Wednesday, January 10, 2007

Thursday, November 23, 2006

Potential pictures for ballettanz

Tuesday, October 10, 2006

Light strip box

Here is a picture of the side of the box: connectors for USB, DC power, BNC (for laser control or laser sensor) and a 9 pin SUBD connector for the light strip. In the middle will be the PIR connectors (don't have a picture of that; it is a 6 pin Mini DIN connector, same as used for PS/2).

Inside the box, I needed to do a little trick to get the light strip cable to the top: see the nice curve to make an opening for the cable:

As two of the boxes are of another model, I did not have to do this for those two.

Upon arrival of the package, you will need to solder the lightstrip that you have already there in Montreal to the SUBD connector. Here is a description of how to do this:

- check which wire of the 9 is the common one, which goes to all sensors (I think it is the red one, but check to be sure).

- cut off the part where the resistor array and the pins are connected. This part is inside the lightstrip box already. Keep the part though, so I can get it back. We may want to use it again someday (recycling is always good). Keep a couple of centimeters of wire on it too, so I could use it to make a backup lightstrip box, or something.

- put the little piece of Schrumpfschlauch (crimp stuff), around the cable, as shown here:

It is good to do this at the start, as it is frustrating to find out you forgot after having soldered it all.

- wire strip all nine wires and tin the ends (put solder on them, let the solder melt into it, so the connection becomes really solid), as shown here:

- then put the SUBD connector on the box

- and start soldering the wire to the box. The most important thing is that the common wire (the red one in the picture, and probably also on your lightstrip) is on the most right upper row.

- The other ones can be soldered to the other pins as you like, though I recommend to do it alternating on the upper and lower pins, so the odd ones go on top, and the even ones to the bottom, as shown here:

- When all wires are soldered, you can move the crimp thingy up. You'll need to cut it a little to move it around the connector, then use a lighter to heatshrink it. Then use some tape to secure it even more.

EL wire box update

Other invertors, so another picture of what it looks like inside the box:

Not on this picture yet: some diodes are needed towards and from the power of the servo to keep the servo from disturbing the +5V and GND levels.

Before I forget to mention: the boxes need 12 V DC, + on the tip.

Wednesday, October 04, 2006

Ideas for the "monolith" - which now replaces the original hanging screen

Here are some images of the kind of visual quality that the "monolith" that is tied to Michael's speaking of the Artaud text should have. The idea now is to have a light box of sorts anchored to the wall at the opposite end of the space from the table/chair.

Lighting (dimmable) would be mounted on the back of the surface that would dim up/down for those scenes where M. speaks the Artaud. When not lit, it should appear as a void in the black wall-hardly noticeable.


Thursday, September 28, 2006

Press Photos for Place des Arts, Montreal

These are the pictures that Chris chose as press photos for Place des Arts. They are shown in reversed order: The top most picture is preference No 6, the one the at the bottom the most preferred. Pictures by Heidi Gilpin and Anke Burger.

Tuesday, September 26, 2006

Sensor network

I worked out a basic setup for how it should work, being the following:

The room control board sends out two kinds of messages: the first is a message with the current status of everything that needs to be activated. This is just one byte, as we have 8 different things we want to turn on and off (2 lasers, 5 el-wires, 1 paper).

The second is a message with a request for information from a lightstrip. It has a byte to indicate from which one it wants to get an update.

So the room control board cycles through sending:

- status update
- request light data 1
- request light data 2
- request light data 3
- request light data 4

The different actuators in the room, listen for the status update, and turn on or off the thing they are supposed to turn on or off.

The light boxes reply with data, when they are asked to transmit data. The data they send is:

8 photo sensor values
2 bytes for PIR data
1 byte for laser data

The bytes for the PIR and laser are created as follows:

upon a state change of the sensor (movement detected or laser break detected), the board rotates the byte one time to the left, and adds the new value (0 or 1).
Just before transmission, the byte gets updated one more time with the current value.

This way, there is some history of whether or not the sensor has detected something in the time in between sending. It should be possible to also detect a change if data gets lost on the way, and only gets received correctly the second time (or third).

So, this is the theory. Didn't get it to work yet.

The box for the laser

Ok, so I broke open the lasers in order to be able to control them electronically.
Then I had to put them into a box again, to make them safe and usable.

Here is the open box:

From left to right we first see the BNC connector, the battery holder, and then on the right:

from top to bottom: the MOSFET (a IRLD024, as recommended by Segor), the internal electronics as found in the pen. The MOSFET is connected to the exact same pins as the actual button of the laser pen. This button is still there (at the back of the tiny PCB) - and can still be used for testing -, as are the remains of the pen.

To close the box, we put the other half on. This may need some wiggling to get it all right. Then there are two screws:

and when we screw them tight, we have a nice black box:

As was quite clearly visible in the first picture, there is quite a lot of space left between the battery holder and the electronics. This space can be used to screw on a connector for the microphone stand.

