I am working with several ultra sound sensors called Ping from Parallax wired to an Arduino board that is connected to puredata. currently I am working with 3 sensors and it works fine.
now I want to work with 5 sensors and the following problem occurs:
- I get "phantom" values on sensors that are not modified: if, for example, I change the value of sensor B, sensor C also changes its values, not exactely syncronized but it jumps to a simlar value range. very annoying indeed.
is that a problem of my script for arduino or a problem of pd? does anybody already have experience with such a configuration? I am using Pd version 0.39.2-extended-RC1 on windows xp.
attached you find the arduino and pd patch for that idea.
thanks, marko
Hi,
your code and patch look good. It is not clear to me, why you are using these delaytimes for the different sensors. but that should not affect your results anyway. the minimum delaytime between two measurements is only 200microseconds. but maybe it also depends on the length of the impulse? would it be possible, that the sensing fields of the 5 sensors overlap, and you get the crosstalk from that? on the other hand, it may also be possible that the problem is in your circuit. you could try to use 10k pulldown resistors. maybe that stops the crosstalk.
marius.
Marko Timlin wrote:
I am working with several ultra sound sensors called Ping from Parallax wired to an Arduino board that is connected to puredata. currently I am working with 3 sensors and it works fine.
now I want to work with 5 sensors and the following problem occurs:
- I get "phantom" values on sensors that are not modified: if, for example, I change the value of sensor B, sensor C also changes its values, not exactely syncronized but it jumps to a simlar value range. very annoying indeed.
is that a problem of my script for arduino or a problem of pd? does anybody already have experience with such a configuration? I am using Pd version 0.39.2-extended-RC1 on windows xp.
attached you find the arduino and pd patch for that idea.
thanks, marko
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Yes, please describe the physical configuration of the 5 sensors Marko. I am working on a similar project using 6 sensors, but we have not done much testing in our proper configuration yet.
My guess is that since all the sensors are hard wired with a specific ping frequency, they will detect each other's pings when placed in close proximity.
~Kyle
On 6/3/07, marius schebella marius.schebella@gmail.com wrote:
Hi,
your code and patch look good. It is not clear to me, why you are using these delaytimes for the different sensors. but that should not affect your results anyway. the minimum delaytime between two measurements is only 200microseconds. but maybe it also depends on the length of the impulse? would it be possible, that the sensing fields of the 5 sensors overlap, and you get the crosstalk from that? on the other hand, it may also be possible that the problem is in your circuit. you could try to use 10k pulldown resistors. maybe that stops the crosstalk.
marius.
Marko Timlin wrote:
I am working with several ultra sound sensors called Ping from Parallax wired to an Arduino board that is connected to puredata. currently I am working with 3 sensors and it works fine.
now I want to work with 5 sensors and the following problem occurs:
- I get "phantom" values on sensors that are not modified: if, for example, I change the value of sensor B, sensor C also changes its values, not exactely syncronized but it jumps to a simlar value range. very annoying indeed.
is that a problem of my script for arduino or a problem of pd? does anybody already have experience with such a configuration? I am using Pd version 0.39.2-extended-RC1 on windows xp.
attached you find the arduino and pd patch for that idea.
thanks, marko
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Kyle Klipowicz -
marius schebella -
Marko Timlin