sensor

[salman]

hello
i have made a toroidal sensor for my proton precession magnetometer.i want to ask that is the toroidal sensor sensitive to external field.bcz i have studied EMF rules that external field of a toroid is zero.will it sense external field

[Willy Bayot]

Re: sensor

It is indeed true that a toroid does not GENERATE much external field when being polarized.
The pick-up signal of a PPM is not generated directly by the external magnetic field.
The proton precession effective inside the water content of the toroid depends on the external global field anyway. This precession (which is INSIDE the tore) then generates a signal in the coil by induction.
Willy

[salman]

Re: sensor

it means that the toroid will detect any metallic object but the difference in frequency will be very low.can i use speaker to listen the difference in frequency.plz guide me in this regard

[Willy Bayot]

Re: sensor

NO, it could only detect FERRO-MAGNETIC objects.
The sine wave signal generated by the proton precession will be VERY SMALL (

[salman]

Re: sensor

its a big problem to measure its frequency.i have to disply to my faculty that the frequency changes with a feromagnetic material.so which cct i should use to measure this minor frequency change.plz help me in this regard

[Willy Bayot]

Re: sensor

You need first to build and test a LOW NOISE audio amplifier with a band pass filter with a center frequency of around 2000Hz and a band of around 100 to 200 Hz. The total gain of the whole chain should be at least around 100,000 to get an output signal of say 100mV sufficient to be triggered by a comparator into a square wave signal and then, entered into a digital input of a micro-controller. The program of this controller should then count the zero crossings of the square wave signal during a VERY precise time of say, 500msec or 1 sec. just after the end of the polarization cycle.
This will give you the proton precession frequency which corresponds to the GLOBAL HEARTH MAGNETIC FIELD VALUE.
To measure the gradient generated by the presence of ferro-magnetic objects, you need to make successive global field measurements at some distance intervals (say, 1 meter) in a short time interval and substract their values. Each difference is the gradient. If the gradients are very small or null, then there is a good chance that there is nothing buried there but if the gradients grow bigger, then there is SOME CHANCE that there is something disturbing the field. Beware, it still could be electrical cables or a local electro-magnetic interference. If you have excluded all the (many) causes of disturbance, then, it could finally be a ferro-magnetic object.

Good Luck,

Willy

[salman]

Re: sensor

i have two queries.
1-how can we be sure that we have removed all the causes of disturbances.plz explain.
2-we r not looking for buried material actually we have to measure the change in the field caused by ferro magnetic material placed closed to the sensor
we have implemented the cct of Phill Bames as it is . except we have changed the resonating capacitor for our required frequency 1.916Khz.the total gain of cct is more than 1000,000.still the signal is 20mV.we r using micro controller 89C52.the band pass range of filter is 1K to 3K as in the cct

[Willy Bayot]

Re: sensor

If you have implemented the circuit of Phil Barnes, you should normally get a signal which is the result of subtracting the sine wave signals of two separated sensors. This result is an amplitude beat depending on the differences in frequency, thus on the field gradient between the two sensors.
What is the shape of the final signal? Is it well over the noise level? Did you try to run it through an FFT process to see its spectrum?
If you have such a total gain and still only 20mV at the output, then, the input signals from the sensors should be quite weak. You could make bigger sensors (more water, more turns).
Actually, measuring changes of field due to ferro-magnetic material put close to the sensor should be much easier than finding buried targets.
The sensors should be put well out of reach of any electrical cables runnning alternating current and out of big ferrous masses. It means that all the tests should always be made OUT OF ANY BUILDING.

Willy

[salman]

Re: sensor

thanx for ur reply.i musing toroidal sensor with turns of 2500(with tuning capacitor of 480nF).presently it is picking up asignal of 2K approximately(seen on the oscilloscope).i have converted the signal into square wave using comparator for microcontroller input.
now the requirement of project advisor is
1-it should accurately measure the local earth magnetic field
2-if u bring ferromagnetic material close to sensor it should measure the new magnetic field.

i want to clarify
1-how can i come to know the signal which i m getting is the actual required signal
2-what should be the specifications of the microcontroller for my requirement
3-will i be able to measure the change in the magnetic field by bringing ferro magnetic material close to the sensor.
4-how much ferquency change will occur by bringing the ferro magnetic material of normal size.

[iftikhar]

Re: sensor

i amworking on constuction of DIFFERENTIAL MAGNETOMETER using PHIL,S DOCUMENT.
so far i have implemented the preamplifer, bandpass filter and amplifier stages succesfully.
in sensor construction i am facing alot of problems. i have not got bottles of exat diameter. i have taken PVC pipe and sealed from both sides and wound the turns on this self made botle. in this way i have constructed both the sensors . i hahe done all the winding with my hands.the hight of sensors is little different but winding area is the same.
now i do not know inductance of both the sensors. will the inductance of sensors be different? will size mater?
how can i tune my sensors to precession frequency?
can you help me in getting exat sensors?

[Willy Bayot]

Re: sensor

First, how did you test the various amplifier stages? What is the total gain of the whole chain? What is its thermal noise level?
You should get a very good SNR from the amplifier to get good final results.
I do not understand why you could not find bottles of adequate dimensions. It should not be EXACTLY the same but around it. Try with glass spice bottles. If you seal the cap with glue, they are OK. Also, there are a lot of small plastic shampoo bottles which should be OK as well even if they are not perfectly cylindrical. I am not even sure that the winding should be segmented like described.
However, the two coils should be as much as possible the SAME (same number of turns, same inside diameter, same length) and they must be well aligned on the same axis in order to cancel the surrounding noise as perfectly as possible when they receive the same signal.
For the tuning of the two coils, fix them where they should be, connect them as they must be (in series but in opposition) and connect them to the shielded cable that you will use later. Connect an AF generator able to deliver a constant RMS sine wave when changing its frequency to a small auxiliary coil and put this coil very close to ONE of the coils to be tuned to get a signal by induction. Connect a scope to the coils to be tuned or, better, the audio card of a PC on which you run an audio spectrum analyzer program (I can show you the one I am using, you can get it free).
Start with a frequency of 500Hz and increment the frequency by steps of 500Hz. Observe the RMS voltage value on the scope or on the FFT peak of the Spectrum Analyzer. The maximum RMS value corresponds to the tuning frequency of the coils. Try first without any cap with which you should get a tuning much too high and then, try with various combinations of caps in parallel starting with the value proposed by Phil. If the tuned frequency is too high, increase the cap value otherwise decrease it.
Willy