what is the best detector to detect iron sheets?

[ahmedamed75]

hi,
I am an engineering student participating in a robotic competition that requires a rover to detect a mine model under the ground, this model is a 10x10cm cube of iron sheet buried in a wet and mineralized sand.
what I see is that most detector schematics is trying to ignore sheets of metal, especially iron, so I am quite confused how to chose the best schematic to create.
however, I thought that an 'All-Metal' detector may be a good choice, so I already created a VLF GEB detector that only response to the phase shift of the received signal. was that a good choice?
And for the coil, what type of coil is the best for this purpose? and what is the appropriate dimensions?
The rover speed is 2m/s.

[kt315]

try load an archive. it is free. https://drive.google.com/file/d/1EIo...usp=share_link

[Skippy]

Presumably this 'fake mine' will be buried at the typical depth that anti-personnel / anti-tank mines are buried, approx 10 to 15 cm down ? In which case, it should be easy to detect, and a search-coil of 15 to 20cm diameter is suitable. I think a circular-shaped concentric coil would be best, as it is completely symmetrical. Your robot will be moving forwards/backwards/rotating , so you need the coil to work correctly regardless of which way it is moved.

[Carl-NC]

If "wet sand" does not include a lot of salt then a VLF should be fine. You can probably get away with a single all-metal channel unless you also have to ignore innocuous targets. A potential problem with rovers is motor noise, especially if you are using stepper motors. You may need to use a noise-cancelling coil method. Coil size depends on other goals, like whether finding targets is a speed competition. In that case you want it as big as practical. But bigger may be unwieldy.

[ahmedamed75]

yes, the mine is actually buried at a depth less than 10cm so it seems like an easy task!
but the robot needs to go fast because of the mission time limit, that is it can't inspect every spot of the 1x1m to tell if there is a mine inside or not as there is a 20x20m field to scan!
so the robot will most of the time be moving only in one line.
I was thinking about a DD coil as I knew that it can cover a wider area, is that right?

[waltr]

Yes DD can have a wider sweep size. Examples:
A 10" Concentric typical has a 5" RX coil so 5" wide
A 10" DD has just under a 10" wide sense.

Are there any "false" targets like non-ferrous that needs NOT to be detected?
Is the Wet fresh or Salt water? Salt is conductive so can affects the detector.

Is the mineralized sand consistent, evenly distributed or clumpy, only is spots?

[ahmedamed75]

If "wet sand" does not include a lot of salt then a VLF should be fine. You can probably get away with a single all-metal channel unless you also have to ignore innocuous targets. A potential problem with rovers is motor noise, especially if you are using stepper motors. You may need to use a noise-cancelling coil method. Coil size depends on other goals, like whether finding targets is a speed competition. In that case you want it as big as practical. But bigger may be unwieldy.

I want to know more about the "noise-cancelling coil method", could you please recommend a link or article?

[Carl-NC]

Since search speed is a goal, I would suggest a Figure-8 coil as wide as you can make it. It several advantages:

• Good RX sensitivity at the extreme widths
• Good noise cancellation (it has a positive lobe and a negative lobe)
• Easy to build, easy to null
• Not dimensionally critical

The downside is that there is a target null at the dead center of the coil, and targets in the positive lobe give a positive response while targets in the negative lobe give a negative response. This can be fixed by taking the absolute value.

[ahmedamed75]

Since search speed is a goal, I would suggest a Figure-8 coil as wide as you can make it. It several advantages:

• Good RX sensitivity at the extreme widths
• Good noise cancellation (it has a positive lobe and a negative lobe)
• Easy to build, easy to null
• Not dimensionally critical

The downside is that there is a target null at the dead center of the coil, and targets in the positive lobe give a positive response while targets in the negative lobe give a negative response. This can be fixed by taking the absolute value.

when you say that the response is reversed at each lobe, you mean the amplitude of the received signal right? what about the phase shift?
I mean can still use a GEB channel with it?

[Carl-NC]

Yes, GEB will still work but to get it to ground balance you will need to tilt the coil so one lobe is closer to the ground. First order, this coil has a natural cancellation of ground even without GEB, but setting the GEB to some nominal position will make it even better.

[ahmedamed75]

thank you, I will try it.

[kt315]

take a chance on STEP MAGNETOMETER. main your problem modern mine does not keep the metal parts. it is whole plastic.
you have to learn this new for you aspect.