Go Down

Topic: Metal imaging system (Read 3470 times) previous topic - next topic

I'm new to Arduino, but have a little experience with electronics.  Not much.  I can solder, I know basic components, and one time I etched and built a PCB for a metal detector.

I'm trying to prototype a metal detector that uses "high" resolution images instead of sound to detect metal.

I've done a lot of research and I think I know how to do it, theoretically speaking.  Now I'm hoping some kind-hearted, brilliant person can help me move from theory into reality.

The basic idea is to have an array of tiny metal detectors - hundreds of data points - that feed into a image.  I think I can accomplish this with Hall Effect sensors distributed in a square array.  I'm starting off with 100 sensors.  The sensors will be linear(not latching), to output a gradient voltage depending on the magnetic field they encounter from the metal object.  I may have to also induce a magnetic or electric charge into the environment to charge the object.  I'm not sure what that will look like yet, or what it may do to the sensors.

I would hook these up with wires to some sort of multiplexer.  There would also be an op amp to pick up the variations, and something to strain out the signals against the environment (can't remember what these components are called). The multiplexer would feed into the Arduino.  The Arduino would run the signals through a program that outputs to a computer or LCD screen to show an image of the gradient in voltages.   Hopefully this gradient, with enough data-points, will resemble the outline of the metal object being surveyed. 

If you think I'm on the right track, please let me know what to buy and how to hook it up the right way.  If you know a better way, I'm all ears.



I'm trying to prototype a metal detector that uses "high" resolution images instead of sound to detect metal.

Cut out the meaningless jargon, and put the words in the right order.
You're trying to form images from an array of sensors that, like most metal detectors, do not involve sound.
"Pete, it's a fool looks for logic in the chambers of the human heart." Ulysses Everett McGill.
Do not send technical questions via personal messaging - they will be ignored.


If you think I'm on the right track,

Sorry I don't.
Lots of things wrong.
For a start a Hall effect sensor will not detect metal. It measures magnetic fields. You might think you get a magnetic field variation in the presence of metal but first off it only works with magnetic metals, iron, nickle and cobalt, it will not work with other metals like copper, zinc, gold, silver, lead, tin and so on. Second the amount of interaction you get between a lump of metal and the Earth's magnetic field is too small to detect with a Hall effect sensor. You need a flux gate magnetometer.

Traditional metal detectors work by sending an electro magnetic field into the ground and picking up disturbances in that field. That works with all metals and in fact all conducting materials.

Good points.  Right now, I just want to image something, so ferrous metal-only is fine.  Later I can switch out for tiny coils.

I mentioned in the post that I will probably have to induce a magnetic field into the environment.  Here is an example of how someone did this to find pipes behind a wall.  http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3274295/

Let's say I do away with the hall effect sensors and go to a traditional metal detector coils, but make many small ones...how do I feed hundreds of sensors into the arduino?  A multiplexer?  I think if I used coils, I would try to do a phased pulse to induce the electromagnetic field.  Can Arduino send a pulse output, and then in the next iteration, take a sensor input from the same pin?


Having lots of tiny coils is not going to do anything for you. What happens is that they will all cross couple and you will get the same signal from each coil.
This is why they don't use this method for detecting metal.

Yes they would.  I was planning to send a pulse from one coil and receive the signal from another coil.  I would build the image by phasing these one after another, like a cathode tube on a TV.  That's why I was asking if an arduino can use a coil as an output then switch and use it as an input later in the series. 

To clarify, yes they would cancel each other out...I agree...so I was going to only use two, maybe three, at a time.  But because their position is different, and they are small, they would provide a resolution.  This method is used in phased-array Eddy Current testing.

Here is eddy current testing...http://www.olympus-ims.com/en/eddycurrenttesting/


The first step for you is to develop the working sensor, then deal with how to control it from the arduino. Once you have one sensor working, you can look at the possibility of adding more.
Google forum search: Use Google Advanced Search and use Http://forum.arduino.cc/index in the "site or domain:" box.

That's a good point, if I am using the coils.  I was planning to use the 100 hall effect sensors I just bought, which should work already.  I just want to figure out how to hook them all up.  Multiplexer?


A  CD4051 is the bog standard analog multiplexer such as in the patent circuit.


Just hook up one coil or one hall effect sensor and get a feel for how it works. I think you will be disappointed at the range if you can get anything out of the set up at all.
Yes I have worked with eddy current flaw detectors for tubes, those coils are huge and the current was large.l I am talking about 40A coil current with a range of a few mm.


I made a Proton precession magnetometer some years back, before the days of Arduino, when micros were black magic.

They are a lot of work to make, but evidently have a good range. 

I dont think you could have an array of them, but you could use a single one to scan a surface ( mine was to scan a beach for treasure shipwreck, but they found the wreck before I finsihed my project, so I dumped it )


At what kind of speed do you want to get your results? You may be better off getting multiple channel external adc's and use spi,  then you'll be able to sample faster and do other things while the adcs sample,

My biggest problem is how to plug hundreds of sensors into the arduino.  I'm imagining soldering wires from the sensors to a bunch of different ribbon tapes and then feeding those into 8/1 multiplexers.  That's not enough sensors, so then I feed those into sub-multiplexers?  There has got to be a better way.

Go Up

Please enter a valid email to subscribe

Confirm your email address

We need to confirm your email address.
To complete the subscription, please click the link in the email we just sent you.

Thank you for subscribing!

via Egeo 16
Torino, 10131