This May I’ve decided to embark upon a project to create a piano which makes light out of situations. And I might need your help / expertise!
My parents are moving house and my piano which was in North Wales is now sitting in my front room in London, having been bumbled across some cobbles and hauled up the stairs by some very triangular muscled men from Kent. It’s very exciting. But it got me to thinking that I’d like to do more than learn new music, brush up on pieces I haven’t touched for years and give the odd lesson here and there.
I’d like to make a piano which spews light, cleverly, depending on what keys are pressed. It should be fairly seamless and not overt, so the workings need to hide inside, but the basic premise is:
Someone plays the keys -> Some magic processing happens -> The music is interpreted, tastefully, in light form and slips out from above the keybed, over the top of the piano and beyond. Here’s a crap drawing of what it might look like:
Not only would it be a fun project but it would also be a great thing to contribute to Burning Man festival in Nevada, which I’m hoping to attend for the third time either this August or next. We camp with Illumination Village, a group of talented like-minded souls who bring joy to the festival through incredible fire art, beautiful photography and unicorn stampedes, among other things.
Back to the piano. There are some major logistic challenges ahead, and I think the key things are:
How to sense what keys are being pressed?
My early research shows there are a number of ways to do this. Using either pressure sensors or vibration sensors. I’m not sure if the latter would be suitable due to string cross-vibrations but the latter would certainly be cheaper. Then there’s the issue of there being 88 inputs and ideally it’d be great to be able to sense them individually so these would need to be processed in some way.
There are systems such as PNOScan and TFT midi record which can be fitted to a piano – they are an array of sensors and a strip which sits in the keybed. it takes a little time to do but it’s by no means impossible. They also provide a controller box which spews midi out, and my music geek education means I’m on safe ground when it comes to midi data and how to process it, which is a good thing. The downside is cost: the PNOScan system is available in the UK for a whopping £750 which feels a bit steep. Especially when there are other bits to consider…
How will the processing work?
I already have an arduino and have just got an midi shield for it which will allow me to process midi in, however it’s a little board with none of the pots or connectors soldered on. I’m also getting my electronics foo back on early in the process :) The first step will be to plug my midi keyboard into this module, attach it to my arduino and laptop and start doing some basic processing with a simple LED array, which brings me on to…
How will the lights work?
This is the bit I’m least sure about. There are plenty of LED rope lights and multi-colour LED ribbons but I don’t yet know exactly how to control them or whether they are individually controllable. Ideally there would be a way to address the lights individually via a controller to allow some funky processing, eg if someone plays an F# then that one light would come on, then the next in the series, then the next, so the effect would be of the note emanating from the top of the chain through to the bottom. This will be the biggest challenge for me. At the moment I’m thinking of building a string of say 20 of multi-coloured LEDs wrapped in ping pong balls, attached to a controller, and repeating that a number of times (possibly 88, possibly only one per octave)
How much will this thing cost?! And what are the other constraints?
I’m beginning to realise how much art projects on the playa can set people back. It’s going to cost a pretty penny. I think I’m setting my budget at about £2k but that will have to cover:
- Electronics to provide note on / off information from each key
- Arduino kit & midi shield
- Array of multi-colour LEDs
- LED controllers (DMX?)
- A second-hand upright piano to be located in San Francisco
- Hire of van with tail lift and ropes for securing
- Paint and decoration
- Tools to make modifications to piano
- Potential post-festival storage of equipment and/or piano
There’s also the small matter of being able to get the electronics from London to a little north of Gerlach which means the eletronics and piano fittings need to be robust and fit into something no larger than a suitcase (and I’ll probably have some ‘splainin to do at customs), and it will need to be able to withstand getting somewhat dusty. OK, very dusty.
What should it be called?
This project needs a name. That’s one of the first things to sort.
What’s happening this week?
With the midi shield here and my soldering kit on the way I’m going to create a little prototype which goes all the way from receiving midi. This will give me a good idea of:
- Whether I should be trusted with a soldering iron
- Whether arduino is good enough to be able to process the midi and do simple and more complex light manipulation
- Come up with some names
Are you still reading this?
If so you might have some useful knowledge which I’ll definitely need to get this thing going! Get in touch in the comments or email simon at this domain name.