Author Topic: Tricking Metal Sensor  (Read 1018 times)

0 Members and 1 Guest are viewing this topic.

Offline bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2237
  • Country: us
Tricking Metal Sensor
« on: July 15, 2020, 03:23:27 pm »
Long story short, I'm looking to trick a metal sensor on a machine to think it's detecting metal.

What could I use to do this?

I'm thinking maybe just sending voltage through a coil that's next to the sensor. Probably a simple circuit like a transistor, a pull up on the collector, driving a coil. I'm uncertain how metal sensors work, so I'm uncertain what would be affective.
 

Offline minifloat

  • Contributor
  • Posts: 34
  • Country: de
  • loss of precision
Re: Tricking Metal Sensor
« Reply #1 on: July 15, 2020, 07:36:49 pm »
Guess what you don't want to use is an actual piece of metal. SCNR

Do you have more info about the metal detector?
Which metal specific properties is it using?
Ferromagnetic
Eddy current
...

br mf
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: Tricking Metal Sensor
« Reply #2 on: July 15, 2020, 08:22:30 pm »
I suspect it detects eddy current.  I would try using an adjacent coil and shorting it.
 

Offline bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2237
  • Country: us
Re: Tricking Metal Sensor
« Reply #3 on: July 16, 2020, 01:25:50 am »
This is for a friend who has a work related problem.

Tubing goes through a "metal detector" to detect metal, but the metal parts tend to get lodged, and they'd like to increase the size of the tube. Unfortunately larger sized tube will not fit through the "metal detector" (which is a ring and looks like a custom built module).

My idea was to bypass the entire thing, use a larger tube with a metal detector, and the output go to a coil (?) that will trick the real detector into thinking it has a piece of metal passing by it.

I thought of a metal detector that outputs a high when it detects (or some sort of pulse) feeding the base of a transistor, a pull up on the collector feeding a coil that sits in the actual metal detector, and obviously a snubber diode across the coil.

This way the metal triggers the metal detector, turns on the transistor, energizes the coil with the pulse, creates a magnetic field, and turns off.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Tricking Metal Sensor
« Reply #4 on: July 16, 2020, 02:08:27 am »
If this is an industrial machine the existing metal detector may run on 24 VDC and provide a DC level back to the controller.  If you find a larger metal detector it may very well have a similar interface and it should be just a matter of working out connectors and wiring.

Here's a general introduction to the types of industrial sensors and how they can be used: http://www.ia.omron.com/support/guide/41/introduction.html
 

Offline bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2237
  • Country: us
Re: Tricking Metal Sensor
« Reply #5 on: July 16, 2020, 02:23:07 am »
Quote
If you find a larger metal detector it may very well have a similar interface and it should be just a matter of working out connectors and wiring.


My thought is not to replace the existing metal detector, but use an external metal detector to detect the metal, and that output triggers a circuit that energizes a coil located in the metal detector located at the machine.

This way the machine thinks it sees a piece of metal, but it's really seeing a magnetic field generated by another sensor that sees the actual piece of metal.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Tricking Metal Sensor
« Reply #6 on: July 16, 2020, 06:08:39 pm »
I get what you want to do and thus avoid messing with the existing wiring.   Do you know what kind of metal sensor it Is: inductive, capacitive or magnetic?

If inductive, then placing an air core inductor close the the sensor and then either shorting its leads (or not) may be detected.  It'll take some experiementation to work out the number of turns and distance from the sensor.

I don't know about the capacitive sensor - I suspect that some sort of antenna and inverting amplifier to make it electrostatically disappear.  Sort of like an electrical guarding circuit.  Maybe just use a solenoid to flip a metal disk in front the sensor.  This could work with inductive sensors too.

If magnetic it probably has something like a reed switch or Hall effect sensor and a bias magnet inside.  Placing an inductor near the sensor and then running a DC current through it that either adds or subtracts from the bias field.   Again, it'll take some experimentation to deterime the coil characterisistics and current level and polarity.
« Last Edit: July 16, 2020, 06:11:05 pm by duak »
 

Offline bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2237
  • Country: us
Re: Tricking Metal Sensor
« Reply #7 on: July 17, 2020, 02:41:16 am »
Quote
inductive, capacitive or magnetic

I'm uncertain. I'm guessing it's inductive, but it's for a friend, so I haven't seen it, only been described.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf