Wide Area Loop Coil Design

How close to a (perfect) circle do you need?

There's a firm here that regularly uses such large coils to probe geology.

I'd consider thinking about the structures in a regular "umbrella" for inspiration. The number of ribs you choose would determine the closeness of a "perfect" circle your result. (i.e., make each chord flex at it's center so it can fold in on itself -- like an umbrella)

You've not indicated how rigid the coil itself needs to be. I.e., could a form that holds it taut between a small number of points be adequate? Does the coil need to be self-supporting? etc.

You might just say what you are trying to do. Someone might have a suggestion of doing something different , but getting the same result. It kind of sounds like you are wanting to make a cave radio. Or maybe see what music is good for growing roses.

Dan

I would like to make a wide area circular loop of approximately 3 metres diameter.

It will be fed with audio sinewaves via a 100W (per channel) car stereo amplifier.

Could I please have some suggestions on:

1. The best way to construct this so it would be moveable. I also need it to function as a former into which to loose wind the coil.

1. Given the continuous signal, a suitable resistance that would not ooverload the 8 ohm output of the amp. I would prefer to have all the load resistance in th coil and not use a series resistor.

2. What would the minimum suitable wire gauge for this current?

Thank you,

Mark Granger

You can use a loop of square steel tubing (or other metal), as long as the loop is broken at some point by an insulator. This is how we construct DF antennae for up to 6m (50MHz) band.

Why not use a standard, jacketed, multiwire cable?

It sounds like you need a Helmholtz coil.

Joe Gwinn

A hexagon is a much better approximation to a circle, and not much more work.

Two hexagons, one above the other in a Helmholtz configuration - as suggested by Joe Gwinn - would give you a bigger sweet spot in which to place the device you want to calibrate.

For 1 m diameter MF/HF loop antennas I have used Hula-Hoops For 3 m diameter a loop made of electric conduits might be usable.

Use a multicore cable and connect each cable conductor in series (off by one) using a multicore (flat)cable with off by one construction to get everything in series.

Such constructions will have quite high interwound capacitance, so make sure you work below the natural resonant frequency.

A sufficient number of turns should solve this problem (remember self capacitance at least for 10+ kHz).

In response to the questions so far, I would like the coil to be as round as possible, and capable of being picked up moved intact. It does not need to fold up or be portable. Just slid on the floor.

My first thought was PVC conduct bent into a circle, and then to nest the wire inside through a slot in the outer edge. The PVC couplings between lengths would also need to be slotted.

A similar idea was to use black irrigation tubing, which is available in continuous lengths. But it tends to kink easily.

I liked the suggestion for 5 turns of 24 gauge wire. However, I would prefer to use enamelled winding wire instead of stranded. I assume the result would be the same.

Any further suggestions would be much appreciated.

The coil is intended for calibrating a magnetic sensor.

Mark Granger.

use a flexible former, possibly a square or hexagnal former that is hinged at the corners. for portability fold it so all the segemnts are parallel

alternatively flexible a circular former can be tripled over itself forming a 3 layer 1m diameter bundle, possibly you could use the back part of a square PVC conduit welded into a hoop shape, the stuff comes in 6M lengths IIRC, so you'd be a little under 3M diameter

Any thickness will work. such a loop has enough surface area to dissipate 100 real watts of power as heat, it will have no trouble with 100 car-audio watts.

The thicker the wire you use the more turns it will take to achiieve 8 ohms resistance, and consequently the more powerful the magnetic field from the coil will be. the limiting factor here is your budget.

the stuff used for water supply has thicker walls and is more resistant to kinking, it seems to come off the roll with a native curvature near 6m diameter but can be formed into tighter curves (eg: hula hoops).

There is an eqivalent enamelled wire available, they might call it

Octagons if you're using PVC pipe fittings, 45 deg joiners are available.

Helmholtz coils wound in opposition give you a very good constant field *gradient* along the axis of the coils. With the field at the mid plane being zero... or whatever the local B field is.

What kind of magnetic sensor are you calibrating that you need such big coils? Is it a big sensor? what kind of sensitivity? Earths field? ~500mGauss.

George H.

ARRGGG!!! not a single 'original' posting by OP appears in my news server!! eternal-september

Calibrate a magnetic sensor? just how BIG is that sensor? If your construction yields better than 1% absolute accuracy, kudos. You want accuracy? you're going to have to monitor current accurtely AND construct with some tight tolerances AND keep it off the floor [concrete usually has rebar in it. AND keep it away from conductive/magnetic items - file cabinets, steel chairs, aluminum studs, you'd be surprised at how much that stuff can modify your fields.] With a little effort you should get tobetter than 0.1%

Helmholtz coils hold their 'uniform' field to within 1% for about 1/3 the space inside the coils. Soa 3m Diameter will give you 1% over the volume of 1m. You can confirm the relationship between ANY construction you intend and get incredible accuracy using femm 4.2, which is free. Yeah, that 'uniformity' caught me off guard too. There is another coil shape that provides a bit more uniform field. It's a cylindrical shape where all the coils are same diameter, but coils with specifc number of turns are spaced down the cylinder at specific locations. Increments kind of like a 'physical' ButterworthTchebyshev filter where the field is more 'uniform' over a larger volume BUT has plus/minus error to it.

Using femm, you'll find out all kinds of nuances that irritate, like the fact that the larger wire is no longer where you 'thought' it was, it is now distributed over the coil's cross section.

For Mike, as I said I can't see any of your original postings, contact me OFFLINE and I'll send you some femm models of Helmholtz coils. That way you can hit the ground with your feet running, so to speak, shorten that learning curve.

Hi Robert, Well there is one throw away line about sensor calibration.

Here's a copy,

My first thought was PVC conduct bent into a circle, and then to nest the wire inside through a slot in the outer edge. The PVC couplings between lengths would also need to be slotted.

A similar idea was to use black irrigation tubing, which is available in continuous lengths. But it tends to kink easily.

I liked the suggestion for 5 turns of 24 gauge wire. However, I would prefer to use enamelled winding wire instead of stranded. I assume the result would be the same.

Any further suggestions would be much appreciated.

The coil is intended for calibrating a magnetic sensor.

Mark Granger.

I might have to play around with femm 4.2 some day. We make 12" diam. Helmholtz coils, and need ~ 10^4 uniformity over about the central cubic inch. So a 5 G field and less than 1 mG linewidths. After making ~100 or so, there were ~10-20% of the coils that were not as good as needed. After much futzing around, I found that I could get them back on spec by tweaking the angle between the two coils. (A few shims, 10-20 mil, in one of the rods that sets the coil separation distance.) It would be interesting to know how much a translation in the x or y direction, (with Z being the axis of the coil) would give a similar error.

George H.

Use a cable with multiple conductors and connect them as a coil at the join.

The thick walled black (polyethylene?) agricultural irrigation pipe does not kink.

On a sunny day (Sat, 28 Mar 2015 06:56:01 -0700 (PDT)) it happened George Herold wrote in :

The common way is to tune the coil for the frequency used.

amp - R -- L -- C -- coil --- ground

If it is an 8 Ohm audio amp, then make R 8 Ohm, tune for resonance, now you have a stable system.

L + Lcoil and C and you know that formula I hope, some sing wiz 0mega,

It is no good to connect the amp directly.

You can see the band with from that other formula, B = O Mega / el or some sing.

J.P. Weapons of Math Destruction

yeh, something like a standard ethernet cable

-Lasse

Or something with fatter conductors, if he wants a lot of field. 10x #16, something meaty like that. He probably wants something like 8 ohms total to optimize field strength per watt.