Upright piano speaker simulator addon idea

RIP The biggest problem which I encourage you to study is the attach-detach problem.

I played with it more last night and I think there isn't a good way to get the sound to come close to the timbre of another real piano. I might keep trying with a hybrid approach of soundboard + speakers near the hammers. I feel that the big players (Yamaha, etc) would have already done something like this if it were simple, but they probably wouldn't do such user unfriendly things like manual attach/detach methods, or 10 transducers, or something crazy that I might personally put up with.

If the soundboard thing works (and I'm not that hopeful), I'd probably just design a lever mechanism attached to the bottom of the piano structure where I can shift a lever and couple the transducer to the soundboard. It's not ideal, but so far I haven't heard much difference in sound quality when I just hold the transducer on with different pressures. I'm using a mic stand and a shockmount, which is basically a spring. Don't know how this added spring/mechanism will affect the sound.

Last night I also just put the transducer on a cardboard box and laid it on top of the piano. The frequency range was better, so I still may continue with this scheme of tuning a large speaker (2x4 feet) and placing on top of the piano lid. Still experimenting with whether or not the soundstage is similar to the real thing.

Frédéric L Do you want to compete with Resonance : https://resonancepiano.com/en/ ?

Interesting. Not trying to compete. I don't have any interest in making money from this, just an interest in trying to DIY a system that is really good for personal satisfaction. Thanks for the link, I may learn something…

Why don't you just buy a pink foam insulation sheet at the local hardware store, like that used in the Newer Hampshire TechIngredients videos?

That cheap foam will be easy to mount and move around transducers and you can ignore any foam damage.

I seem to recall the TechIngredients video illustrated the basic impact of moving the transducers around and adding more transducers with sand. That might provide some basic transducer guidelines for this application.

You can cut the foam roughly shaped to your piano using a saw or maybe a heated metal string. Then mount the foam behind the piano. If you were using a M&H grand piano, for example, rather than hang the foam from the bottom of the piano just cut some wood straps for legs like a table; the legs may have some audio impact (positive and negative) but are good enough for testing and you could move them around.

Take my comment (like all my other comments...) with a large grain of salt, because I really have no idea about soundboards and transducers, but in my naivete I would assume that the wood soundboard will have a large filtering effect upon the higher frequencies (which are easily dampened because a large piece of wood will not be the best medium to propagate high frequency vibrations). The piano sound (recorded by microphones) is already the result of that filtering. If that filtered sound is "injected" again in a soundboard it will go through a second cycle of filtering... My point being that probably the signal to input to the transducers would need to be enhanced in the high frequencies in order to the final result to be realistic. Maybe Pianoteq is in fact the best option for a transducer system because the higher harmonics can be enhanced. Is this the reason why hybrid systems have both speakers (for mid-high frequencies) and transducers (for good spatial spread)? Does this make sense?

n-player Why don't you just buy a pink foam insulation sheet at the local hardware store,

That is essentially what I was planning to do originally - try out different materials like this. But when I just put the transducer on top of a cardboard box without adhesive, I realized that I could easily try it on the soundboard for fun. I haven't experimented too much with it, but IF the soundboard could project the sound, this would be the ideal solution because the sound would project up and down at the same time. Using additional substrates like foam on top, around, behind, underneath the piano will, I think, give a different image. For example, putting the panel above the lid will allow sound to project up like a real piano, but not beneath to the floor as the piano is in the way. Also, the panel will also project downward and reflect off of the lid, which may cause some issues. Again, I'm just theorizing here.

BTW, good to see you here n-player.

@vagfim You're right, I was just intending to EQ the hell out of the signal to see if I could get anything near close. But I think the additional substrate, like foam, will be easier to control and tune. I dont have access to every part of the soundboard because of the piano's structure underneath, but it might be fun to try anyway to see where it leads. Too many projects to fiddle with…

I also speak from a layman’s perspective, but I believe Steingraeber has been working on a ‘hybrid’ using only a full soundboard, and they have chosen Pianoteq (perhaps for the reasons mentioned by vagfilm); but still, the good news is that some piano makers (not Steinway) are working on these type of instruments and hopefully it’ll just be a matter of time before we see a hybrid with a full soundboard and no speakers whatsoever.

