I was wondering if any of you could recommend sources for sensor fabrication. I'm interested in building my own sensors for use in performances w/ computer interface. I'd love advice about books to read, websites, or even where to start (i.e. do I need to be an electrician first). Thanks,
-i
hi,
i use a lot the basicstamp from www.parallax.com to make interfaces. i usually conncet them via serial port and comport object to pd/gem.
lg
e
On Thu, 30 Oct 2003, 0 wrote:
I was wondering if any of you could recommend sources for sensor fabrication. I'm interested in building my own sensors for use in performances w/ computer interface. I'd love advice about books to read, websites, or even where to start (i.e. do I need to be an electrician first). Thanks,
-i
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
they have some sensor systems. maybe also help. or do it yourself packages. what kind of sensors do you think of?
Marius.
----- Original Message ----- From: "0" root@0x09.com To: pd-list@iem.at Sent: Friday, October 31, 2003 3:22 AM Subject: [PD] Sensors
I was wondering if any of you could recommend sources for sensor
fabrication. I'm interested in building my own sensors for use in performances w/ computer interface. I'd love advice about books to read, websites, or even where to start (i.e. do I need to be an electrician first). Thanks,
-i
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
Hi,
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
Regards,
Chris.
On Thu, Oct 30, 2003 at 07:22:07PM -0700, 0 wrote:
I was wondering if any of you could recommend sources for sensor fabrication. I'm interested in building my own sensors for use in performances w/ computer interface. I'd love advice about books to read, websites, or even where to start (i.e. do I need to be an electrician first). Thanks,
-i
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
On Fri, Oct 31, 2003 at 09:25:34PM +0800, Chris McCormick wrote:
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
Interesting, but I'm not sure it's possible to poll the parallel port as fast. With my parapin object, I use 1Khz by default, which correspond to the maximum resolution of timers in PD (but I might be wrong). http://www.pure-data.org/Members/odradek/parapin -- Marc
I really don't understand why people are so hooked up on using MIDI or serial/parallel ports for interfacing sensors with pd. It strikes me that the absolute best way if you are at all worried about sample rate/quantisation amount is to use the audio signal inputs of your soundcard. some simple circuitry and then you have maybe a 96kHz 24bit input - super fast control and with same latency!
am I mistaken? has anyone used this to good effect? I was thinking that the terminatorX style turntable things are a prime example of where a mouse just doesn't offer enough resolution to do even a half decent job....
matt
-=-=-=-=-=-=-=-=-=-=-=-=- http://www.loopit.org/ -=-=-=-=-=-=-=-=-=-=-=-=- ----- Original Message ----- From: "Marc Lavallée" odradek@videotron.ca To: pd-list@iem.at Sent: Friday, October 31, 2003 2:01 PM Subject: Re: [PD] Sensors
On Fri, Oct 31, 2003 at 09:25:34PM +0800, Chris McCormick wrote:
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
Interesting, but I'm not sure it's possible to poll the parallel port as fast. With my parapin object, I use 1Khz by default, which correspond to the maximum resolution of timers in PD (but I might be wrong). http://www.pure-data.org/Members/odradek/parapin -- Marc
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
On Fri, Oct 31, 2003 at 02:29:54PM +0000, matthew jones wrote:
I really don't understand why people are so hooked up on using MIDI or serial/parallel ports for interfacing sensors with pd. It strikes me that the absolute best way if you are at all worried about sample rate/quantisation amount is to use the audio signal inputs of your soundcard. some simple circuitry and then you have maybe a 96kHz 24bit input - super fast control and with same latency!
am I mistaken? has anyone used this to good effect? I was thinking
that
the terminatorX style turntable things are a prime example of where a mouse just doesn't offer enough resolution to do even a half decent job....
There's many ways to interface sensors. Using an audio card is definitely possible, but it's not a silver bullet. Many audio cards cannot be used an input interfaces, often because the input stages are expecting audio signals and nothing else.
