Piezo Coaxial Cable as driveway sensor

Thread Starter

bradenjoseph

Joined Mar 6, 2016
13
Hi Guys - new to this forum and have a rudimentary understanding of electronics. I have considered several different ways to install a driveway sensor that's about 300 feet from the house. I don't want to use the Milton bell tubing because I'd have to take it up in winter (and it's $$), I am not keen on visible optic systems either. I looked at a giant metal detector, but then I found piezo electric coaxial cable, so I'm trying to work out the circuity with that.

I think I have it mostly figured out (probably lots of mistakes all over the place), but the main problem I'm having is that the cable generates an AC signal that I convert to DC and regulate with a LM7808, then send the signal up the driveway to the house where it triggers a BJT transistor to complete the DC circuit of the alarm bell.

The problem is, when the piezo signal is cut (i.e. I change the AC input to 0), there is still charge and voltage in the system after the wheatstone bridge which continues the input/output of the LM7808.

What am I doing wrong that the circuit won't re-open when the AC is turned off?

See attached two PDFs - one with 0V AC and one with 12V AC.


Thanks!
B
 

Attachments

SLK001

Joined Nov 29, 2011
1,549
Why can't you post regular .jpg files? I'm not about to download and view a .pdf file that I know nothing about its source.
 

Sensacell

Joined Jun 19, 2012
3,432
Your concept of the power levels available from this cable is flawed.

It's a capacitor, it will generate some voltage when compressed, but the impedance is VERY high, such that it cannot supply any significant current to a load. You will need a high-impedance amplifier to detect this signal- powered by an external power source source. The capacitors you have connected are enormous in comparison to the cable capacitance, the minuscule charge generated by the cable will be totally swamped by this capacitance.
 

Thread Starter

bradenjoseph

Joined Mar 6, 2016
13
Your concept of the power levels available from this cable is flawed.

It's a capacitor, it will generate some voltage when compressed, but the impedance is VERY high, such that it cannot supply any significant current to a load. You will need a high-impedance amplifier to detect this signal- powered by an external power source source. The capacitors you have connected are enormous in comparison to the cable capacitance, the minuscule charge generated by the cable will be totally swamped by this capacitance.
Thanks for your response. So I guess I'll go back to looking at the metal detector.

But to my original posting, what is causing the ac voltage to continue once the AC is set to zero??
 

Sensacell

Joined Jun 19, 2012
3,432
I have no idea what you mean?

Not enough information to formulate a meaningful response.
Post the data sheet for the piezo coax cable- maybe we can help with a simple circuit idea?
 

SLK001

Joined Nov 29, 2011
1,549
You could build a seismic detector. Take a 4" by 10' piece of PVC pipe and glue a cap on one end. Just inside the other end, mount a mylar membrane with a magnet afixed to the center. Wind a coil so that the magnet can move thru the center and feed the wires from the coil thru two holes in a second cap. Glue on the second cap and seal the thru holes with silicone to make the pipe water tight.

The tube acts as a seismic detector, with sound being guided down the pipe where it will vibrate the membrane/magnet, thus creating a signal. Properly constructed, these detectors are very sensitive and can pick up foot falls in their vicinity when buried (it basically is a giant microphone).
 

AnalogKid

Joined Aug 1, 2013
10,986
Not sure what you mean by metal detector, but a buried coil of wire like a traffic light sensor is reliable. Basically, it is the L of an L-C tank, and there are several ways that can be turned into a sensor. Google patents has a ton of information.

ak
 

Thread Starter

bradenjoseph

Joined Mar 6, 2016
13
Also, this one:
http://www.disensors.com/downloads/products/Piezo Polymer Coaxial Cable_176.pdf

It shows the applications ranging from airplanes to foot pressure of the cable. Perhaps I generated confusion. The piezo coax is the "AC"/nf cap at the top left of my drawing, the cable running up to the house would be whatever AWG i am going to need to carry sufficient current or voltage to close the alarm bell circuit.

Thanks
B
 

blocco a spirale

Joined Jun 18, 2008
1,546
Have you tried
Hi Guys - new to this forum and have a rudimentary understanding of electronics. I have considered several different ways to install a driveway sensor that's about 300 feet from the house. I don't want to use the Milton bell tubing because I'd have to take it up in winter (and it's $$), I am not keen on visible optic systems either. I looked at a giant metal detector, but then I found piezo electric coaxial cable, so I'm trying to work out the circuity with that.

I think I have it mostly figured out (probably lots of mistakes all over the place), but the main problem I'm having is that the cable generates an AC signal that I convert to DC and regulate with a LM7808, then send the signal up the driveway to the house where it triggers a BJT transistor to complete the DC circuit of the alarm bell.

The problem is, when the piezo signal is cut (i.e. I change the AC input to 0), there is still charge and voltage in the system after the wheatstone bridge which continues the input/output of the LM7808.

