A
Anonymous
Guest
This one is for Eric and the other circuit guru's.
I have an interesting idea as to how to accurately identify iron with a PI. Start with a master oscillator and divide it by 7 in a first divider and divide the inverted oscillator signal by 9. Use the divide by seven output signal to transmit positive going pulses and the divide by nine output signal to transmit negative pulses from the same coil. The pulses would both be generated on the positive edges of of the dividers outputs. Due to its hysteresis, an iron target will have a non linear response. This will cause a beat frequency to be generated in the receiver equal to the difference of the two frequencies. eg: For an oscillator running at 10 KHz then the transmit frequencies are 1,111.1Hz for the divide by nine and 1,428.5 Hz for the divide by seven. Notice that the pulses cannot collide with the ratios shown. The beat frequency will then be 317.4 Hz. A band pass filter can be used to seperate the beat frequency in the receiver. The presence of the beat frequency would indicate an iron target. A conductive target such as a coin will provide very little, if any energy at the beat frequency. The amplitude of the beat frequency signal could be compared if required to the amplitude response of a standard PI sample taken from either the positive or the negative channel. The ratio would permit one to determine a conductive target with a small non linearity such as a coin right under the search-head or a piece of jewelry to a weak iron signal buried way down deep. Another interesting idea is to sum the beat frequency signal from the bandpass filter into the receivers audio. This will provide a dual audio response. The presence of the 317.4Hz beat tone would let you know that you were detecting iron. Hmmm, could this be improved upon? As is, this will be a most useful feature for a PI. I still get the feeling though that this is just scratching the surface of the problem! Let's get our thinking caps on and see where we can go from here. Any ideas? Dave. * * *
I have an interesting idea as to how to accurately identify iron with a PI. Start with a master oscillator and divide it by 7 in a first divider and divide the inverted oscillator signal by 9. Use the divide by seven output signal to transmit positive going pulses and the divide by nine output signal to transmit negative pulses from the same coil. The pulses would both be generated on the positive edges of of the dividers outputs. Due to its hysteresis, an iron target will have a non linear response. This will cause a beat frequency to be generated in the receiver equal to the difference of the two frequencies. eg: For an oscillator running at 10 KHz then the transmit frequencies are 1,111.1Hz for the divide by nine and 1,428.5 Hz for the divide by seven. Notice that the pulses cannot collide with the ratios shown. The beat frequency will then be 317.4 Hz. A band pass filter can be used to seperate the beat frequency in the receiver. The presence of the beat frequency would indicate an iron target. A conductive target such as a coin will provide very little, if any energy at the beat frequency. The amplitude of the beat frequency signal could be compared if required to the amplitude response of a standard PI sample taken from either the positive or the negative channel. The ratio would permit one to determine a conductive target with a small non linearity such as a coin right under the search-head or a piece of jewelry to a weak iron signal buried way down deep. Another interesting idea is to sum the beat frequency signal from the bandpass filter into the receivers audio. This will provide a dual audio response. The presence of the 317.4Hz beat tone would let you know that you were detecting iron. Hmmm, could this be improved upon? As is, this will be a most useful feature for a PI. I still get the feeling though that this is just scratching the surface of the problem! Let's get our thinking caps on and see where we can go from here. Any ideas? Dave. * * *