Photo sensor for the lasers

(to detect Michael moving through them)

These are the boxes for the photo sensor, one closed and one open:

As a connector I took a normal BNC connector, as they connect well, and cables are common. I bought some 3m cables, but I am not sure whether that will be long enough.
The photo sensor is mounted into a little tube, and the careful observer may recognize the tube from what was once a laser pen. Yes, this is recycling.

A close-up:

I tried to mount the sensor itself, as well as possible in the middle of the tube. Also notice the foam underneath the tube, to keep the tube in place.

There is still some space left to screw on a connector for the microphone stand.

Before closing the box, another piece of foam goes on top:

So that in the end the tube is pushed towards the middle from both sides. The lid only fits on one way, so that is easy to figure out. No need to screw anything either, as the lid closes tightly. You'll need a screwdriver to open it.

As I mentioned, I used the remains of the laser pen for the tube. This means that for transport the tube can be closed:

and before use opened again:

One more note on the photo sensor: the laser emits at 532 nm. This means that it would be most handy to have a photo resistor that has its peak sensitivity on that wavelength as well. From the Reichelt catalog, I get the following photo resistors with a peak at 530 nm (close enough):


They slightly vary in resistance.

I will check out if my local electronics store Segor has these as well, or others that match the right wavelength, so that I can finish the sensors properly before sending them out.

Transmitter connection

As per Chris' request, I made some pictures of how the transmitter could be connected.

First, your solder some wires between the important pins as shown in this picture:

The pins are (counting from the bottom:)
Pin1: GND (black wire)
Pin2: +5V (red wire)
Pin6: RA2 (green wire)
Pin13: RA4 (yellow wire)

The wires can be a bit shorter probably. Just see what fits nicely for the box, keeping in mind that it shouldn't be awkward to open the box in case of need.

then, connect one side to the transmitter:

and the other to the CUI board:

and voilĂ :

interesting, as I was writing this and looking at the pictures again, I realised I had wired the green wire wrong... so I needed to make the pictures again...


MOSFET circuitry

So I went to Segor, the local electronics store, and asked for some advice how to the switching on and off. The guy behind the counter was really nice and recommended the IRLD024 (costing 70 eurocents a piece; probably cheaper if I get 10 or more) and said I could use them both for the 3V (lasers) and 12V (el-wire). He drew me the pin description, and I went home with two of them (and all of the other stuff I bought) to try them out.

At home, I set to work. The laser was easy: connecting the PIC output to the gate (G), the ground to the source (S) (ground both from the PIC as well as from the laser), and the other side of the laser button to the drain (D) was all there was to it.

The EL-wire was slightly more difficult. I first tried putting the MOSFET in between the 12V of the DC input to the box to the 12V of the input to the invertor, but this didn't work. So I ended up with this working connection:
PIC output to the gate (G)
GND (board) to the source (S)
GND of the power input to the invertor to the drain (D)

So, in the coming week, I need to get more, so I can make these switches for the other laser and for the other EL-wire boxes.

Friday, September 22, 2006

Progress on the hardware side

Most of this stuff I already wrote to Chris, but just to keep you all informed:

I have succeeded in peeling open the laser pens, to get to the electronics. So all I have to do now is attach a MOSFET switch with some wires to turn them on and off electronically with the CUI.
When I say peeling open, I really mean peeling. I used my wire cutter and then basically peeled it open further like a banana; a sturdy one but still :)

I also made an alternative housing for the laser pens, which now wait for the MOSFET circuit to be built in.

For the EL-wire CUIs, I found a box and drilled some holes in it. Everything will fit in nicely there, with on one side a DC power connector and a LED, and on the other side the servo motor with the wire on it.

That box is now also waiting for the MOSFET circuitry.

As for the MOSFET stuff: I am in the process of choosing and buying these.

Then I have also built up some more CUIs, so I can do the testing of the sensor network. There I am also still waiting on more receiver boards to arrive (due to some change of mind about frequencies, I have a bunch of mismatching transmitters and receivers now).
Some of the CUI boards aren't working yet, so I have some debugging to do there too.

I ordered new PCBs to be made, though I am unsure at which stage they are now. There was a problem with the drilling information, so I sent them more info, but they didn't get back to me yet. I guess I can check that out on Tuesday, as on Monday I'll be busy with a WFS symposium all day.

Anyway, I'll try to get as much stuff done as possible this weekend, but I still have to prepare some stuff for the WFS thing too. So maybe I'll take some extra days next week for the Schwelle project, as on the weekend I'll have to take the sunday for WFS.

Berlin out.

spatialization ideas

Scene 1: Room Breathes
Room sounds pulse back and forth between the on-stage speakers and those behind the audience as room breathes. This could be enhanced by a similar pulse between reverb wet/dry.

Scene2: Michael Enters
Michael's entrance shifts the spatialization balance from an equal movement (pulse) between on-stage/off-stage speakers, to a more off-stage or on-stage mix with a wetter or dryer reverb mix (depending if we want the room to contract or expand as he comes on-stage), I can't decide. I think if we have the sound associated with Michael suddenly appear as he appears, we could push the room mix to the back speakers and increase the reverb wet mix. This might allude to michael diplacing the room slightly.

More to come....