Hi,
A while back I was in a jam, and needed to come up with a way to locate underground cables fast. I had to improvise and make do with what I had.
I google search found information about a concept, in which you would inject an audio signal with a suitable carrier frequency into the cable, and then use a portable AM radio to locate the cable.
So, I modified a proto picaxe 18x board I had. Stripped off any unnecessary parts, and soldered in a fet driver. The fet driver became the "radio transmitter" in this. To generate the carrier frequency, I set up the pwm on the 18x to the maximum possible, giving a 1MHz carrier wave output. Good enough, as there didn't seem to be any other transmissions going on with that f. If there had been conflicts, a simple solution would have been to poke a new calibration value to the osctune register, and offset the carrier slightly.
To modulate the carrier wave the options with picaxe were quite limited, in fact the only feasible solution was just to toggle the pwm on and off as fast as possible. That ended up being about 1kHz. To help follow the signal with the radio, there are 2 glitches to the audio, one with the loop re-starting, and the other in the led blink (300ms pause in the tone). Those make it quite easy to tell apart from static.
I tried various ways of connecting the device. Maybe the best results came by grounding the other lead with an electrode, and connecting the other lead to the "hot" wire. Simply floating the other wire also worked surprisingly well. The caps allow connecting to a live wire as well. If you want to try this, I suggest building it into a sealed plastic box, so none of the internal conductive parts can be accidentally touched.. should the caps ever fail(Y2 rated should not, but one can never be too careful..) Safest way is to use only on wires that have been separated from the breaker panel.
It was quite easy to follow up the whole lenght of 300m of the cable, except for the first 10 or so meters, where the signal was so strong it was swamping the radio. Further than that, and it was easy to pinpoint the cable to about 40cm accuracy. Depth of the cable varied from 70cm to 1 meter.
Attached are the (dirty) program, and a block diagram type schematic.
A while back I was in a jam, and needed to come up with a way to locate underground cables fast. I had to improvise and make do with what I had.
I google search found information about a concept, in which you would inject an audio signal with a suitable carrier frequency into the cable, and then use a portable AM radio to locate the cable.
So, I modified a proto picaxe 18x board I had. Stripped off any unnecessary parts, and soldered in a fet driver. The fet driver became the "radio transmitter" in this. To generate the carrier frequency, I set up the pwm on the 18x to the maximum possible, giving a 1MHz carrier wave output. Good enough, as there didn't seem to be any other transmissions going on with that f. If there had been conflicts, a simple solution would have been to poke a new calibration value to the osctune register, and offset the carrier slightly.
To modulate the carrier wave the options with picaxe were quite limited, in fact the only feasible solution was just to toggle the pwm on and off as fast as possible. That ended up being about 1kHz. To help follow the signal with the radio, there are 2 glitches to the audio, one with the loop re-starting, and the other in the led blink (300ms pause in the tone). Those make it quite easy to tell apart from static.
I tried various ways of connecting the device. Maybe the best results came by grounding the other lead with an electrode, and connecting the other lead to the "hot" wire. Simply floating the other wire also worked surprisingly well. The caps allow connecting to a live wire as well. If you want to try this, I suggest building it into a sealed plastic box, so none of the internal conductive parts can be accidentally touched.. should the caps ever fail(Y2 rated should not, but one can never be too careful..) Safest way is to use only on wires that have been separated from the breaker panel.
It was quite easy to follow up the whole lenght of 300m of the cable, except for the first 10 or so meters, where the signal was so strong it was swamping the radio. Further than that, and it was easy to pinpoint the cable to about 40cm accuracy. Depth of the cable varied from 70cm to 1 meter.
Attached are the (dirty) program, and a block diagram type schematic.
