[MOVED] Inductive sensor (Eddy Current) with DC output

[faisal78]

Hi
I am looking to build a simple inductive sensor to detect the distance of a metal object.
Detection range is between 1mm-20mm.
I was planning on using a simple coil wound inductor to achieve this, i.e. maybe in the range of 6mH.
However, the trick is the sensor output, I'd like it to be DC analog output.
The oscillator I plan to use a simple 555 timer in the range of 100kHz.

Thanks

 

[erikl]

Re: Inductive sensor (Eddy Current) with DC output

... sensor output, I'd like it to be DC analog output.

Use a precision rectifier circuit. S. e.g. Razavi: "Fundamentals of Microelectronics" , or the TI "Handbook of OpAmp Applications" .

 

[jiripolivka]

Sensor output will depend on the circuit used. For eddy-current proximity sensor, the simplest circuit is a comparator which is triggered by pulse amplitude. Pulse amplitude decreases with a conductive object approaching. A simple comparator can be adjusted to respond to a chosen object distance, and typically has a TTL output. You can generate the pulses with a 555.

A simpler circuit can use a LM324, one opamp as a pulse generator, another as a comparator.

 

[faisal78]

Hi, I don't quite understand how the comparator & pulse generator would be connected to the eddy-current sensor. can you please help elaborate jripolivka.

Another thing, I managed to hack up a resonant circuit using 27uH 40cm coil & 220pF.
I am injecting a 2MHz pulse and measuring DC current in series with the pulse generator.
The LC tank unloaded measures 900uA.

4cm 0.9mA
3.5cm 0.86mA
3cm 0.83mA
2.5cm 0.8mA
2cm 0.74mA
1.5cm 0.03mA
1cm 0.035mA
0.75cm 0.5mA
0.5cm -0.5mA

THis seems to be working correctly? However, does DC current measurement represent metal proximity?
I am also wondering why after dropping to 0.035mA @ 1cm distance, it comes back up to 0.5mA?
Also how is it possible for a -ve negative current?

 

[jiripolivka]

Measuring oscillator DC current shows a certain response to LC circuit loading by the approaching object. But as you can see, it is not exactly the direct proportion you would wish.
Therefore as I wrote, instead of measuring DC current of your oscillator, add a RF detector of the RF voltage across your LC circuit. As the conductive object gets closer to LC circuit, the eddy-current load causes the detected voltage to drop.
Then use a comparator to detect the drop against a reference voltage. This voltage you can adjust so the sensor will respond to a given material at a certain distance from your LC circuit.

 

[faisal78]

Thanks for you advise jiri.
I have managed to make it work now using a RF diode detector on the output of the oscillator to achieve DC.
This was using a winding wire and I achieved ~20uF out of it.
Now, I am trying to proceed and use a printed spiral inductor on the PCB.
However, with the amount of space, 10mmx10mm... and based on single layer spiral inductor calculator available online, i can only achieve ~2.2uH of inductance with ~20 turns.

Is 2.2uH sufficient to generate enough magnetic field to be used as a eddy current sensor?
The conductive object that i am trying to detect, is within 1/2 of the coil winding diameter (<5mm) guideline.
But there does not seem to be a inductor value or Q and its affect to my system performance.
Any advice

 

[FvM]

2.2 uH might work with a circuit that precisely measures the inductor loss tangent respective real impedance part, but rather unlikely with the simple methods you are using up to now. In metallurgy, there are also eddy current measurement methods utilizing pulse response.

As another problem, coil resistance becomes a relevant part of the loss in this setup, so you get stronger temperature dependency with the small coil. I would rather go for a usual inductive sensor geometry with a ferrite pot core half.

 

[jiripolivka]

At this low frequency a printed coil looks like not an optimum technology. Inductor Q is better with a wire-wound coil.
Using a ferrite core may help but it focuses the RF field on-axis and close to the core end, so try to adjust your sensor for best results.

 

[faisal78]

2.2 uH might work with a circuit that precisely measures the inductor loss tangent respective real impedance part, but rather unlikely with the simple methods you are using up to now. In metallurgy, there are also eddy current measurement methods utilizing pulse response.

As another problem, coil resistance becomes a relevant part of the loss in this setup, so you get stronger temperature dependency with the small coil. I would rather go for a usual inductive sensor geometry with a ferrite pot core half.

Hi, can you clarify on the pulse response measurement methods?
How would this circuit look like?

 

[jiripolivka]

Hi, can you clarify on the pulse response measurement methods?
How would this circuit look like?

I did no modeling but rather experimented. I think the coil inductance alone is not too important but the detected material conductivity. Sometimes you can adjust the oscillator to be sensitive to detected material, other oscillators "hold" better but so doing the sensor was not too sensitive.

 

[FvM]

Hi, can you clarify on the pulse response measurement methods?
How would this circuit look like?

Unfortunately I don't know any details. I heard about it in a presentation many years ago. I don't think that it's a simple method.

 

[jiripolivka]

Pulsed eddy-current methods are used to measure coating on metal surfaces. Pulse decay is measured and paint thickness derived from it.
In general, pulse shape and delay can be used in a distance sensor, too. Only the circuitry is a bit more complex.See Arduino library.