Go Down

Topic: Need help with shielded wires for capacitive sensing using long cables (Read 8152 times) previous topic - next topic

viemaob

Hi,

I am building an installation with four capacitive sensors made with Bare Conductive ink - each connected to a Fio Xbee via 6-8 meters long cables (unshielded). I was expecting interference - and also I am getting real slow sensor readings (2000ms delay).  I have been through changing the resistor (between the send and the receive pins) and the capacitor values (between the receive pins and GND), plus the GND is connected to a water pipe. At the moment I am using 10M Ohm res. and 100pF caps. Still much interference and slow readings.
I could try shielded wire, though I am not sure how to connect the shielded wire to the capacitive sensor. What am I exactly doing with the shield..?
Also other recommendations are very appreciated regarding my setup.
Extra info: I am reading the inputs in MaxMSP via serial port, and using Arduino Fio w/ Xbee to Xbee Explorer.

Peter_n

Using capacitive sensing and using 6 meters wires seems impossible to me.
You can try antenna cables, maybe it will work with 2 or 3 meters.

There should be an Arduino (or capacitive sensing chip) near the sensors. There is no other way (in my humble opinion).

viemaob

I cannot cut the cables shorter in this particular installation - and a bit of interference could also be for my advantage. Though I need stable input readings, and less delay… It takes around 2000-2500 ms to update the readings, and that is quite slow. Of course, the less the interference the better - that is why I was asking about shielded wire, and how exactly to hook up to the sensors / circuit.

Peter_n

Can you get an Arduino Pro Mini in there ?
http://arduino.cc/en/Main/ArduinoBoardProMini
If a few cables fit, so will the Pro Mini.

viemaob

The sensors are hanging from 3-4 m high trees, and the sensor cables are gathered in the centre (3-4 m away from the trees) in an Arduino Fio, that sends the data to the computer via Xbee communication. I chose the Xbee way, as I am comfortable with it, and being wireless also fits my concept and the use of the space.
Could you help me explaining how exactly you imagine to add the Arduino Minis..?

I really would give the shielded wires a try, I have been reading many suggestions about it - I am though not quite sure, how to implement the shield itself..

Peter_n

That won't be easy. The meters of wires are sensors themselves.

I guess that if people touch an object with conductive ink, something is happening ?
Perhaps you can get an Arduino Mini Pro inside that object.
Or maybe you should use a completely different sensor.

viemaob

It is okay if the cables are also working as sensors - though I still need to find a way to reduce the interference somehow, and avoid getting these very slow sensor readings… The shorter the wires, the better the results, I know - though I will give it a try with shielded wires - a test cannot harm. I am though still not quite sure how to work with the shield - should I connect it to the GND of the board..? The board itself is though only grounded via a nearby water pipe at the moment.
I am not an advanced user, as you can tell - I would appreciate more detailed explanations / recommendations :)

Peter_n

I'm just guessing here, but I think some old coaxial TV cable cable can be used. Connect the sense wire from the Arduino to the core wire and the Arduino GND to the woven shield. That will have an influence on the capacitance, but at least it is a fixed influence.

viemaob

Thanks, I will connect the new cables tomorrow and see how it changes things… Will give a feedback - I am also concerned about too much capacitance, but let's see…

viemaob

this did not work out well - probably i am connecting smth wrong, probably it is a too much capacitance..? the sensor value goes to -2, as it exceeds the set max. update time.

Peter_n

That could be the capacitance.
I don't know how to solve this. Perhaps with extra hardware or in a different way, and still use also the long wires as part of the sensor.

viemaob

I did a test today in the installation space, and managed to cut the wires approx. 2 meters shorter… Readings are much faster, and there is still some interference, but I can work with it, I think… Fingers crossed now, that it all works out. I am grounding the Fio board via a water pipe in the space - probably that helps as well….

Go Up