Educate me please

[kernels]

Hi gents,

I have been playing around with building a PI detector for a bit of fun, lots of schematics and info around the place.

I can't seem to find as much information on the reasons behind some of the specs given for coils, would appreciate you gurus enlightening this youngen.

1. Why is the inductance always around 300 to 400uH ? My best guess: Eddy currents induced in target objects when the main magnetic field is generated (or collapses ?) only "live" for a short amount of time, if the inductance is too high, the current keeps going too long after you turn off the pulse, which makes the target object invisible because its eddy currents had already disappeared ?

2. How can you shield a PI coil without the shielding material looking like a target object.

3. Does the resistance of the coil matter ? It will affect the time constant of the inductor as well as the maximum current that will flow through the coil, nothing else ? So you could use a super low resistance coil as long as your coil has enough inductance that the current never damages the fet because the current is turned off just as you reach the desired peak current ?

4. Does the current matter ? Are you aiming for the biggest magnetic field possible ? (more current and more windings give you a bigger field, but more windings also equals more inductance and slows everything down ? )

5. What is the problem with very low inductance coils ?

Appreciate any info.

[kt315]

Why is the inductance always around 300 to 400uH
===
last detector of Eric has 1mH coil.

[kt315]

2. How can you shield a PI coil without the shielding material looking like a target object.
===
answer in his 1 question. lowest coil resistance gives the shield it-self.

[lucifer]

Hi,
1. You can have a coil with any inductance, from a single turn coil to the millihenry range. However, increasing the coil turns (hence the inductance) adds parasitic capacitance as well. More turns and more parasitic capacitance make the coil slower which means it may not be sensitive enough for all targets. It seems that a 300-400uH coil is good for a wide range of targets.

3. Coil resistance (along with supply voltage) determines the peak current. It also determines how fast the current rises in the coil (L/R) during TX and how long you need to keep your MOSFET turned on. You need to keep it "on" long enough so that the current in the coil has reached >90% of the peak current.

4. More current means stronger magnetic field around the coil. But don't be fooled that doubling the current will double the detection depth, depth will only increase about 2%. In addition higher currents might cause thermal and other problems.

5. Low inductance coils are less sensitive. The more turns the coil has, the more sensitive it is. And increasing the turns too much will cause the coil to pickup a lot of noise and ground signal.

Cheers!

[kt315]

i offer to read ITMD before the questions getting up. also, there will be new release of the book, Carl has reported.

[waltr]

5. Low inductance coils are less sensitive. The more turns the coil has, the more sensitive it is. And increasing the turns too much will cause the coil to pickup a lot of noise and ground signal.

Also a typical PI uses a single coil for TX and RX. This is another place to do trade-offs.
More turns = more RX sensitivity but has downs sides as per #1.

Carl did an analysis in the post some years ago of more verse less turns. After factoring in all the trade-offs 300-500uH is the sweet spot.

3. Coil resistance (along with supply voltage) determines the peak current. It also determines how fast the current rises in the coil (L/R) during TX and how long you need to keep your MOSFET turned on. You need to keep it "on" long enough so that the current in the coil has reached >90% of the peak current.

I did a lot of experimenting with coil series resistance when building my HH2 (see: https://www.geotech1.com/forums/show...ake-on-the-HH2).
Minimum resistance did not obtain the best sensitivity (distance) or sampling soonest. Adding some R in series increased detection distance and allowing sampling sooner and lower peak coil current.

Then there was the Targets excitation and response, Found best to have a TX ON pulse that allowed the coil current to level off from about 5Tau of Target Tau.
Added series R helped get the coil current to near peak faster (L/R). All this then helped getting a fairly fast cycle time. My HH2 runs at 1600Hz. Faster cycle time improves the Integration stages obtain a larger signal.

So, long story is one must consider the entire PI detector circuit from TX to Coil to RX through to Audio. Then decide what trade-offs obtain the best result for the type of detecting you are going to do. Is it small gold nuggets (very low tau) or silver coins (higher tau).

[kernels]

Great answers, thanks gentlemen, I will keep playing around, I moved from electronics to software a decade or so ago, so really enjoying getting my hands on the scope again.

Last question, is there a book that delves into the physics of the PI detector, I feel like looking at the electronics I am still one level away from REALLY understanding what is going on at magnetic and electric field level.

Cheers

[Carl-NC]

[Carl-NC]

Also look at this thread:

https://www.geotech1.com/forums/show...rget-Responses