New approach to digital piano actions

If we lock up “One-Man,” Del Vento, and CG in a room with a huge bag of cash…….magic!

Look at what they’ve achieved with nearly zero budget😉; imagine if we hand them a bag with $500,000,000 and leave them to their own devices, so long as they deliver the real “acoustic-killer.”

What I would not give to see “Larry’s” face when I show him we’re not just “there,” but we actually went there twice, slapped the acoustic, and came back!

P.S.

We don’t literally need to “lock them up;” this is just a figure of speech. They can go about it as they wish, once again, so long as they deliver; and yes, I do have some suggestions, like the NanoString (it’s like a string, yet so much better!)

Did anyone watch that video all the way through?
I didn't have the patience ... so I'm left with one question: What novel result is he trying to accomplish?

MacMacMac Did anyone watch that video all the way through?
I didn't have the patience ... so I'm left with one question: What novel result is he trying to accomplish?

I watched it all the way through, and I have the same question. Understanding his words was too great a difficulty. Maybe, when I have more time I can get through it and understand it.

RIP I think I've watched this some years ago, maybe a similar project or maybe the same thing is actually older. And BTW, we've hypothesized about the possibility to simulate a real piano action through actuators/solenoids applying CPU-controlled resistance that follows the forces in a real acoustic pianos long ago on PW, I think it might have been my idea. Or maybe not, doesn't matter 🙂

Pete14 imagine if we hand them a bag with $500,000,000 and leave them to their own devices, so long as they deliver the real “acoustic-killer.”

Humans are human. Leave them with such a big pile of cash and they will end up on a luxury yacht covered in white powder and surrounded by dubious women 🤣

CyberGene Humans are human. Leave them with such a big pile of cash and they will end up on a luxury yacht covered in white powder and surrounded by dubious women 🤣

What if we allocate some money for a little R&R; let’s say, 100,000,000?

And yes, you could also work from a luxury yacht covered in white powder and surrounded by dubious women; so long as the women help you achieve your goal, I don’t mind the occasional “happy ending.”😆

MacMacMac Did anyone watch that video all the way through?
I didn't have the patience ... so I'm left with one question: What novel result is he trying to accomplish?

Ralphiano I watched it all the way through, and I have the same question. Understanding his words was too great a difficulty. Maybe, when I have more time I can get through it and understand it.

Perhaps because I'm a scientist, I found this obvious, but here it is in layman's words ( @CyberGene came close)

Instead of making an action which tries to reproduce the actual resistance of a well regulated grand, with all its quirks (typical digital folded action, with letoff simulator) or instead of using an actual well regulated acoustic action (upright or grand, as done in hybrid), let's use a "generic" action whose mechanical characteristics are chosen in real time, computationally and actuated by a mechanical device. As such, you can program the key to "feel" like whatever you please: grand, upright, historical piano, harpsichord, organ, you choose.

Does anyone know what type of haptic force actuation was used? Is that an electromagnet or something?

CyberGene
This is just a tease; imagine the possibilities if you ‘get on board with it’ (figure of speech).

Trust me, there’s a lot of that “white powder” hidden somewhere in the lower deck(s), and plenty of beautiful sunrises in the horizon; not to mention, dubious women waiting to make you happy.

Won’t you join me on the adventure of a lifetime (leading to the death of the acoustic); all we need is to raise some capital, buy the yacht, get on sailing, and somewhere between here and there, come up with the ‘perfect digital’ (to be built inland further down the road).

Pete14 OK, you won me! Where should I send my IBAN? Hopefully @RIP will join us. Ohh, and BTW, I will fire that DJ with his stupid whatever-that-dj-gadget-is and we will listen to Bach and Scriabin on that yacht, that's a must!

CyberGene
Great!

Now we only need to enroll Mac to help us raise the capital, and Drew-r to keep Mac from stealing that capital!🤣

About your demand(s), is it okay if we play Bach to a “latin beat?”

