Detect a magnet?

[jimmy101]

I wonder if you could use a compass based detector?

A quick test with a very crappy compass and a 5/8"D rare earth magnet shows that the compass will deflect ~45 degrees from north with the magnet to the west at 8" from the compass.

If you took a compass and oriented its base vertically it would point down when it crossed over the buried magnet (Or up if the magnet was pointing in the opposite direction.)

Might need two compasses, both mounted with their bases vertical (instead of the normal horizontal) and their two bases at 90 degrees to each other. As the 'bot drives around at least one of the compasses will point north (the needle will be horizontal). The other compass will be confused when it is pointing east/west since it can't rotate in the proper direction. You need two compasses so that at least one is always aligned properly with north no matter which way the 'bot is pointing. When the compasses pass over the buried magnet both needles will point down (or up) instead of at least one needle being horizontal. If you use orienteering compasses then both the top and bottom of the case are clear plastic. So, you could use a phototransistor to "see" the needle when it is pointing down (or up), or you could "see" when both needles are not horizontal.

A more "high tech" approach would use a solid state compass module like those found in some cars and other navigation systems. Digikey sells a compass module that might work;
Honeywell HMC1055 COMPASS SENSOR 3-AXIS KIT, $36
http://www.digikey.com/scripts/DkSearch ... 42-1036-ND
The PDF of the catalog page is http://dkc3.digikey.com/PDF/T072/P2114.pdf

The robotic hobbyists already have "fairly cheap" modules that should work for detecting a magnet (anything that can detect the Earth's magnetic field should be able to see a strong magnet at several inches.) You might want to take a look at http://www.superdroidrobots.com/shop/it ... 8&catid=35, for $52.

Parallax (the microcontroller people) also sell a compass module for their Basic Stamps ($40 from Mouser Electronics, http://www.mouser.com/catalog/631/46.pdf)

The Lego Robotics System has a compass module, $47 from http://shop.lego.com/ByTheme/Product.as ... 1034&cn=17
You can probably rummage around on the net and find info on this sensor's specifications and how to hack it for other uses.



Is there anything in the description of the problem that says which way the magnet will be pointing? It'll make a huge difference if the magnet is oriented with the poles pointing up/down vs. side to side vs. "the direction the poles point will be random". Any decently thought out challenge should specify one of those three conditions.

[beebs111]

would there be a way to make a very powerful electromagnet and then measure resistance, and then just see when the restiance is the strongest? that seems like a pretty good soloution to me. other than that you could try very high mesh iron that would move if passed over a magnetic field, triggering an alarm or something of the sort.

[rednecktatertosser]

I dont thinnk anyone has said it, so ill throw it in, you could try some sort of GPR device. (thats Ground Penetrating Radar for those who dont know)

[paaiyan]

would there be a way to make a very powerful electromagnet and then measure resistance, and then just see when the restiance is the strongest? that seems like a pretty good soloution to me. other than that you could try very high mesh iron that would move if passed over a magnetic field, triggering an alarm or something of the sort.

I'm pretty sure that would instefere with electronic components, bigtime. If I were to go that route, I'd just make an uber strong one that would pull the magnet out of the sand, then have to robot plant the flag there, because that's now where the magnet is, lol.

GPR might work, I'm not sure how easy that could be made though, and if there are other metal objects in the sand, that would interfere a lot. Also, the bot would have to be able to interpret the reading from the GPR, that could get really complex.

[rednecktatertosser]

well he said it could be remote controlled, i figured GPR would be eaisily adapted to that, and if they know what the magnet looks like then they could probably recognise its its signature on the GPR monitor.

[PCGUY]

This is more of an FYI since I know it won't work with the magnet being buried but...

I was talking with someone who was testing the fire alarm at work the other day. He had a tube that could spray smoke at smoke detectors to set them off, and a magnet. I asked what the magnet was for, and he showed me how he could trigger the heat sensors on the ceiling with a little magnet close to them.

Would be nice if the magnet didn't have to be so close.

[robert23]

@redneck: We have thought about that, but no one in our group knows anything about GPR. Also, we will most likely be in the fully autonomous class, meaning it's not going to be remote controlled.

As far as how the magnet will be positioned in the ground, we have no idea. We also don't have any idea how powerful the magnet will be, and that would help us out a lot. A big neodymium magnet would actually be pretty easy, but it could also mean trouble for electronics, so I don't really know what to expect. I will be able to find out more about the comp once school starts back up, my department head is involved in the organization that hosts it.

[beebs111]

if you wanted to pull a weak magnet out of the sand, you could try one of these super magnets, idk what your budget is, but there is one of the "super dangerous" ones that only costs $75

[jimmy101]

This is more of an FYI since I know it won't work with the magnet being buried but...

I was talking with someone who was testing the fire alarm at work the other day. He had a tube that could spray smoke at smoke detectors to set them off, and a magnet. I asked what the magnet was for, and he showed me how he could trigger the heat sensors on the ceiling with a little magnet close to them.

Would be nice if the magnet didn't have to be so close.

Sounds like the heat sensors are basically thermostats that use a bimetal coil or just a coil of wire. Heat it up and it expands and a contactor on the end moves and touches a fixed contactor. Basicly like an old house thermostat. His magnet probably grabs the contactor on the end of the bimetal and pulls it to the fixed contactor.

I wonder if that was a clever design feature or something people "in the field" figured out as a way to test the sensors?