Metal detecting

[Bern]

I have a stream of material, in which there are occasionally small pieces of both ferrous and non-ferrous metal. These are from about ½-inch, down to where they are hard to see with the naked eye. I need to detect them. – Idea #1; Drop material through a 1¼-inch plastic pipe. Have three coils around pipe, approximately ½-inch apart. Drive outside two coils with 150 KC, 180 degrees out of phase, so signal would be cancelled in center coil. Any metal would cause an upset with the canceling effect as it fell through the coils, thus creating a detectable signal change in the center pick-up coil. - Idea #2; Use one coil, as a tank circuit of an oscillator. With a PLL, monitor for a change in frequency. - - - Which will work the best, as far as sensitivity, stability, reliability, etc? Is there a better way? How do this type of standard metal detectors do it? How do the “BIG BOYS”, do it? Thanks for your ideas and time.

[jwax]

Bern, if you're going to detect particles down to grains of sand-size, and up to 1/4", I think ultrasonics would be a better option. Depends also on flow rate, and the actual contents of the liquid. Any more details?

[Bern]

jwax, thanks for the reply, sorry I wasn’t clearer. The stream is not liquid, it is small plastic parts and pieces of many sizes and shapes. --- I am coming up with a total blank on how Ultrasonics could be used, either in or without liquid. What principle and technique are you thinking?

[Bern]

jwax, thanks for the reply, sorry I wasn’t clearer. The stream is not liquid, it is small plastic parts and pieces of many sizes and shapes. --- I am coming up with a total blank on how Ultrasonics could be used, either in or without liquid. What principle and technique are you thinking?

[russlk]

Most cheap metal detectors use the BFO system with a single coil. I doubt it would be sensitive enough for you. The 3 coil system is the best bet but I have no experience with it. There is another system called pulse induction but I tried that with no success.

[upsmaster]

Some use magnetic seperators for this problem,or vibration or size weight differences.. the plastic parts fall on to a screen which they can not pass through but the metal will some times aided by vibrating the screen or the container that the material stream falls into ... or strong pm magnets that filter or seperate the metal from the plastic parts. One way is to direct a stream of air across the material stream if the weight difference is enough from the plastic parts you can direct the parts or the metal in a different direction and seperate the two.
joe

[Bern]

Thanks Joe, some good ideas’, however, I have both ferrous and non-ferrous material, so magnets won’t do the trick. As far as airflow, there is such a difference in size and shapes, for both the product line, and contaminations, that the “effective” SG overlaps the various material and particles, both good and bad.

[Externet]

Hi.
Ferromagnetic particles are not a problem.
Other metals do have something in common against ANY plastics: specific gravity.
I would tackle the separation with a non-electronic approach, a wet centrifuge; or launching the material mixture with some velocity into a flowing liquid. The sinking location will be very different for metals and plastics. A properly positioned baffle would keep the materials separated.
Sorting heavier pieces of plastic and lighter pieces of metal is the key, that could be achieved by a submerged mesh conveyor in the 'grey' sorting area.
Seems you are into recycling... the melting heat will dry the plastic.
Well, this was for the nuts instead of the volts.
Hope helps somewhat,
Miguel<p>[ May 03, 2004: Message edited by: Externet ]</p>

[Bern]

Thanks for the idea, Miguel. I will be running over a thousand pounds a day, and drying would be another process I would rather not get involved in. Another possible problem, when a vary irregular and jagged tiny piece of metal comes it could hold on to an air bubble long enough to change its “effective SG”, Other than that, your idea defiantly has given me food for thought.

[rshayes]

The three coil approach with a Faraday shield around the pipe is probably the best approach. It would be excited by RF, probably in the low MHz range. You want inductive coupling, since plastic pieces should have no effect on magnetic fields. Both ferrous and non-ferrous metals should change an RF field, by eddy currents if nothing else. Balancing the coils may be tricky, and the degree of balance will determine the smallest detectable particle.

[jwax]

Do you simply want to count these metal pieces, and if so, what sort of rate would they be going through the detector? One/hr, one/millisecond etc.?

[Bern]

jwax, They are mixed in with the material, so are radom when they come through. They could be anywhere from side by side, to none. We would like to see none, but that is the problem. When detected, they will be removed from the "good" material stream.

[perfectbite]

A non-slurried thousand pounds a day through a 1+1/4" plastic pipe? That is a LOT of stuff even for a 24 hour day. I think your best bet would be to spill the stuff out as it comes in and vibrate/shake it out over a series of slightly concave trays (that way the heavier stuff would tend to gravitate toward the lowest part). Have the sensors located in the 1+1/4 plastic pipe right at the drop off edges (make the plastic pipe the drop off edge itself) and have sensor controlled solenoids, directly located beside the falling stream, push or chute divert whole metal containing sections of the stream onto another slightly concave chute/tray which in turn will go through the same sensing process and so on until you have separated out, through stepped elimination, the tramp metal. If the stream material is so irregular in shape as you say then running it dry through a very confining pipe section would cause more jamming than processing even if you 'tefloned' it. Did you want to have the metal stuff have very close proximity to the sensing coils by using the 1+1/4" plastic pipe?

[josmith]

Somewhere i saw a demonstration where a spinning aluminum disk caused a magnet to levitate. The point being that moving a non magnetic metal through a magnetic field will generate a force. I envision your material falling past a very strong magnet in a narrow column. The non magnetic material would be deflected away form the magnet,the plastic would fall straight down and the magnetic material would be deflected toward the magnet. With a little (perhaps very much) R&D you might be able to end up with three seperate material streams.

[haklesup]

I saw some TV footage of how a recycling plant worked. The material was fed down a conveyor belt where it would fly off the end and form a ballistic arc (shape influenced by gravity only)<p>This arc of falling debris was subjected to a magnetic field that would attract ferrious materials tward one hopper and an electric field that would deflect Al to another hopper. The remaining materials went on to other sorting procedures. Al cans literally leaped off the conveyor when passing the field. Not sure of the specs on this though.<p>This method kind of reminded me or mass spectomitry only on a much larger scale.