Something like this; only for Bach and Scriabin:

Surely, we need to keep them dubious women moving their ass(ets), for that’ll keep the moral up and the tide low! 🛥

Pete14 that's not even half as bad as I imagined it would be. I know how tolerant @RIP is towards spam, so I'm waiting for a bit more of this to gather and will move it into the purgatory 😀 (and it started with me...)

Purgatory, here we come!

Now, if we only had the money, perhaps the notion of working off a yacht would not be so far fetched.

Of course, I’m talking about the design/concept, but the actual prototype would be built inland.

If you guys built a hybrid, for which Yamaha and Kawai charge a bundle, I could see you taking it up a notch, once again, if the conditions were there; namely, money & time (and a little fun by the sun).

dore_m Does anyone know what type of haptic force actuation was used? Is that an electromagnet or something?

I don't know know. The guy has posted a number of videos, but I don't recall he mentioning it. I guess at the current experimental level, it does not matter? Do you have any specific reason for asking or is it just curiosity?

RIP Do you have any specific reason for asking or is it just curiosity?

Just wondering how scalable this idea is. It would be nice to be able to change the feel of your action without much trouble. It does say "electro-mechanical", and you can see the fine copper wires.

dore_m Just wondering how scalable this idea is. It would be nice to be able to change the feel of your action without much trouble.

YES! I totally agree with that.

I hope that the research of this person has been funded by a private grant by a major digital piano company, or a small start-up, and I hope that it ends up in a commercial product. I'll surely consider purchasing it if it comes out (sounds much more interesting the the LUMI, the Roli and all that kind of kind-of-lame kickstarter stuff).

It looks like this work is a Ph.D. thesis done at the university of Louvain, concluded last fall. It's website is extremely slow almost unusable, ant the internet archive has stored very little (perhaps someone in Europe can give it a try and see if geographical proximity gives them better luck). If you google "immc uclouvain haptic piano" you will find a number of relevant links. So far the only one which I've been able to read is https://dial.uclouvain.be/pr/boreal/object/boreal:254404 and I'm trying to download the 100MB dissertation. If I succeed, I'll let you know (please do the same).

Back to your question, he does say that he's using a linear actuator and reading https://en.wikipedia.org/wiki/Linear_actuator I assume it's a linear motor, which is the one that makes most sense for the application. My understanding is that most of the control electronics for linear motors is based on desired position, rather than desired force, hence some transformation between the two domains would be needed. Looking from 10,000 ft, it seems nothing that it would not be possible to offload to a microcontroller, and not being massively harder than Cybrid.

RIP My understanding is that most of the control electronics for linear motors is based on desired position, rather than desired force, hence some transformation between the two domains would be needed. Looking from 10,000 ft, it seems nothing that it would not be possible to offload to a microcontroller, and not being massively harder than Cybrid.

I agree, and I think it's a custom solenoid, a strain gauge, and a control loop. I'd think the trick and the patent (if there is one) is on the linear motor part.

Its essentially like taking a Pianodisc player piano solenoid assembly and reversing it so that instead of actuating the piano keys to hit strings, it provides backforce on the keys for haptic feedback/touch force curves.

RIP Perhaps because I'm a scientist, I found this obvious, but here it is in layman's words ( @CyberGene came close)

Instead of making an action which tries to reproduce the actual resistance of a well regulated grand, with all its quirks (typical digital folded action, with letoff simulator) or instead of using an actual well regulated acoustic action (upright or grand, as done in hybrid), let's use a "generic" action whose mechanical characteristics are chosen in real time, computationally and actuated by a mechanical device. As such, you can program the key to "feel" like whatever you please:

I have not watched any of the video, yet. I intend to but i’m a teency bit behind on many chores right now. However, i have watched what some select others have said about that video, to include @CyberGene @MacMacMac , @Ralphiano, @Pete14 and @RIP; with respect to a lay man’s understanding…..