What am I doing wrong that the circuit won't re-open when the AC is turned off?

See attached two PDFs - one with 0V AC and one with 12V AC.


Thanks!
B
Your input circuit is completely over the top and unsuitable for this application. You're only looking for a pulse so there is no need to full-wave rectify, regulate and smooth the input from the cable, you're not building a DC power-supply and the piezo cable is nothing like an AC power source.

If you configure a 555 as a monostable and bias pin 2 at 1/2 the supply voltage, it will trigger on the negative cycle of the incoming pulse from the cable. You will need to protect the input from over-voltage by adding a couple of fast diodes to clamp the input to the power rails.

Incidentally, there is no "wheatstone bridge" in your design, but there is a bridge rectifier.
 
Last edited:

Thread Starter

bradenjoseph

Joined Mar 6, 2016
13
Have you tried


Your input circuit is completely over the top and unsuitable for this application. You're only looking for a pulse so there is no need to full-wave rectify, regulate and smooth the input from the cable, you're not building a DC power-supply and the piezo cable is nothing like an AC power source.

If you configure a 555 as a monostable and bias pin 2 at 1/2 the supply voltage, it will trigger on the negative cycle of the incoming pulse from the cable. You will need to protect the input from over-voltage by adding a couple of fast diodes to clamp the input to the power rails.

Incidentally, there is no "wheatstone bridge" in your design, but there is a bridge rectifier.

Do you mean something like this?
 

Attachments

blocco a spirale

Joined Jun 18, 2008
1,546
Not quite, you have tied pin 2 to 0V. Fix it at 1/2 the supply voltage by using 2x 47k resistors to make a voltage divider. I have no idea what the actual output of the piezo cable will be when a car drives over it so you will have to experiment with resistor values or install a preset pot to allow fine adjustment. The input to the 555 just has to drop momentarily below 1/3 of the supply voltage to trigger.
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
I have no idea what the actual output of the piezo cable will be when a car drives over it
I'm trying to get a feel for the likely output voltage pulse from a piezo cable as per the link in post #8, and the signal conditioning required.
Assume a 2m length.
That will have a capacitance of ~2000pF.
Assume a vehicle can exert a 1N stress (~10Kg force) on the cable via the driveway it should produce 20pC of charge. So the voltage produced would be 20pC/2000pF = 10mV. That's going to need amplification and some serious filtering to eliminate the inevitable noise pick-up from both the piezo cable and the 300ft run of connecting cable going to the house. Nearby lightning strikes could result in nasty voltages on the cable, so safety/suppression components would be advisable.
 

blocco a spirale

Joined Jun 18, 2008
1,546
I might have been over optimistic, I was expecting more than that, but I agree, a little amplification may be necessary; the OP may just need to do a little experimentation. The 300ft of cable problem can be effectively eliminated by moving the electronics to the end of the piezo cable.
 
Last edited:

Thread Starter

bradenjoseph

Joined Mar 6, 2016
13
I'm trying to get a feel for the likely output voltage pulse from a piezo cable as per the link in post #8, and the signal conditioning required.
Assume a 2m length.
That will have a capacitance of ~2000pF.
Assume a vehicle can exert a 1N stress (~10Kg force) on the cable via the driveway it should produce 20pC of charge. So the voltage produced would be 20pC/2000pF = 10mV. That's going to need amplification and some serious filtering to eliminate the inevitable noise pick-up from both the piezo cable and the 300ft run of connecting cable going to the house. Nearby lightning strikes could result in nasty voltages on the cable, so safety/suppression components would be advisable.
Alec
I'm trying to get a feel for the likely output voltage pulse from a piezo cable as per the link in post #8, and the signal conditioning required.
Assume a 2m length.
That will have a capacitance of ~2000pF.
Assume a vehicle can exert a 1N stress (~10Kg force) on the cable via the driveway it should produce 20pC of charge. So the voltage produced would be 20pC/2000pF = 10mV. That's going to need amplification and some serious filtering to eliminate the inevitable noise pick-up from both the piezo cable and the 300ft run of connecting cable going to the house. Nearby lightning strikes could result in nasty voltages on the cable, so safety/suppression components would be advisable.
Alec, here's what I'm looking at in real numbers:
Specs:
950pF/m
20pC/N

2 m = 1900pF

A car weighing an average of 4000lbs (1,814kg) would generate about 17,800N. 17,800N x 20pC/N (2e-11) = 355,863.71pC
So the voltage produced would be 1,900pF/355,863.71pC = 56nV
Same calcs with a 4536kg massive truck would produce 2.14mV

So I guess those are my ranges. I will definitely bury the electronics next to the cable for sure, and then separately deal with the voltage drop over the 300ft based on the wire I select.

Blocco, I will try playing around with the 555 as you suggest and let you know
 
Top