We’ve been waiting on Kawai to do this, but maybe they will be beaten to it by Steingraeber or some other piano maker.

P.S.

I know that some believe that a ‘soundboard’ for a digital is a gimmick, and that speakers alone can do the job, but I humbly disagree, once again, from a layman’s perspective.

Pete14 I know that some believe that a ‘soundboard’ for a digital is a gimmick, and that speakers alone can do the job, but I humbly disagree, once again, from a layman’s perspective.

I can neither agree nor disagree. I've heard virtual instruments played through various speakers that were good, bad, or in between. But I've never heard one played through a soundboard.

Yes, I've heard the Kawai CA- series pianos, which play through a combination of speakers and soundboard ... but the samples in those pianos are not top notch. And I did not notice any mode setting for "soundboard only".

So who's to say that a virtual instrument with only a soundboard will produce acceptable results?
And who would think it would be acceptable when driven by Pianoteq?

Pete14

We’ve been waiting on Kawai to do this

Kawai Aures is doing exactly that. No speakers, just a soundboard with some transducers:

Soundboard Speaker System with 4 transducers (Low-Mid Range × 2, High Range × 2)

Yes, that’s true, but I’m hoping for ‘pure digital’ with a soundboard and no speakers (NV-20).

The Trans/Aures comes attached to a full acoustic, strings and all, and I don’t want to pay for all that stuff. Now, I do understand how for those wanting to have the silent capability for late-night practice whilst also having a full acoustic for ‘the day’ something like the Trans-Aures would be appealing.

MHirsch Kawai Aures is doing exactly that.

Yes, you are correct: I was forgetting the Aures line... And for some reason I don't think that Kawai is using different samples in those transducers. So maybe a soundboard is truly a good medium for full-range transmission. Like I said, this was unsubstantiated bias...

So that was why I was trying to distinguish between the sound projection (unlike a cone speaker) and the timbre. I tried a Yamaha transacoustic once, and wasn't impressed by the timbre that the speaker put out, but did like the way it projected like a piano soundboard (as it was using the piano soundboard).

With fine tuning, I might be able to achieve something like this:

But still, I don't feel the sound is good enough. But possibly a "hybrid" using the piano soundboard and some separate tweeters and a subwoofer placed in the correct places. Maybe. Or just the foam substrates instead. Dunno yet.

Pete14

But Pete, what makes you think it’ll be a lot cheaper? NV10s, a pretty much acoustic instrument without strings and all, is already at the same price level as a K300 with an Aures system. To me, where’s no point in buying an almost acoustic piano but not really acoustic, and not really cheaper.

The only reason to even look at NV10s for me was the action - but its sound was let’s say less convincing than the sound of a K300 (no Aures) standing next to it. K300 is 3.3k € cheaper, by the way.

vagfilm

Well, I played a K500 Aures.

It’s acoustic sound, compared to the digital, was a whole lot better - natural, powerful but still mellow. Digital sounded like there’s a much bigger piano inside, trying to get out. I was not impressed like at all.

That’s been a year ago, things might have changed with Aures2, but there’ll be Aures3 - better, more natural and authentic, <marketing stuff>. And then what, an endless upgrade loop?

MHirsch Digital sounded like there’s a much bigger piano inside, trying to get out

It seems like a better job could be done with speaker placement/quality/calibration/etc.. with these ‘upright’ type hybrids, and yes, you are correct, they are overpriced.

Perhaps the prices won’t come down much, but at least I expect future upgrades to improve on the known weaknesses (starting with the speakers, etc.).

