Optimizing PI Coils

[Carl-NC]

If this is a bench-only coil I would not worry too much about self-eddies. They will have a very minimal effect on the turn-on current, and will have a constant effect on the turn-off flyback. Where they cause havoc is in the flyback over mineralized ground.

[green]

Green,

What is the inductance of the TX coil.
What is the total resistance of the TX coil drive circuit, including:
1. Coil resistance
2. MOSFET coil driver on resistance
3. Any series resistance in the coil circuit

With this information we can see the coil current rise Time Constant.

Thanks
Joseph J. Rogowski

4.4 ohms, 325uH

[green]

If this is a bench-only coil I would not worry too much about self-eddies. They will have a very minimal effect on the turn-on current, and will have a constant effect on the turn-off flyback. Where they cause havoc is in the flyback over mineralized ground.

I think lead shot 18 is balanced. Lead shot 17 left recordings aren't quite balanced, right side is balanced. At less than 2us, signal goes opposite polarity (X signal ?) as it does after 2us (R signal ?). I'm thinking coil wire is being seen as a target at less than 3.5us if not balanced. Think I could reduce the delay to maybe 2.5us if both ends look similar but opposite(balanced ?). What could be causing the problem if it isn't the coil wire being seen as a target? Maybe not important but reducing delay from 3.5us to 2.5us would increase the signal for #9 lead shot 4 times. Just trying to learn something.

Some one replied in another forum that he could detect #9 lead shot at .5inches with a GPX 6000 with a 11 inch mono coil. Think this must be all physics, I'm missing the part that allows detecting #9 lead shot at .5inches with a 11 inch coil.

[green]

A test with a couple spiral wrapped coils I have (no lead wire or shielding). Was expecting different results. Wasn't thinking the AWG19 coil would be similar to the AWG28 coil. Was thinking of ordering some AWG26(7/34) to try. Know I wonder if it would allow a lower delay. Was thinking wire time constant had something to do with the decay, looks like coil time constant not wire time constant. Where's my thinking wrong?

[bbsailor]

A test with a couple spiral wrapped coils I have (no lead wire or shielding). Was expecting different results. Wasn't thinking the AWG19 coil would be similar to the AWG28 coil. Was thinking of ordering some AWG26(7/34) to try. Know I wonder if it would allow a lower delay. Was thinking wire time constant had something to do with the decay, looks like coil time constant not wire time constant. Where's my thinking wrong?

Green,

I am not sure that what I will say next will answer your question but you should consider that the TX pulse rise time constant is more horizontal than vertical when the TX pulse turns off.

Your 325 uH coil with a 4.4 ohm total resistance represents a 73.8uS TC to raise to 63 percent of maximum. At 2TC or 147.6 uS it will raise to 85 percent of max and at 3TC or 221.4 uS it will raise to 95 percent of max. When the current raises to 3TCs the current rise is almost horizontal and any current rise eddy currents in targets should have died down so that when the TX pulse turns off, then the energy in the discharge pulse can stimulate your targets. The more vertical the discharge slope will better stimulate smaller, lower TC targets.

What is your TX pulse width?
At what coil rise TC are you turning off the TX pulse?
What is you pulse frequency?
What is your delay time from TX turn off to RX turn on?

If you look at most mono coil PI metal detectors, they mostly wait to about 3 TX current rise TCs to turn off the TX current. Maybe Carl could jump in and further explain this standard.

Joseph J. Rogowski

[green]

Green,

I am not sure that what I will say next will answer your question but you should consider that the TX pulse rise time constant is more horizontal than vertical when the TX pulse turns off.

Your 325 uH coil with a 4.4 ohm total resistance represents a 73.8uS TC to raise to 63 percent of maximum. At 2TC or 147.6 uS it will raise to 85 percent of max and at 3TC or 221.4 uS it will raise to 95 percent of max. When the current raises to 3TCs the current rise is almost horizontal and any current rise eddy currents in targets should have died down so that when the TX pulse turns off, then the energy in the discharge pulse can stimulate your targets. The more vertical the discharge slope will better stimulate smaller, lower TC targets.

What is your TX pulse width?
At what coil rise TC are you turning off the TX pulse?
What is you pulse frequency?
What is your delay time from TX turn off to RX turn on?

