Interactive dance floor

Good afternoon.

I'm looking into the feasability of an interactive dance floor, something that would look like this:

The floor itself needs to be quite resilient to support the weight of a number of people - how much weight can a LED handle? I'm also curious about which sensors to use. I ruled out vibration and infrared sensors because of the lights and sounds inside a disco.

So, what are your thoughts? Doable? Too expensive/complex?


how much weight can a LED handle?

Don't put the LEDs under any mechanical stress. Use Lexan as a basic tile top and construct the floor out of cells of plywood with the tile let in the top.

I'm also curious about which sensors to use.

I would go for a mechanical strain gauge, or some sort of pressure sensor like in this project:- Or a quantum pill (be careful how you search for that)

Too expensive

What is your budget? I would recon on about $50 per cell so it depends on how many cells you have.

If it's a dance floor I think you also need to be concerned about how slippery it gets, especially with drinks spilled on it, whether your lights need to survive being danced on with stiletto heels and whether there's a risk of the heels getting caught between the lights.

Thick enough Lexan to deal with two 250 pound people jumping on it may run you more than $50 for just the cover. The LEDs and drivers are probably pocket change compared to the mechanical needs of a construction like that. FWIW, 2x2 feet of 1/4" thickness polycarbonate is about $60, and acrylic is about $70, if memory serves.

Very cool project! Do you have a budget?

This is a BIG project, but I wouldn't say it's "complex". If you can build one pressure-activated light-cell, you can build 100 or 300. You could do some interesting things with a combination of pressure sensitivity, music triggering, random (and not so random) pattern generation etc. Once you have a matrix, there's a lot you can do. (BTW - The dance floor for Saturday Night Fever was operated manually behind the scenes.)

My gut feeling is that jwatte is about right with 1/4" lexan (polycarbonate), assuming an unsupported square size of about 6". We seem to be using American measurements & dollars... Are you in the U.S.? From Tap Plastics, it's about $14 per square foot.

A strain gauge seems like the "obvious" choice, but from what little I know about them, I think they have very low-output and you'd need lots of amplification. It's not hard to get amplification with an op-amp, but with low signal levels you can get noise problems, and the setup just might be finicky. I was looking around to see what I could find, I found some [u]Force Sensitive Resistors[/u]. Or,you might be able to do something with short-travel tactle switches, but I'm not sure... I think you'll have to do some experiments to see what works for you. Again, you've just got to design & perfect one cell and then duplicate it a couple hundred times...

I'm not sure if LEDs are the obvious choice. This is what I'd call a "Major Effect" and you probably want LOTS of light. You can get more light for less cost with incandescent. Also with incandescents, you can screw-in a larger (or smaller) bulb if it turns-out that you want more (or less) light. At low power, the driver circuits are cheaper for LEDs, but with high power LEDs the driver circuits can get expensive. So, I think an optoisolator and TRIAC switching AC voltage is probably cheaper. On the other hand, if this thing is going to be portable, incandescents might be too bulky and might need too much power for something that has to plug-in along with your other sound & lighting equipment... I've got about 1000 Watts of lighting, and I have to be careful about adding anything else (like a fog/haze machine) because I could blow a fuse.

Doug hit spot on this is a really cool idea.

I agree that LED's are not the obvious choice. Polycarb is expensive and it is going to get scarred up so there might be replaement costs associated as well. If this is for a permanent installation, you might consider using glass wall blocks. They are usually 8" x 8" and are priced about the same as 1/4" polycarb. This would make for a much more permanent install. The slip and fall issue is going to be relevant no matter what you use if you want a transparent or opaque surface. Another problem is that you need the top surface to be smooth so a contoured surface or a support system that holds

This is all going to come down to cost. At $14/sqft just in the raw material, not including any framing or support, for a 24' x 24' dance floor, this is $8,064.

If you google interactive dance floor, it looks like most of them are made with a 1/2" (technically 15mm) tempered glass top.

Thanks everyone for your answers! I'm going to take a look at the materials and sensors you mentioned.

@Grumpy_Mike: Takin' a look at your MIDI Footsteps strain gauges. @PeterH: Good thinking, thanks for pointing the slippery problem out. @DVDdoug: I'm in Europe. With any luck, there'll be shipping possibilities or deals just as good nearby. @Sacman: Looking into these glass wall blocks...

Thanks again.

As you are in Europ you can get the Qtc pills here

Just a thought, but if you intend this to synchronise with the music and dancers, it may be possible to get the effect you want using overhead motion detection camera and pulse input from the music using standard visualisation algorithms to control the lights, and just mount them beneath a transparent floor covering. That may still not be easy to do, but I think would probably be much easier and safer than trying to make the floor also act as a sensor.

Take a look at Reactable.

This would be an awesome concept to translate to a dance floor but really complicated. The basic concept is that it uses a camera system under the table to look for distinctive patterns on the bottom of the blocks. It takes the pattern’s orientation to the center to determine some factors and allows input via the table, again picked up by the camera, to control others. The results are then projected on the bottom the opaue glass table.

To translate this to a dance floor would eliminate the need for the camera to scan for distinctive shapes but merely scan for a shape indicating the presence of a foot. It would then project whatever you wanted on the floor. The biggest challenge to this kind of setup is the distance the camera would need to be from the floor surface to get a reasonable field of view. This kind of setup would also be completely computer controlled since there is no reasonable way to translate this to a uController world. It would also likely be done in panels as you would be limited by both camera field of view as well as by how large a panel of either tempered glass or poly you could get. Of course, if you could build a dance club around the dance floor, some of these limitations would be eliminated since you could go as deep as you reasonably wanted to get the field of view needed.

Imagine being able to use non-pixelated graphics to represent literally anything you wanted on the dance floor. During a slow song, the floor emits hearts around each set of feet or it can send small hearts randomly from each foot. Or when there is a more club beat going, it cold track a foot slide versus a step and display this differently.

The things a good dance crew could do with this would be amazing! Imagine each dancer has a unique pattern on the soles of their shoes so you could choreograph the floor to their dance and make it dancer specific and addi n visuals that pulse with the music as well as the dance.

Too cool!

The only part of the floor that needs to be really hard is the top. 1/8" or 1/16" lexan over thicker plexiglas would probably work. And you'd want a thick coat of hard clear floor wax over that if you don't want to replace it too soon.

Not so cheap but reliable, I am finding that piezo disks make great force/pressure-change sensors. When wired up with 4 diodes, 2 transistors, and 4 leds, I can tell when pressed more and when pressed less and about how much. They can take quite a tap and they can tell through solid material when it has been tapped, check the door knock detector. Just a guess but knocking on the door right on the other side where the disk is might get a stronger return than knocking on the door a foot (30 cm) to the side of that spot. But I am guessing there.

Not so reliable but cheap(er), play around with capacitive sensing. It's a dance floor, not a precision data collector.

1/8" or 1/16" lexan over thicker plexiglas would probably work.

Except plexiglas (acrylic) is actully more expensive than lexan (polycarbonate) :-)