Interesting coil design article

[bklein]

I touched on some of this in the Litz wire discussion but this is worthy of a separate thread I think...

Take a look at this article.
http://www.sensorsportal.com/HTML/DIGEST/december_2010/P_SI_122.pdf

I'd like your thoughts as to what fundamentals apply to a PI coil design if you just focus on the RX function of the target signal. (That is, ignore the fact that just moments before there was a DC current flowing through this same coil and a huge back EMF was likely created. I don't want to hear about it. I want to ignore it. I'll deal with it later!)
I'm trying to get at how PI coil designs are developed from scratch.
Say you want a 9" or a 5" coil... What next?
Wind as many turns as it takes to match the Surfmaster PI coil inductance? Why, are you sure? Will more turns give you better sensitivity? Why not. Worried about current through small wire guage? Why? - you're feeding into a high impedance preamp if using voltage mode.
Sorry, my brain just hasn't grasped this all yet. I understand the build techniques to minimize winding movement and capacitance but I am confused as to where number of turns stops being productive. This article points out 3 different approaches for LF and HF frequency ranges. Its throwing a lot at me - a lot to absorb - but I'm sure it pretty much has a lot of good fundamentals in there.

Barry

[Carl-NC]

If you are talking about a mono coil, it's impossible to separate the TX from the RX. They have different requirements, so the coil design is a compromise. Examples:

For the TX, field strength is proportional to N for a given current, so you want as many turns as possible.

But current is inversely proportional to N due to wire resistance, so fewer turns are better for higher current.

But higher current burns through batteries faster.

Smaller wire is better for self-capacitance and faster sampling. But it has higher R which hurts speed.

Larger wire is needed for current capacity, but slows things down due to higher C.

For the RX, coil should have lots of turns to produce a stronger target signal. But this also produces a stronger ground signal, and a stronger EMI signal.

Generally, sampling speed is the main factor in the coil design. Peak current is secondary, everything else tends to fall out.

[Woody.au]

My experiments point to tin plated copper wire being a faster coil than a litz wire coil. Maybe the insulated strands on litz cause inter wire capacitance that slows the current.
I have not noticed this effect on small RF coils but only large detector coils.

[ghaj]

If you are talking about a mono coil, it's impossible to separate the TX from the RX. They have different requirements, so the coil design is a compromise. Examples:

For the TX, field strength is proportional to N for a given current, so you want as many turns as possible.

But current is inversely proportional to N due to wire resistance, so fewer turns are better for higher current.

But higher current burns through batteries faster.

Smaller wire is better for self-capacitance and faster sampling. But it has higher R which hurts speed.

Larger wire is needed for current capacity, but slows things down due to higher C.

For the RX, coil should have lots of turns to produce a stronger target signal. But this also produces a stronger ground signal, and a stronger EMI signal.

Generally, sampling speed is the main factor in the coil design. Peak current is secondary, everything else tends to fall out.

Dear Carl,
I am trying to make a SD-2000 board based on ZED circuit. Unfortunately I don't access SD-2000 loop. So I am forced to make one myself. But I don't know loop spec like number of turns, diameter of the loop (I guess it to be 18"), size of wire used, and so on. I just know that loop resistance and inductance must be 0.4ohm and 300uH. Do you know anything about the mentioned spec and how to get the required ohm and uH of loop? Tnx.

[IBGold]

Hi Ghag,

You require either Tinned or Litz wire in the range of 32/.2 to 36/.2 or equivalent AWG with strands no thicker than .2mm for your TX or Mono coil if using Tinned wire try to get Marine wire with polyolefin insulation.

Regards, Ian.

[ghaj]

Hi Ghag,

You require either Tinned or Litz wire in the range of 32/.2 to 36/.2 or equivalent AWG with strands no thicker than .2mm for your TX or Mono coil if using Tinned wire try to get Marine wire with polyolefin insulation.

Regards, Ian.

Hi Ian,
1- what about diameter of coil?
2- No. of turns (Number of windings I mean)?
3- Can I make the loop out of normal wire with equal sectional area? (I mean A=32*0.2*0.2*pi/4=1mm2 ==> AWG17)
4- Are wire gauge and No. of windings the same for both monoloop and Double-D coils?
Thanks a lot.