6

Visitors of two public exhibitions A and B are equipped with compasses.

Exhibition A:

I want to guide these visitors along a certain path throughout the exhibition by "overriding" the earth magnetic field so that they can follow there compasses "north" along the path. The path where this deviation should work on is no wider than 1.5 meters. This deviation can be turned on and off by a switch (thus can not be accomplished by permanent magnets).

Exhibition B:

I want to notify the visitors of events by generating a short magnetic pulse that will shortly deviate their compasses. This should work over a ground circle area of at least 1 meter radius.

Update 13th April: The direction of deviation is not of importance.

Annotation:

Update 13th April: The path will guide the visitor along a rectangular playing area of about 20mx20m. Due to the nature of the installations, the only options to install hardware are the sides and the floor (max. height 20mm).

Assumptions:

  • Local earth magnetic field strength is below 65 uT.
  • There are no other electromagnetic influences.
  • Height of compasses above ground are 2 meters maximum.

Research so far:

I built a small solenoid magnet with about 400 windings, 4.8V and iron core. Although getting up to 1400 uT at the magnet itself, it rapidly drops in the distance (between ~1/r^2 and ~1/r^3) to a level where it cannot be distinguished from the earth magnetic field; reaching 70uT as soon as 20cm.

In order to reach compasses in 2 meters height (200cm distance, 20cm*10) I would have to amplify the strength by a factor of 1000 (10^3, because of 1/r^3) in case I assume a Dipole (Wikipedia). I doubt this can be accomplished with my small self-built magnet.

What would be the most cost- and time-efficient way to accomplish exhibition A and B?

In particular I'm interested in:

  • how to simulate such an environment on a computer,
  • what kind of magnet type/layout is suitable,
  • can this be accomplished with off-the-shelf equipment,
  • are there any dangers for the visitors.

I'm happy to provide more details if needed.

InteractiveCube
  • 61
  • 2
  • On a sidenote: Normal drop for magnetic field of a solenoid (if I remember correctly) is ${1}/{r^2}$, so the result you're getting is to be expected. If I remember correctly, it's true for many simple magnetic field shapes. The best way to approach this would be to analyze magnetic fields and try to calculate shape which would drop off with a ratio different from ${1}/{r^2}$. There are some which drop off with rate of ${1}/{r}$, but I can't remember any right now. – AndrejaKo Apr 12 '12 at 19:54
  • I can't help but think that a magnetic field strong enough to affect a compass at 2 meters would erase credit cards at a closer distance. If your magnets are embedded in the floor then any woman that sets her purse down will be very disappointed (and mad). –  Apr 12 '12 at 20:22
  • @AndrejaKo - \frac and other multi-line TeX constructs don't work in comments. – Kevin Vermeer Apr 12 '12 at 20:51
  • @AndrejaKo: Any ideas on how to analyze those magnetic fields using a simulation program on a computer? – InteractiveCube Apr 12 '12 at 21:43
  • @DavidKessner: Good input, we wouldn't want a credit card eraser floor :) Any ideas on the limits of the magnetic field strength in order to avoid that? – InteractiveCube Apr 12 '12 at 21:43
  • @Kevin Vermeer The first frac worked fine for me, but it was a bit difficult to read. Are you sure it isn't just some random bug in my TeX? – AndrejaKo Apr 13 '12 at 06:06
  • @InteractiveCube Unfortunately, no. All we did on university was by hand. As for the credit card problem, it would be a good idea to actually read some magnetic stripe card datasheets to see at which levels will there be problems. If I remember correctly, there were some studies on that topic and the conclusion was that the field strength was supposed to be so high that it would be highly unlikely to find it anywhere in normal use. I'll see if I can find the numbers for that study. – AndrejaKo Apr 13 '12 at 06:10

1 Answers1

2

There are probably easier ways to do this than using real compasses. You can still make things which are 'compass like" in appearance. Infrared would be amongst the easiest. RF would also work.

But ...

A large coil which fits the path shape should do what you want. Large diameter is the key to large distance. If you want to cover an area of about 1 metre you could have a 1 metre diameter or edge coil either on / in the floor or above their heads. Magnetic field will be proportional to amp turns. More turns = less Amps needed for a given field strength.

Best result would be from a coil above and below or either side BUT you will get usable results with a single coil. Two coi;s allow you to achieve a uniform field across the space concerned. Searching for Helmholtz coil will give you many ideas - but there are other types that are suitable.

Wikipedia Helmholtz Coil

enter image description here enter image description here

  • Contours showing the magnitude of the magnetic field near the coil pair. Inside the central 'octopus' the field is within 1% of its central value B0. The five contours are for field magnitudes of , , , , and

More here ....... and here

Pictures galore - all linked to web pages.

Wow .........

It is widely acknowledged that the fields generated by such arrangements are safe for living creatures, even using substantially more current than you will need. BUT you will always find people who are suspicious of such equipment.

Russell McMahon
  • 150,303
  • 18
  • 213
  • 391
  • 1
    This is interesting, but if I understand this correctly then the person holding the compass would have to walk through the coils. Is that correct? –  Apr 12 '12 at 20:26
  • Helmho_l_tz :-) – Curd Apr 12 '12 at 20:40
  • @DavidKessner - Yes, that's correct. I don't think that coils above and below the walkers (as posited by Russell) will suffice, you need to modulate the field parallel to the ground - They need to walk between the coils for the field to point directly to the left or right, or through one pair of coils and between another for intermediate directions. – Kevin Vermeer Apr 12 '12 at 20:47
  • I've maintained an industrial compass tester with 3D coils 75 cm on a side through which an industrial assembly line passes (though I wasn't a part of the design process). The coils don't have to be circular, you can have square or rectangular coils and still get good results. – Kevin Vermeer Apr 12 '12 at 20:47
  • @KevinVermeer - yes. I didn't mean to suggest they had to be any given shape. While the field will not properly assume the shape of a random coil this is an extremely undemanding application and relatively rough coils would suffice. – Russell McMahon Apr 12 '12 at 21:00
  • @DavidKessner - between, not through. And with only one coill "at some distance from". – Russell McMahon Apr 12 '12 at 21:03
  • @RussellMcMahon - You'd need to pass through one pair of coils and between another (varying the current through each to get the desired direction; the fields are additive) for the compass to point in any direction other than straight through the coils. – Kevin Vermeer Apr 12 '12 at 21:11
  • @RusselMcMahan + others: Thanks for your answers and comments, greatly appreciated! I updated my questions regarding the options to install hardware (sides of a 20m*20m area and within floor). Having the people run through the coils would be a good idea, but unfortunately messes with the exhibition. If I understand it correctly a single coil within the floor would not allow me to guide the visitors (exhibition A) but might provide an option to shortly deviate their compasses (exhibition B)? If so, how would I go on to calculate the parameters of such coil with max height 20mm? – InteractiveCube Apr 12 '12 at 22:10
  • 1
    @RusselMcMahan: Could you expand on your ideas of infrared and RFID? I see problems regarding infrareds line of sight, RFIDs detection range (detecting multiple tags at high frequency and large distance) and costs. As it's a public exhibition, funding is very limited... but it's for a good cause :) – InteractiveCube Apr 12 '12 at 22:17