Rudy
Admin
By popular request, I am making this a sticky.
Whoever made a claim that shorter pulse delays lead to more depth lacks an understanding of physics, geophysics and electronics.
Having said that, a detector capable of stably operating at shorter pulse delays will, everything else being equal, be more sensitive to small gold items and other so called "low conductors". Target signals are spread in time from the transmit pulse shut off. The signals will reach a peak sometime after the transmit pulse is turned off and this time is given by the geometric shape of the target, its mass and its admittance (a vector quantity combining its conductance and reactance). The signals from low conductors (including ground mineralization) are the first ones to decay after the transmit pulse is turned off. High conductors --such as silver-- peak later and will hang around a lot longer because their high conductivity do a better job of supporting Eddy currents for a longer time. Ideally, for maximum pickup, we would like the detector's sampling delay to not happen after the target's signal has totally decayed. One could envision the delay time acting as a sort of discriminator, progressively discriminating out lower conductors as we increase the delay time.
Now, we can't start sampling immediately after removing the transmit pulse. Inductors (which is what the coil is) resist an instantaneous change in current flowing through them and will respond with a very large transient voltage spike, trying to keep the current through them flowing. During this voltage spike the detector can't sample anything, it is blinded by said transient.
We may ask what determines the length of time this transient condition exists. Although there are things an engineer can do to minimize the length of time the transient takes e.g. clamping diodes, damping resistors, etc., the coil's inductance and parasitic capacitance play a large role. For depth and reception sensitivity, we want a large inductance, but how large is limited by things like coil wiring resistance, power source, design target pulse repetition rate, etc. Parasitic capacitance is something we can do without. Certainly the coil's self-resonant frequency needs to be dealt with, but in general, a larger than necessary capacitance is undesirable and may lengthen the time at which the detector is able to sample. In this regard it should be noted that --for the same inductance value-- a coil made from a spiral PCB trace will have a significantly larger parasitic capacitance than a well designed wire coil. This is plain physics and it's due to the much higher dielectric constant (er) of the PCB substrate (usually an FR4 type copper clad material), compared to enameled wire and air as the dielectric. A detector with a PCB coil can not in general be used at lower sampling delay values as one made with a well engineered wire coil. It does however have a cost advantage as PCB spiral coils can be manufactured in a batch process and each unit has a near identical behavior, cutting down on manual tuning.
Wether a particular ground/gain combination allows the delay to be set at 10µs is immaterial. It could well allow a delay of 15µs and pick up a small gold item that may not be "seen" by a detector with a longer delay, simply because the signal has peaked and has practically decayed to nothing before the longer delay machine gets to sample the signal. Perhaps this is the reason why you may have heard some people say the shorter delay machines go deeper?
As you said, pulse delay is but one parameter of a machine's performance. Filtering, good low noise amplifiers on the receive side, peak coil current (related to pulse width and pulse repetition rate as well as design constraints) are also important, as are others not even mentioned.
There is a lot of controversy over "Pulse Delay," and its importance to depth and small target sensitivity. Unfortunately, most people don't understand that a setting of 10us pulse delay, is very noisey and unsteady on wet ocean sand. So even though you may be getting more sensitivity to smaller targets, you might not hear them.
Now, move your coil from the wet sand and into the water and you will have to increase your pulse delay to at least 15us to smooth out the threshold. By the time you have reached a depth of five-feet in saltwater, you'll need a pulse delay of about 20us.
...
The pulse delay must be coupled with power to the coil, and circuitry that filters and reads the return signal. The performance of a machine is NOT based soley on pulse delay. That is my opinion, and why I think I would rather have a set pulse delay of 20us, than 15us.
...
Whoever made a claim that shorter pulse delays lead to more depth lacks an understanding of physics, geophysics and electronics.
Having said that, a detector capable of stably operating at shorter pulse delays will, everything else being equal, be more sensitive to small gold items and other so called "low conductors". Target signals are spread in time from the transmit pulse shut off. The signals will reach a peak sometime after the transmit pulse is turned off and this time is given by the geometric shape of the target, its mass and its admittance (a vector quantity combining its conductance and reactance). The signals from low conductors (including ground mineralization) are the first ones to decay after the transmit pulse is turned off. High conductors --such as silver-- peak later and will hang around a lot longer because their high conductivity do a better job of supporting Eddy currents for a longer time. Ideally, for maximum pickup, we would like the detector's sampling delay to not happen after the target's signal has totally decayed. One could envision the delay time acting as a sort of discriminator, progressively discriminating out lower conductors as we increase the delay time.
Now, we can't start sampling immediately after removing the transmit pulse. Inductors (which is what the coil is) resist an instantaneous change in current flowing through them and will respond with a very large transient voltage spike, trying to keep the current through them flowing. During this voltage spike the detector can't sample anything, it is blinded by said transient.
We may ask what determines the length of time this transient condition exists. Although there are things an engineer can do to minimize the length of time the transient takes e.g. clamping diodes, damping resistors, etc., the coil's inductance and parasitic capacitance play a large role. For depth and reception sensitivity, we want a large inductance, but how large is limited by things like coil wiring resistance, power source, design target pulse repetition rate, etc. Parasitic capacitance is something we can do without. Certainly the coil's self-resonant frequency needs to be dealt with, but in general, a larger than necessary capacitance is undesirable and may lengthen the time at which the detector is able to sample. In this regard it should be noted that --for the same inductance value-- a coil made from a spiral PCB trace will have a significantly larger parasitic capacitance than a well designed wire coil. This is plain physics and it's due to the much higher dielectric constant (er) of the PCB substrate (usually an FR4 type copper clad material), compared to enameled wire and air as the dielectric. A detector with a PCB coil can not in general be used at lower sampling delay values as one made with a well engineered wire coil. It does however have a cost advantage as PCB spiral coils can be manufactured in a batch process and each unit has a near identical behavior, cutting down on manual tuning.
Wether a particular ground/gain combination allows the delay to be set at 10µs is immaterial. It could well allow a delay of 15µs and pick up a small gold item that may not be "seen" by a detector with a longer delay, simply because the signal has peaked and has practically decayed to nothing before the longer delay machine gets to sample the signal. Perhaps this is the reason why you may have heard some people say the shorter delay machines go deeper?
As you said, pulse delay is but one parameter of a machine's performance. Filtering, good low noise amplifiers on the receive side, peak coil current (related to pulse width and pulse repetition rate as well as design constraints) are also important, as are others not even mentioned.