0
$\begingroup$

I am trying to measure an oscillating magnetic field, of a solenoid and in the air gap of a transformer core. I have a Hall effect sensor, but its frequency range is limited, so I have put together a couple small coils which I connect to an oscilloscope to measure the field. The transformer core setup is within the frequency range (I have just been going to 40 KHz) of operation for the Hall sensor so I have used both and compared results to characterise a 3 turn coil's output voltage for given flux density. Similarly I have characterised the second coil 1 single turn based on measurement comparison with the 3 turn.

(Which in my mind seems like a sensible enough approach, and as the coils are the same radius and I got approx. 1/3 the output reading for the 1 turn compared to the 3 turn appears correct, but if I am mistaken please correct me).

I am running the solenoid at much higher frequencies (up to 20 MHz) and with a lot more power. I would like to know if I can assume that a coil set up for measurement has a linear relationship with field strength. The Hall sensor has a defined range for linear operation. I do not think this is the case for a measurement coil but I am not sure and would like to confirm.

And to ask whether the frequency of the magnetic field is something that needs to be factored in as well for my measurement?

Also, when trying to measure lower strength fields with the loops my output signal is very noisy. Why is this, and would it be improved with increased number of turns?

Any help, advice or tips would be appreciated.

$\endgroup$
  • $\begingroup$ 20 MHz is getting to be a pretty high frequency for an electronic circuit as you have to be aware of issues such as stray capacitances, skin effect, impedance matching, signal reflections, etc.. What's the purpose of your measurements? Do you really need to go that high in frequency? $\endgroup$ – Samuel Weir Sep 6 '18 at 17:11
1
$\begingroup$

This is an electrical engineering question. I think there are forums for asking such questions. Anyway, I will try to answer as I happen to also be an electrical engineer.

Your coil idea is good. Coils without magnetic cores are very very linear, and they work up to very high frequencies, until the capacitor effect becomes apparent. After all, there are capacitance between turns, between different sections of the same turn, etc. This capacitance is not a problem at low frequencies (eg, 40KHz), but they must be considered at high frequencies(eg, 20MHz). Once this effect is considered, you can use your coil to measure high frequencies without a problem. A coil is still linear with the capacitors. It is just the magnetic field strength needs to be calculated from the coil output with this capacitor in mind.

One turn coils have low capacitance. Multiple turn coils might have high capacitance. The 1/3 relationship of the output voltage of your two coils might change when measuring high frequencies. There are ways to make low-capacitance coils (say, turns are separated with a distance).

When you are measuring low strength fields, the output voltage of your coil is low. That's why it is noisy because your noise level did not change, thus you have low SNR (signal to noise ratio). To deal with this problem, you need to amplify the signal with low noise circuits. Operational amplifiers, or Op-Amp's, are invented for such applications. One design is called "instrumental amplifier". Take a look of those.

BTW, your transformer might not work well under 20MHz, if it was designed for 40KHz in the first place. That is another story.

$\endgroup$
  • 1
    $\begingroup$ excellent points from @verdelite. here is something else to try with regard to the noise issue. see if you can isolate the chassis ground from the (-) input of the scope- so the first stage of the o-scope's internal preamp is operating more in differential mode. this sometimes helps. $\endgroup$ – niels nielsen Sep 6 '18 at 17:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.