I have never been impressed by soundboards in digital pianos, the kawai NV5 sounds a little bit better than a Kawai NV10, but still, we are talking about a sound which is light years away from a real acoustic piano with vibrating strings. And using a better vst wouldn't improve things, the result you get will be always a boxy sound.

“Light years”… why the pessimism, Pianissimo?

We’re but a decade away from transformational improvements (the digital overtakes the acoustic).

These ‘string vibrations’ you talk about are way overrated, and if anything, easily replicated and simulated without the need for strings.

Yes, I agree, we’re still a little ‘boxed in’, but give us a decade and we will blow your mind and ears away (figuratively).

We have to admit that it is of essence for the acoustic piano industry to start moving out of the analog age and into the digital future, or they will become extinct.

“Adapt to your changing environment or disappear”, said Darwin, so you see, I’m not making things up.

We all know in the near future all will be wireless, virtual, and stringless, and there’s just no room for hand-made romanticism.

P.S.

Remember how just a few years ago a device like an iPhone was considered impossible?, “a computer in your pocket? You, sir, must be crazy”; well, here we are, and the iPhone14 is now a reality.

Pete14

I am not pessimistic, just realistic, I would be delighted to be surprised. But so far I think a nice big speaker will sound better without any soundboard help or any transducer.

dore_m So that was why I was trying to distinguish between the sound projection (unlike a cone speaker) and the timbre. I tried a Yamaha transacoustic once, and wasn't impressed by the timbre that the speaker put out,

Just an observation: if its a certain timbre you seek, until they synthesize drivers that speak organically by way real wood, timber may be requisite for any little “piano engine that could” …..but what do i know ? 🙂

Had the time to try two different thicknesses of that pink/purple insulation (1" and 0.5") both in 150 compression strength. I drove them with Dayton Audio DAEX25Q-4 Quad Feet 25mm Exciter 20W 4 Ohm

I confirm what everyone else said and I add only the following three points:

1. The soundstage is incredible, absolutely astonishing (with the right recording). A random (but good) CD of a Teresa Milanollo compositions made a violin player literally materialize in front of me. Other recordings with more reverb, probably mic'ed from a distance, made me think I was in a cathedral. Never experienced anything like that with speakers, definitely feels like "being there". My wife, who is usually totally numb to my audio experimentation, looked at me in awe when I forced her to go in that room and have a listen. Pianoteq blues demo also made the piano materialize just behind the panels, I felt I could touch it.
2. At least with the exciters I used, the dynamic range is somewhat limited compared to regular good speakers (I have floorstanding Polk Audio ones, mid-bottom of the line, good but nothing super fancy). Not that you cannot "pump" high volumes, but the sound becomes excessively distorted at the risk of buzzing when the moving membrane is hitting the end of the travel
3. There is not a huge difference between the 1" and .5" panel thickness. I personally find the thinner a bit more loud and a bit more clear. Family member where split 50/50 on which was best, but the difference is certainly minor.
4. Because of 2. it's important that whatever material you pick you get the most volume possible, because "cranking up" does not work. Of all the materials I tried (everything I had lying around the house and the garage that I could attach the exciter and hang) the pink insulation was best. I was expecting decent results from drywall, and it was ok, but definitely too quiet to make it worth the hassle with its friablity and weight.

I am now waiting for a quote from the company I mentioned earlier to try the 400 or 600 and the 1000 compression strengths (price permitting).

I plan to put the best among these options in my DIY hybrid piano.

Cool! I ordered my Dayton Audio drivers last month thinking that I would receive them in February - I was planning early. But I have received all of them now and did run a few experiments. I also had a 50+ inch long cardboard box, so i tried as well as the PS foam. (2foot square). I did notice that in my horizontal orientation (my intended orientation) that you need some space on the bottom side to create more bass. It's going to take some effort to optimize. I want to try the ceiling tiles (4foot x 2foot) and see if I can get a decent piano stereo image with two exciters. That is about as far as I've taken it - had to clean up the living room because family is coming over soon. Project is going on the backburner for now.