Here's a good example of using an audio input as an interface input: http://www.inf.ethz.ch/~mringwal/lirc/ This PDF explains that some audio cards are using a DSP to filter the input signals: http://www.minidisc.org/contrib/oded_files/AudioCardIRSampler.pdf
I'll try this audio input lirc interface, and see if the idea could be extended to other interface circuits. -- Marc
On Fri, 31 Oct 2003, Marc Lavallée wrote:
There's many ways to interface sensors. Using an audio card is definitely possible, but it's not a silver bullet. Many audio cards cannot be used an input interfaces, often because the input stages are expecting audio signals and nothing else.
BTW, if you have electric signals that are of very low frequencies not easily dealt with using audio cards, or maybe even signals that don't have frequencies per se, then you may modulate those using a high frequency wave, and then analyse the resulting signal in software to find out its volume. I have successfully used that to pick up the signal from a photosensor, through an old audio card, and into a Ruby script, a few years ago. This should be easy to do. I used a 555 back then, but I guess it can be done using about 2 capacitors and 3 transistors, if I'm not completely incompetent wrt electronics.
Mathieu Bouchard http://artengine.ca/matju
On Fri, 31 Oct 2003, matthew jones wrote:
I really don't understand why people are so hooked up on using MIDI or serial/parallel ports for interfacing sensors with pd. It strikes me that the absolute best way if you are at all worried about sample rate/quantisation amount is to use the audio signal inputs of your soundcard. some simple circuitry and then you have maybe a 96kHz 24bit input - super fast control and with same latency!
Soundcards filter out all dc ("constant bias") on their inputs. You will not get good measurements from unmodulated, slowly ( <10Hz) varying signals. You could modulate the sensor signal in some way, but this requires more analog electronics.
am I mistaken? has anyone used this to good effect?
Yes, I've used a soundcard to sample ultrasound (sonar) receivers directly. This is (in my case) a 40kHz modulated signal. This shifts the demodulation from hardware to software... So 96kHz is usefull - not only for audiophiles :)
best, Johannes
Actually, USB mice can have quite good resolution and sampling rates. The problem is the OS drivers limit the resolution to screen pixels but there are tricks to get around that. I think the sampling rates are on the order of 1-5 khz with a resolution of 400-800 pixels per inch.
http://www.semiconductor.agilent.com/cgi-bin/morpheus/home/home.jsp?pSection...
.hc
On Fri, 31 Oct 2003, matthew jones wrote:
I really don't understand why people are so hooked up on using MIDI or serial/parallel ports for interfacing sensors with pd. It strikes me that the absolute best way if you are at all worried about sample rate/quantisation amount is to use the audio signal inputs of your soundcard. some simple circuitry and then you have maybe a 96kHz 24bit input - super fast control and with same latency!
am I mistaken? has anyone used this to good effect? I was thinking that the terminatorX style turntable things are a prime example of where a mouse just doesn't offer enough resolution to do even a half decent job....
matt
-=-=-=-=-=-=-=-=-=-=-=-=- http://www.loopit.org/ -=-=-=-=-=-=-=-=-=-=-=-=- ----- Original Message ----- From: "Marc Lavallée" odradek@videotron.ca To: pd-list@iem.at Sent: Friday, October 31, 2003 2:01 PM Subject: Re: [PD] Sensors
On Fri, Oct 31, 2003 at 09:25:34PM +0800, Chris McCormick wrote:
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
Interesting, but I'm not sure it's possible to poll the parallel port as fast. With my parapin object, I use 1Khz by default, which correspond to the maximum resolution of timers in PD (but I might be wrong). http://www.pure-data.org/Members/odradek/parapin -- Marc
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
zen
\
\
\[D[D[D[D
there is even a "gamers mouse" with 2000dpi:
http://www.gamesdomain.com/gdreview/zones/reviews/pc/jan01/razerrev.html
gotta love its name too ... the "Razer Boomslang 2000"
-josh
Hans-Christoph Steiner wrote:
Actually, USB mice can have quite good resolution and sampling rates. The problem is the OS drivers limit the resolution to screen pixels but there are tricks to get around that. I think the sampling rates are on the order of 1-5 khz with a resolution of 400-800 pixels per inch.
http://www.semiconductor.agilent.com/cgi-bin/morpheus/home/home.jsp?pSection...