In reverse order: 1) DelVento , quoted above, seems to be saying “ you know how it is often pointed out to lay DP enthusiasts that when they adjust the ‘key touch’ setting on their DP, this does not actually change what they feel as far as their action’s weight / softness V stiffness, it only makes a logical change in the DP’s computer/sound generator so that the pianist’s usual keypress forces are translated to a different velocity / dynamics response that provides an illusion of such?” ……well, our proposed generic DP action design, if successful, WILL mechanically do that! …. what DP actions have heretofore never been designed to do; 2) foreseeing this future THIS! moment, Pete14 is already planning some celebratory R&R, for which MacMacMac will be managing the funds and other pertinent data, Ralphiano will get back on this and our CWLO - chief wrench & lighting bolt officer - is tentatively on board with this.

I politely make a motion for MacMacMac to be at the helm of said pleasure craft, to further his lumbar’ectomy rehab sitting and standing at the wheel as he sees fit, and further motion the wheel / helm area be fully equipped with his DP & bench of choice ….. and further motion that i be appointed duty as far away as practical from monitoring any thing $ related

… …. but i could be mistaken in all of this.

RIP I assume it's a linear motor, which is the one that makes most sense for the application. My understanding is that most of the control electronics for linear motors is based on desired position, rather than desired force, hence some transformation between the two domains would be needed. Looking from 10,000 ft, it seems nothing that it would not be possible to offload to a microcontroller, and not being massively harder than Cybrid.

Yes, it would need to be a linear motor to get any kind of useful responsiveness.

There are a large number of drive schemes, most of which should still apply to linear motors, rather than rotary. Some I believe would allow direct control of force, but that's dancing quickly out of things I have experience with.

The relevant sort of motor control is usually done in a 1:1 ratio between motors and microcontrollers. Maybe you could handle a couple of motors per controller, depending. On the bright side, I think as a side effect the control loops could do a spectacular job of estimating key velocity.

A year or so later here ----- interesting vid - about haptic feedback. I think that my haptic feedback when playing piano is my ears. The air-pressure waves are providing physical feedback -- vibrating something in my ears, and then into nerve/acoustic feedback.

Also - haptic feedback in keys of pianos probably isn't going to be much use if you're playing a bunch of different notes with various fingers, and at various different velocities. But at least there's the thought of haptic feedback in keys. If it helps somebody for at least something, then that's fair enough.

And just because I decided to now embark myself into this anymore (solenoids available for free!!) let me dump here a number of things which I wanted to study but never did, in case they can be useful to the next tinkerer.

https://www.manualslib.com/manual/1256748/Yamaha-Disklavier-Mx1z.html?page=34#manual

https://www.youtube.com/watch?v=CCY6hRhjNpo

https://www.mdpi.com/2075-1702/8/4/76/htm

https://dial.uclouvain.be/pr/boreal/object/boreal:254404

https://www.actuonix.com/10mm-stroke.html

https://www.h2wtech.com/category/voice-coil-actuators#specifications1

https://p3america.com/lmc8-series/?gclid=CjwKCAjwtp2bBhAGEiwAOZZTuPqK_G2fJk9oEgaK0SwhQ5He0Zib9v1B1H3CsS6eX3swI3IYtEKBnhoCtQQQAvD_BwE

https://github.com/andrebroekman/FlyByPi/blob/master/GettingStarted/User%20Guide.pdf

https://www.motioncontroltips.com/whats-the-difference-between-torque-mode-and-velocity-mode-in-servo-control/

https://digital.lib.washington.edu/researchworks/bitstream/handle/1773/34887/he-v3n2.pdf?sequence=1&isAllowed=y

https://www.digikey.com/en/articles/why-and-how-to-use-current-dacs-actuator-drive-closed-loop-control

https://www.electronics-tutorials.ws/io/io_6.html

https://www.emworks.com/application/linear-solenoid-actuator

https://www.mouser.com/ProductDetail/Ledex/195204-228?qs=l7cgNqFNU1h0QIylMDhndA%3D%3D