If you look at most mono coil PI metal detectors, they mostly wait to about 3 TX current rise TCs to turn off the TX current. Maybe Carl could jump in and further explain this standard.



Joseph J. Rogowski

Tx pulse width_160us
Pulse frequency_1006 Hz
What is your delay time from TX turn off to RX turn on? Looking at amplifier out, so no delay time except maybe the amplifier.

Wondered why I didn't see AWG19 decay (2.49us time constant) this morning. Tried again, I see the decay if the coil is closer. Don't know what I am doing wrong, maybe another stupid pill.

top left picture_no target

[green]

Tx pulse width_160us
Pulse frequency_1006 Hz
What is your delay time from TX turn off to RX turn on? Looking at amplifier out, so no delay time except maybe the amplifier.

Wondered why I didn't see AWG19 decay (2.49us time constant) this morning. Tried again, I see the decay if the coil is closer. Don't know what I am doing wrong, maybe another stupid pill.

top left picture_no target

[lead shot21.png] The 6inch piece of AWG19 has a tau of 2.5us, same as calculated. The AWG19 coil with Rd=1180 (highly overdamped) has a tau of 2us, why is the coil decay different than the 6inch piece of wire?

[lead shot22.png] rows 2, 3 and 4 are the coil I'm using now. Call it Rx 133mm diameter (inside diameter because spiral coil calculator uses inside diameter). Want to try different size wire, so thinking for comparison mean diameters should be the same. Rows 6 an7 are the figure8 Rx coils. Rows 8 or 9 are the Tx coil. Row 8 is AWG30 magnet wire, row 9 PVC coated stranded. Should I expect the coil resonance (calculated capacitance) to be similar to the AWG28 magnet wire coil or quite different? Will the new coils allow a lower delay before sampling? Should I be trying something different? Goal is to detect #9 lead shot.

[Carl-NC]

What do you mean "The AWG19 coil ... has a tau of 2us?"

[green]

What do you mean "The AWG19 coil ... has a tau of 2us?"

Different lengths of AWG19 wire chart 2.5us decay. When I hold the spiral wrapped coil with 1180 ohms across it as a target and charted the decay it was around 2us.

[Carl-NC]

Probably the signal strength of the coil overwhelms the signal strength of the wire eddies.

[green]

Probably the signal strength of the coil overwhelms the signal strength of the wire eddies.

That could explain the test coil needing to be closer to see the wire eddies? At longer distances wire eddies aren't stronger enough to see when the coil has decayed, coil decays faster than wire eddies for AWG19. Is there a ratio for wire eddies time constant to coil time constant when selecting wire size for winding a coil PI coil? If mono coil is 300uH and Rd is 1000, what should be the maximum wire size? Valid question or am I thinking wrong?

[green]

That could explain the test coil needing to be closer to see the wire eddies? At longer distances wire eddies aren't stronger enough to see when the coil has decayed, coil decays faster than wire eddies for AWG19. Is there a ratio for wire eddies time constant to coil time constant when selecting wire size for winding a coil PI coil? If mono coil is 300uH and Rd is 1000, what should be the maximum wire size? Valid question or am I thinking wrong?

Wound one of the Rx coils with AWG30 wire. A chart with some spiral coils I have, shielding and added lead wire increase capacitance.

Still wondering if there is a way to determine minimum wire size when making a coil.

[green]

I noticed some variation in signal amplitude when testing some different pieces of #9 lead shot. Connected a smaller coil to get more signal and tested 19 different #9 shot. Amplitude varied typically less than 2mV when repeat testing the same piece. Didn't expect amplitude to vary as much as I saw, all shot came from the same shot gun shell.

[EL NINO]

Mr. Green, it would be interesting to consider the individual shots No. 9 on the exact weight ... and add to the comparison .., maybe some shots are out of the expected weight tolerance ..

....it can also be a mistake in uniformity in the production / casting / of some shots.

[green]

Mr. Green, it would be interesting to consider the individual shots No. 9 on the exact weight ... and add to the comparison .., maybe some shots are out of the expected weight tolerance ..

....it can also be a mistake in uniformity in the production / casting / of some shots.

Don't have a scale accurate enough to weigh #9 shot. Searched jewelers scales, .01gram resolution. #9 lead shot is about .049grams, maybe .01 gram resolution isn't fine enough?