.hc
On Fri, 31 Oct 2003, matthew jones wrote:
I really don't understand why people are so hooked up on using MIDI or serial/parallel ports for interfacing sensors with pd. It strikes me that the absolute best way if you are at all worried about sample rate/quantisation amount is to use the audio signal inputs of your soundcard. some simple circuitry and then you have maybe a 96kHz 24bit input - super fast control and with same latency!
am I mistaken? has anyone used this to good effect? I was thinking that the terminatorX style turntable things are a prime example of where a mouse just doesn't offer enough resolution to do even a half decent job....
matt
-=-=-=-=-=-=-=-=-=-=-=-=- http://www.loopit.org/ -=-=-=-=-=-=-=-=-=-=-=-=- ----- Original Message ----- From: "Marc Lavallée" odradek@videotron.ca To: pd-list@iem.at Sent: Friday, October 31, 2003 2:01 PM Subject: Re: [PD] Sensors
On Fri, Oct 31, 2003 at 09:25:34PM +0800, Chris McCormick wrote:
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
Interesting, but I'm not sure it's possible to poll the parallel port as fast. With my parapin object, I use 1Khz by default, which correspond to the maximum resolution of timers in PD (but I might be wrong). http://www.pure-data.org/Members/odradek/parapin -- Marc
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
zen
\[D[D[D[D
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
On Fri, 31 Oct 2003, Chris McCormick wrote:
Hi,
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
How about timing the pulse width (from the capacitor discharges) directly, instead of counting the frequency? The legacy joystick port works this way (also using polling).
Still, if you want to build your own sensor-to-pc interface, I think using a microchip PIC microcontroller (on serial port, on a MIDI port, or the USB-series) for analog inputs is a more elegant solution. It requires the use of a programmer once though, but those are not expensive either. It is less components, less handwork, less CPU cycles wasted on polling, and more resolution. Have a look at www.ucapps.de for a well-documented and free design.
I started messing with those chips a couple of months ago, and I'm pleased with them. I've programmed them for analog-to-midi, as a 12-channel infrared-light-barrier-controller and as multichannel pulse generator.
regards, j#|@
Hey Johannes,
Have you heard of the AID project? (www.interaccess.org)
We're making a GPL PIC16F877 based general sensor platform for artists in particular. We already have 8 channels AtoD, 2 PWM ports, I2C and a 8bit bus to communicate with up to 16 digital "peripheral" cards. Its got only Serial interface for now, but we're working on USB as well.
I've also started a suite of control abstractions for PD, based on the comport and ascii externals.
We're always looking for help (especially in programming- we're using hitech C) Please jong the mailist or drop me a line if your interested in contributing.
Ben
----- Original Message ----- From: "Johannes Taelman" Johannes.Taelman@UGent.be To: "Chris McCormick" chris@mccormick.cx Cc: pd-list@iem.at Sent: Saturday, November 01, 2003 5:59 PM Subject: Re: [PD] Sensors
On Fri, 31 Oct 2003, Chris McCormick wrote:
Hi,
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
How about timing the pulse width (from the capacitor discharges) directly, instead of counting the frequency? The legacy joystick port works this way (also using polling).
Still, if you want to build your own sensor-to-pc interface, I think using a microchip PIC microcontroller (on serial port, on a MIDI port, or the USB-series) for analog inputs is a more elegant solution. It requires the use of a programmer once though, but those are not expensive either. It is less components, less handwork, less CPU cycles wasted on polling, and more resolution. Have a look at www.ucapps.de for a well-documented and free design.
I started messing with those chips a couple of months ago, and I'm pleased with them. I've programmed them for analog-to-midi, as a 12-channel infrared-light-barrier-controller and as multichannel pulse generator.
regards, j#|@
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
Hi, Christan Klippel has made the CVBOX interface: 88 channel AD converter with 8, 10 or 12 bits of resolution. Connects to the parport. Schematic, circuit boards and externals for Pd (Linux/Win), jMax and Max (Win) available online. Try: http://www.nullmedium.de/dev/hardware/cvbox.html
Christian also is working on a USB version but I don't know when this will be finished.
