combined reactance of the transformer and transformation ratio however it can be done.
As the frequency raise's the impedence lowers.
The THD of a an esl is generaly that of the electronics driving them.
And usally < .1% thd and typicaly < 0.01% or less across the entire audio bandwidth.
I would like to hear common box speaker do that with no resonance's typicaly found in boxed systems. jer
Question: Is it difficult to end up with a well matched pair when you are making everything from scratch like that?
I just wouldn't want to do it. Not with gadgets like MiniDSP available for active xo and 6 channel PEQ to boot. Even pretty decent amps can be had cheap. Reminds me: What kind of spl efficiency do these things have? Are they friendly to low watt amps like SET, Tripath, chip, etc?
That is pretty amazing. However, a lot of people seem to dig the sound of speakers with some harmonic distortion, especially even harmonics. Has anyone ever accused these of sounding dry or anything like that? For monitor application, it's only a good thing. Box-originated resonances, of course, have no socially redeeming value at all, and one reason I'm building some OB full rangers (high SPL) (bass augmented) (active xo).
To Tylersphile: I didn't mean to go off topic there. Actually, I'm very interested in the topic myself. I get what you're doing. At the risk of sounding too simplistic, it sounds like you've found yourself with an extremely nice empty "cabinet" and you want to do it justice. Admirable, as is the willingness to sail into uncharted waters. I respect a big challenge.
Last edited:
I had another whole page of stuff written and it raining and the power glitched and lost it all!
so will answer quickly and we should consider continuing this dicussion on the other thread we started for respect to the rest of the diyer's here.
It is very easy to make matched panels by using the same dimensions and following the same cronstruction procedures for every panel.
I don't have an SPL meter anymore.
But provided the bias voltage I am using and the very high transformation ratio of 1:200 to 1:400 I would have to say that it is fairly high.
As a midrang/tweeter they can take everything That my crown DC300A can muster up.
And can cause pain although I am quite immune to such levels .
But the average person would not be able to stand the loudness even as far away as 10 or 20 feet let alone nearfield.
They way the are setup up they should only purpose an 8 ohm load.
But they are actualy around 1 ohm at 20khz because of the capacitance of the transformer is about 6 to 8 times that of the panel.
This will be corrected when I do a rewind of the transformer.
I started out using my cheap AIWA 80 watt amp and used it with good sucsess at a fair listening level.
But because of the very low impedence at the high frequency's it would shut down from self protection quite often.
And has even brought that crown to its knees due to heat.
Because of the low impedence.
But this is a small problem that can be fixed easily. jer
so will answer quickly and we should consider continuing this dicussion on the other thread we started for respect to the rest of the diyer's here.
It is very easy to make matched panels by using the same dimensions and following the same cronstruction procedures for every panel.
I don't have an SPL meter anymore.
But provided the bias voltage I am using and the very high transformation ratio of 1:200 to 1:400 I would have to say that it is fairly high.
As a midrang/tweeter they can take everything That my crown DC300A can muster up.
And can cause pain although I am quite immune to such levels .
But the average person would not be able to stand the loudness even as far away as 10 or 20 feet let alone nearfield.
They way the are setup up they should only purpose an 8 ohm load.
But they are actualy around 1 ohm at 20khz because of the capacitance of the transformer is about 6 to 8 times that of the panel.
This will be corrected when I do a rewind of the transformer.
I started out using my cheap AIWA 80 watt amp and used it with good sucsess at a fair listening level.
But because of the very low impedence at the high frequency's it would shut down from self protection quite often.
And has even brought that crown to its knees due to heat.
Because of the low impedence.
But this is a small problem that can be fixed easily. jer
The sound radiation is predominantly from the sound board, which is excited by the string's vibrations.
The direct radiation component originating from the strings is fairly low.
THE ULTIMATE GRAND PIANO!
Might I make a suggestion, Why not build a full range planar using neos, and make it the same size and shape of the soundboard of the grand piano. You wouldn't need any crossovers or transformers.As it radiates from front and back it would be like the original soundboard.
Might I make a suggestion, Why not build a full range planar using neos, and make it the same size and shape of the soundboard of the grand piano. You wouldn't need any crossovers or transformers.As it radiates from front and back it would be like the original soundboard.
ULTIMATE GRAND PIANO
Tylersphile,
If you could send me the dimensions of the soundboard cut out and the surrounds, I could if your interested build a full range planar of that size, using neo magnets. I have been building these speakers for the last 20 years. And I have 13 of these speakers in the living room at the moment. All plugged in to my ONKYO receiver. They sound great even with mp3 cards.
Tylersphile,
If you could send me the dimensions of the soundboard cut out and the surrounds, I could if your interested build a full range planar of that size, using neo magnets. I have been building these speakers for the last 20 years. And I have 13 of these speakers in the living room at the moment. All plugged in to my ONKYO receiver. They sound great even with mp3 cards.
I am in the process of building a very small baby grand, using 3 diaphragms using 1,560, 25 x 6 x 3 mm neos. It,s about 41" x 31" at the moment. When I have finished it, I will let you know how it sounds, using the game setting L C R to drive the 3 diaphragms individually. Using the separate tonal settings on my Amp to help with the separate sounds of the diaphragms, integrating them separately.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.