Coil resonance vs performance

[bugwhiskers]

Will the detector be enhanced if the coil is pulsed at it's natural resonant frequency ?

Will the xmit power be improved or stay the same if you increase the coil drive pulse width beyond the "flat top" of the MOSFET's protection diode ?

[bbsailor]

It's all about what happens between the pulses.

A pulse induction coil that operates at the coil's natural self resonance (between 500Khz and 1000 KHz) will only have the pulses spaced one or two microseconds (uS) apart. All the processing would need to occur in that space. This space is too short to stimulate a target with enough energy to cause eddy currents to be generated in the target and also have enough time to process the receive signal. Also, once a damping resistor is added across the coil it stops being a resonant circuit and will have a certain bandwidth based on the value of the damping resistor.

Even continuous wave metal detectors need to operate in the VLF range 1KHz to 20KHz) to penetrate the ground enough to obtain a response from a metal target.

Go to the following web link for an excellent patent explaination of pulse induction coils, damping resistors and sensitivity to small objects. http://www.freepatentsonline.com/7075304.pdf.

The two paragraphs below, taken from that patent, provide a good description of PI coil bandwidth and damping to make a sensitive coil for small objects.

"[0007] The induced eddy currents in a metal target are proportional to the change in magnetic field with time (.DELTA.B/.DELTA.t) at the metal target location. For high sensitivity, one would like to have dB as large as practical and .DELTA.t (the change in time) matched to the metal object's time response (bandwidth). For a small metal object with a fast time response (high bandwidth) the optimal detector sensitivity would be achieved with a small .DELTA.t matched to the small metal objects response (matching bandwidth of sensor and target). For a large metal object with a slower time response the optimal detector sensitivity would be achieved with a larger .DELTA.t matched to the metal object's time response. The magnetic field (B) is proportional to the current (I) in the transmitter coil and the number of coil turns (N), thus B.about.IN. More coil turns (N) increases the magnetic field at the target depth for a fixed current. However, increasing the number of coil turns also increases the kick-back voltage across the transmitter coil and switch due to the increased inductance. The voltage across the transmitter coil and the electronic switch turning off the coil current is V=Ldi/dt and L.about.N.sup.2. More coil turns also increases the capacitance C of the coil due to the potential (voltage) differences that exist between the individual turns of wire which makes up the coil.

[0008] Consider the transmitter coil. The same effects apply also to a receiver coil that is being excited by a transmitter coil. At the moment of current change in the transmitter coil, a high voltage appears across the coil. A fixed shunt resistor R is typically placed across the transmitter coil to dissipate the current in the coil. The resistor is called the damping resistor since it is used to dampen or suppress coil oscillation caused by the LCR circuit formed by the coil. The larger the shunt resistor, the greater the current dissipation and the faster the current decay. Fast current decay allows for small metal targets to be more easily detected since the coil has a higher bandwidth. If the damping resistor value is set too high, [my spelling and grammar correction added] current is forced into the coil where the capacitance and inductance combination causes voltage/current oscillations: the oscillations will mask small metal target signals. A small damping resistor slows down the coil decay and lower the sensitivity of the coil to small metal targets. Controlling the damping resistor effects the performance of the PIMD."

The value of the damping resistor is critical when attempting to operate at very low delays. The coil takes about 5uS after current shut off to spike and decay to zero plus a few more uS for the front-end amplifier to stabalize. Once it reaches zero at the output of the front-end amplifier, because of the amount of energy in the flyback spike, the signal wants to oscillate, like a ball being dropped from some height. The damping resistor works like putting a pillow under the ball to dampen it's bounce and attempt to quicken the front-end amplifier's recovery time. As you reduce the delay setting on your PI machine, you are sampling closer and closer to the bounce point. Once you get to the point where you are starting to sample the tail-end of the bounce, the PI machine will lock-up, indicating that you have reached the natural delay of that coil. When the MOSFET "flat-tops" it starts to conduct, generates internal heat, and tends to exted the time it takes for the signal to reach a stable sampling point. Anything that adds capacitance to the coil circuit such as: amount of coil turns, wire insulation dielectric, shield spacer dielectric, shield spacer thickness, the shield itself, MOSFET output capacitance and the type and length of the coax used, will tend to slow down the potential performance of the PI machine.

Coils with higher self resonant frequencies require less damping, (larger damping resistor values) and allow a little closer sampling to the the point where the signal stabalizes at zero (or the impact location of the ball analogy). Some desired targets, like gold, have very fast decays and require sampling at the earliest possible time to obtain the strongest response from a fast-decaying target signal.

The referenced patent reveals the inventors use of a MOSFET to act as a variable damping resistor to optimize the coil's bandwidth and ultimate performance in detecting specific targets. The inventor describes the use of a computer to control the voltage on this damping MOSFET which controls it's resistance in an attempt to optimize the signal form an unknown target. Once the signal is optimized, the computer then can identify the nature of the target by comparing the optimized target response from a library of target responses stored in the computer. This is all an attempt to recover additional information from the target to aid in it's identification. PI machine designs that integrate the samples over some period of time (a few ms) loose the ability to extract this additional target intelligence.

This long-winded explaination should help you better understand what is happening between the pulses.

bbsailor