Olaf
"B. Bogart" schrieb:
Hey Johannes,
Have you heard of the AID project? (www.interaccess.org)
We're making a GPL PIC16F877 based general sensor platform for artists in particular. We already have 8 channels AtoD, 2 PWM ports, I2C and a 8bit bus to communicate with up to 16 digital "peripheral" cards. Its got only Serial interface for now, but we're working on USB as well.
I've also started a suite of control abstractions for PD, based on the comport and ascii externals.
We're always looking for help (especially in programming- we're using hitech C) Please jong the mailist or drop me a line if your interested in contributing.
Ben
----- Original Message ----- From: "Johannes Taelman" Johannes.Taelman@UGent.be To: "Chris McCormick" chris@mccormick.cx Cc: pd-list@iem.at Sent: Saturday, November 01, 2003 5:59 PM Subject: Re: [PD] Sensors
On Fri, 31 Oct 2003, Chris McCormick wrote:
Hi,
An idea I had recently (though I'm sure it's not original) is to hook up 8 555-timer circuits with variable resistors to the parallel port and poll it periodically to figure out the frequency of the 8 squarewaves on individual parallel port pins. These frequencies could be used as control values in a Pd patch. The advantages are that 555 timer circuits with a variable resistor are very cheap, and simple. Your range would be from where frequency updates are too far between (let's say 100Hz) to the upper-range of the parallel port polling frequency. So if you can poll it at 8000Hz you would probably want to stop at nyquist to get meaningful updates let's say at 4000Hz (not quite certain that nyquist is relevant here though).
How about timing the pulse width (from the capacitor discharges) directly, instead of counting the frequency? The legacy joystick port works this way (also using polling).
Still, if you want to build your own sensor-to-pc interface, I think using a microchip PIC microcontroller (on serial port, on a MIDI port, or the USB-series) for analog inputs is a more elegant solution. It requires the use of a programmer once though, but those are not expensive either. It is less components, less handwork, less CPU cycles wasted on polling, and more resolution. Have a look at www.ucapps.de for a well-documented and free design.
I started messing with those chips a couple of months ago, and I'm pleased with them. I've programmed them for analog-to-midi, as a 12-channel infrared-light-barrier-controller and as multichannel pulse generator.
regards, j#|@
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
sensors are almost always used for show off
if you really want to have a hand on controll a trackball is the best way to go.
and you have 2 axis to play with
sonofsound.com
On Thu, 30 Oct 2003, 0 wrote:
I was wondering if any of you could recommend sources for sensor fabrication. I'm interested in building my own sensors for use in performances w/ computer interface. I'd love advice about books to read, websites, or even where to start (i.e. do I need to be an electrician first). Thanks,
-i
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
On Fri, Oct 31, 2003 at 12:32:52PM -0500, The Tripper wrote:
sensors are almost always used for show off
if you really want to have a hand on controll a trackball is the best way to go.
Maybe, but some people likes to have fun with something else than one of their hand.... -- Marc
or one who really wants to perform with is back to the computer... it's the way to go, some body expression, which is a lot what a performing instrument is about anyway. If not why go to any concert at all? Less tape music, less laptop performances and more physicality to the electronic instrument performance....
My 2 cents I. On Friday, Oct 31, 2003, at 17:17 Europe/Lisbon, Marc Lavallée wrote:
On Fri, Oct 31, 2003 at 12:32:52PM -0500, The Tripper wrote:
sensors are almost always used for show off
if you really want to have a hand on controll a trackball is the best way to go.
Maybe, but some people likes to have fun with something else than one of their hand.... -- Marc
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
Ivan Franco
Research Associate
YDreams S.A.
ivan.franco@ydreams.com http://www.iilab.com/ivan/ http://www.ydreams.com/
-
0 -
B. Bogart -
Chris McCormick -
Hans-Christoph Steiner -
Ivan Franco -
Johannes Taelman -
Josh Steiner -
Marc Lavallée -
marius schebella -
Mathieu Bouchard -
matthew jones -
Olaf Matthes -
rat＠telecoma.net -
The Tripper