Getting ELF Out of a Sound Card Using EMF Cancellation

Amplitude modulation is a nonlinear process (multiplication). A transformer is a linear process (addition and subtraction). How do you recover the original signal from that?

DC coupling from sound cards is well understood: just bypass the coupling capacitors on the board.

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Consider getting a PC soundcard with a SPDIF output (some desktop motherboard audio systems have them built-in), and then one of those inexpensive little SPDIF->line out DAC boards they sell on eBay. Then you can easily mess with the circuit and modify/remove coupling capacitors without messing up a nice internal sound card or audio interface

Wouldn't you then need a transformer capable of faithfully passing the ELF?

piglet

Still won't work because the real issue is that the SC is decoupled with a Cap for most of them.. Make a phase detector that nulls each other at the center freq and then you can modulate a higher freq carrier.. That is about the only sure way to get around it.

Using a PLL could also do it there by you can use the DC correction loop as your output but I think that involves even more components. also the other issues are that many on board cheap sound chips are not linear..

Jamie

some soundcards will pass DC

anyone who unserstands sine waves.

That won't work.

The sum or difference of the two audio sine waves will still be a sine wave.

You'd have to rectify or synchronously demodulate (and filter) to turn a high frequency sine wave into low frequency.

Or just DC couple the sound card DAC.

Modulating a tone with the ELF can certainly work, just use a basic envelope detector on the output to recover the ELF.

Of course this is all because elves don't fit through 3.5mm holes.

NT

What the F's an elf? I'm thinking I might not to 'pass one' myself. sounds painful.

:^) George H.

Hmm I meant might not *want* to pass one. never mind, GH

Gosh, why didn't I think of that?

is

c.

dio

lope detector on the output to recover the ELF.

Use 2 channels and you could make it bipolar too, each channel delivering o ne polarity.

Theoretically, and less usefully I dare say, you could also produce arbitra ry bipolar waveforms with very low bandwidth by multiplying 2 af channels. Varying frequency & phase of those 2 could control the output fully. I'm no t spotting any upside though, the software would certainly be more complex.

NT

probably would be bad for your elf. Extra low frequency.

NT

ELF (excruciatingly low frequency) is 3 to 30Hz. As others have mentioned, the frequency response of the sound card is limited by the input and output coupling capacitors. You can DC couple them: "Simple Sound Card Unipolar DC Modification" "DC coupled soundcard DAC mod"

If you prefer to roll your own, replace the sound card with a VCO (voltage controlled oscillator). Feed your ELF signals into the DC input and either envelope detect or synchronously demodulate the higher frequency output. Or maybe try a lock-in demodulator: "THE LOCK IN MODULATION IN ULTRA LOW FREQUENCY APPLICATION"

More:

If you're doing Schumann earth resonance work at about 7.83Hz, this one might be interesting: "Synchronous extremely low frequency (ELF) receiver"

It amazes me that anyone here indulges the OP. He seems hopelessly infatuated with magical thinking about electromagnetism, and won't take science for an answer. It's such a waste of time. Point and laugh by all means, but please stop trying to teach him anything.

Oops. Never mind the AM envelope detector. The VCO outputs FM and will need an FM detector.

actually..... Thinking more about it....

The old LM1496 chip.. Gilbert cell or the use of a balance mixer diode and transformer array.. Inject Left and Right channels at some freq like around 10KHz and output through a low pass filter.

Use the Lower side band. etc.

Jamie

ELF? Trying to talk to subs? Try green laser!

You can FM the ELF onto a carrier, record that. Done in long-period seismometry all the time (because DC is so ... incompatible with tape recorders). Then, a little PLL demodulates it afterward.

How ELF do you want?

Wouldn't it be easier to set the two channels of the sound card to produce f and f+df at the same amlpitude so that the sum of the two channels will beat at the required ELF and low pass filter to get the beat frequency at the amplitude of the original carrier wave (and the sound card decoupling can cope try 100Hz carrier) after rectifying.

Transformers don't have ideal behaviour.