Originally posted by kt315
View Post
Damping resistors work by suppressing the flyback pulse oscillation as soon as possible. These oscillations occur because the energy in a charged coil by the TX pulse current at turn off time creates a flyback pulse that can peak near 1000V for high power PI designs. Add to this flyback model, the capacitance added by:
1. Coil turn-to-turn wire capacitance
2. MOSFET COSS (output capacitance)
3. Coax cable capacitance (about 30 pf per foot)
4. Coil to shield capacitance
The rule of thumb is to keep the coil alone capacitance about one half the total measured capacitance at the end of the coil cable. This is done by measuring the self resonant frequency of a known coil inductance and then doing the math.
Then you have some collective value of capacitance that tends to absorb the flyback energy and causes that flyback energy to oscillate. Lower total capacitance has less energy to oscillate and requires less damping so a higher value damping resistor can be used. The sooner the TX pulse oscillations can reach near zero volts, the faster the RX circuit can be turned on. This is called critical damping. During this delay time any eddy currents remaining in the stimulated target are decaying. If you have a small target with a 2 uS time constant, in 5 time constants of that target's charged eddy currents will be at zero and have little or nothing to detect at a 10 uS delay. If you could drop the delay down to 7 or 8 uS you could begin to detect that target with some remaining eddy currents.
As you make your delay lower other factors begin to become involved.
1. Operational amplifier time to come out of saturation. Lower gain multiple stage op amps work faster than single stage high gain op amps.
2. Sensitivity to detecting environment. Wet beach hunting is typically done at about 15 uS delay or higher. Shielding a coil isolates the coil from ground capacitance but also adds some additional coil capacitance.
3. External noise at some multiple of local AC current frequency: 50Hz in some countries and 60Hz in some other countries requires a shielded coil.
4. Coil wire itself (including wire soldered joints). How long are eddy currents being retained by the coil wire itself. If your wire holds the eddy currents any longer than the minimum delay, the wire will act as a target.
5. Full stimulation of the target: Theory tells us that the TX pulse discharge time constant (TC) should be 5 times faster than the desired target TC to fully stimulate it. This coil discharge TC is measured by coil inductance divided by the damping resistor value. If a 300 uH coil has a damping resistance of 750 ohms than the discharge TC will be 300/750 or .4uS and would be good to stimulate a target with a TC of 2 uS (.4 X 5).
If you build a coil and can detect small gold and when you reduce the delay and suddenly your PI machine locks up, you are beginning to detect something in the coil itself.
As you can see, when you attempt design a coil to reduce the delay to detect smaller targets, other factors come into play. It is a real balancing act. Choose your battles wisely!!!
Joseph J. Rogowski
P.S.
In keeping with the topic: where is the best location for the damping resistor. It depends on how many coils you plan to use and the range of damping resistor values that will critically damp each coil. If you have a range of coils that will critically damp from 800 ohms (faster) down to 600 ohms (slower coil) put a metal film (1Watt to 3Watt) 800 Ohm damping resistor on the PI circuit board. Make sure that it is not getting too warm as its resistance will get a little higher when warm. Then, when using coils that critically damp at lower resistance values you can add a smaller resistor in parallel to the circuit board value by locating it in the coil connector. Size is critical but since its value will be near 2500 ohms in parallel with 800 ohms its power absorption will be much less and you could get away with using a fraction of a Watt resistor. Use a metal film type.
Leave a comment: