I want to measure an oscillating magnetic field using a coil. Using a small self-spun coil, I can already measure very decent 100 KHz signals by hooking each end of the coil up to an oscilloscope.

I want to eventually pump the frequency up to 10 MHz. Using my extremely basic coil+oscilloscope setup I can already measure this for fairly strong oscillating magnetic fields but the fields that I'll be measuring will be much smaller. I can therefore not just hook my coil up directly to the oscilloscope anymore, but will need to amplify the signal.

My problem is as follows: if I want to push my signal through some filters, then those filters will invariably have some capacitance and resistance. Hooking up my coil to non-neglectable capacitances turns the whole thing into an electric oscillator, whose resonance frequency lies (for decent coils) far below 10 Mhz!

The solution I came up with was first passing the signal through an opamp, to "restart" the circuit:


simulate this circuit – Schematic created using CircuitLab

I do have some opamps that can handle 10 MHz signals. The datasheet for the OPA6999 shows that it works just fine for these frequencies (page 7).

But won't I run into the same problem using this circuit? Won't the opamp introduce capacitance and inductance on the left side of the circuit that could significantly decrease my resonance frequency, unless I use a tiny coil?

  • \$\begingroup\$ Do you truly mean to have positive feedback as in your schematic? \$\endgroup\$ – Marla Apr 11 at 23:19
  • \$\begingroup\$ Yes, this is the circuit for a voltage follower, right? What I mean to do is process the signal on the right hand side (i.e. by adding filters and/or active components) without interfering with the resonance frequency of the LHS coil. \$\endgroup\$ – Heatherfield Apr 11 at 23:34
  • 2
    \$\begingroup\$ @Heatherfield, a voltage follower has negative feedback, not positive feedback. \$\endgroup\$ – The Photon Apr 11 at 23:35
  • \$\begingroup\$ Also, on the OPA699 datasheet, notice the next bullet after it tells you the GBW product is 1000 MHz, it tells you it's stable for gain greater than 4 V/V. This means it isn't stable in follower configuration. \$\endgroup\$ – The Photon Apr 12 at 0:10
  • \$\begingroup\$ Shame... I should have noticed that. What about turning the thing into a V/V converter using the right side schematic in this image? electronics-tutorials.ws/opamp/opamp48.gif That could give me the right gain to make the circuit stable, without compromising on the input impedance I wanted. Do you think I could use such a circuit like that to keep the coil resonating, and still filter the signal that comes through? \$\endgroup\$ – Heatherfield Apr 12 at 0:16

Some years ago I designed and built a gain-of-600,000X amplifier for 100,000Hz, as part of beginning to evaluate magnetic beacons. Is this the level of gain you are considering?

There were 2 modules, each with gain of 8 & 64 (analog-mux selectable), a gain of 1/2/4/8 to finetune the ADC loading, and a gain of 20X low-noise discrete amplifier.

And the frontend had 3:1 analog mux, to select X/Y/Z orthogonal coils.

As a teenager, I built many a 4/5/6/7/8 stage bipolar signal chain that oscillated until I learned to use extreme methods in VDD filtering. I call those "Local Batteries".

More recently, to evaluate silicon substrate crosstalk, I built AM radios ONCHIP with the sole purpose of downconverting a differential-signal (two probes into the substrate) to a few MHz IF and then boosting the IF by 40/80/100dB. This was oscillation-free in the 40 and 80dB modes; the 100dB setting did have some ringing but lacked overt oscillation.


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.