Subaudio Multiplier for Soundcard Input

It's worth a try then. Cheaper, for some reason, than a USB DAQ.

Maybe someone could make a comfortable lving converting these to resell at half the price of a "real" DAQ.

Mark Harris.

Do you mean the citrcuit I have designed will do this "as is". If not, what would you suggest changing?

BTW I had previosuly considerd using an MLT04.

Do I understand correctly that what you described in the first entry is not considered to be a "frequency multiplier? What would you call it?

Mark Harris

How would the acquired FM signal appear in the spectrograph software as a readable relpcate of the orignal ELF?

Mark Harris

Of course. Look at SDR (sound defined Radio)..

But in any case, you can have the sound card monitor a steady tone and that tone can be a AM carrier.

Multiple reads of AM levels of a constant tone on the input can be used to regenerate a stream of PCM values via the amplitude of the tone. A dead carrier would be a tone sitting in a steady state around 50% of the full amplitude. doing a +/- amplitude can allow you to regenerate the original ELF signal, right down to 0 HZ.

But I made a comment earlier about some sound cards. AGC's can mess with you. It works best if you could generate a marker/sync pulse at full amplitude and then generate the actual signal at lets say 75% for

100% modulation. This way decide in software which is which.. But if you think you can insure the softcards settings to not use AGC then you should be able to set the input levels properly. Jamie

It's a voltage multiplier.. one input multiplies the other input and the sum is at the output.

However, don't get to upset about this, it will generate DSBs, fundamentals, third harmonics from the square wave etc..

it'll end up sounding like Jimmy Hendrix..

Jamie..

Do you mean "Software Defined Radio"?

Because of the cap, I specifically mentioned this usb soundcard where the PCB is easily exposed. And I also mentioned a level shift would be required.

A full amplitude carrier AM modulation demodulation can be done by simply

1a.) take the absolute value of each sample (full wave rectification) 1b.) set any negative sample value to zero (half wave rectification) 2.) low pass (maybe 150 Hz) signal to remove carrier residuals 3.) high pass signal (say at 0.1 Hz) to remove DC bias

Phase locking to the reduced carrier can be hard when the audio sidebands are only a few Hz away from the carrier.

Maybe a Costas loop might function for a DSB signal.

A full amplitude a few samples long synch pulse every 50 ms should do the trick. Should be easy to detect and eliminate from the data stream and compensate against AGC gain changes. Since the antialias filter will broaden the pulses, a bit longer section needs to be cut out from data stream.

It might be a good idea to invert every other pulse to remove any DC bias.

It generates double sideband suppressed carrier signal.

To generate normal amplitude modulation (A3E), add a DC bias to the audio input terminal on the multiplier, so that the voltage is always positive compared to the reference input. This will enable the carrier to pass through all the time and hence taking the absolute value of the modulated signal can be used for detection.

The analog input should be low pass filtered well below the carrier frequency (actually below the lower sideband). A sine (not square) wave should be used as a carrier, to avoid sidebands and carrier around three times the carrier frequency.

It is a _voltage_ multiplier.

Exactly as the AM sidebands provided that the modulation index is kept extremely low, to prevent the creation of second order sidebands, which would overlap the first order sidebands.

The OP did not say that the ELF signal is digital.

DSB is difficult, as the re-inserted carrier must have both proper frequency and phase. For a signal without a recognizable sync sequence, the loop is prone to mis-lock.

--

Tauno Voipio

Since U2 seems to have differential inputs for both X and Y, applying a DC bias on X2 should help keeping the X2 (audio) to X1 (bias) difference always at positive or always negative level and hence generate normal AM (A3E).

Yes, excuse me..

Jamie

Big snip Reticon - 1024SAD ???? - it's an analog delay "bucket brigade" chip. Company is no longer around but chips of this type were/are really common as a sound effect for guitars. All digital versions are very common in the music industry. For example there are boxes that can be set to generate any harmony (x2, x3 whatever), for a vocalist, with excellent fidelity. You can find the same function built into alot of audio recording software which allows you to change tempo or key or both. bg

Digitally you can do just about anything, it's only a matter of horsepower and available budget.

More esoteric analog tricks would be, for example, deliberatly distorting, picking one of the harmonic sets and downconverting that. Now you'd have a 2x, 3x, 4x ... 10x or whatever stretch. I just don't see how that would make sense in Mark's case.

--
Regards, Joerg 

http://www.analogconsultants.com/

On Sat, 07 Dec 2013 09:13:39 +1100, snipped-for-privacy@comprodex.com wrote:

--- You can do it with the analog multiplier you have on hand, but just for fun:

Version 4 SHEET 1 880 680 WIRE 416 -352 96 -352 WIRE 416 -336 416 -352 WIRE 608 -304 496 -304 WIRE 656 -304 608 -304 WIRE 496 -272 496 -304 WIRE 608 -272 608 -304 WIRE 256 -240 224 -240 WIRE 272 -240 256 -240 WIRE 416 -240 416 -256 WIRE 416 -240 352 -240 WIRE 416 -224 416 -240 WIRE 256 -176 256 -240 WIRE 496 -160 496 -192 WIRE 608 -160 608 -192 WIRE 608 -160 496 -160 WIRE 208 -48 -64 -48 WIRE 608 -48 608 -160 WIRE 256 -32 256 -96 WIRE -64 16 -64 -48 WIRE -272 48 -304 48 WIRE -144 48 -272 48 WIRE 96 64 96 -352 WIRE 128 64 96 64 WIRE 208 64 208 -48 WIRE 208 64 192 64 WIRE 224 64 208 64 WIRE 416 64 416 -144 WIRE 416 64 288 64 WIRE -272 80 -272 48 WIRE -144 80 -144 48 WIRE -64 112 -64 96 WIRE 32 112 -64 112 WIRE -64 128 -64 112 WIRE 96 160 96 64 WIRE 128 160 96 160 WIRE 208 160 192 160 WIRE 224 160 208 160 WIRE 416 160 416 64 WIRE 416 160 288 160 WIRE -272 192 -272 160 WIRE -144 192 -144 160 WIRE -144 192 -272 192 WIRE 32 208 32 112 WIRE -272 256 -272 192 WIRE -64 272 -64 208 WIRE 208 272 208 160 WIRE 208 272 -64 272 FLAG 32 208 0 FLAG -272 256 0 FLAG 608 -48 0 FLAG 256 -32 0 FLAG 224 -240 IF FLAG 656 -304 RF FLAG -304 48 LO SYMBOL diode 192 144 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D1 SYMATTR Value 1N4148 SYMBOL diode 288 144 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D2 SYMATTR Value 1N4148 SYMBOL diode 128 80 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D3 SYMATTR Value 1N4148 SYMBOL diode 224 80 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D4 SYMATTR Value 1N4148 SYMBOL ind2 400 -240 M180 WINDOW 0 -44 71 Left 2 WINDOW 3 -60 41 Left 2 SYMATTR InstName L1 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=.1 SYMBOL ind2 400 -128 M180 WINDOW 0 -46 82 Left 2 WINDOW 3 -58 55 Left 2 SYMATTR InstName L2 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=.1 SYMBOL ind2 512 -176 R180 WINDOW 0 -34 78 Left 2 WINDOW 3 -55 47 Left 2 SYMATTR InstName L3 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=.1 SYMBOL ind2 -48 224 R180 WINDOW 0 -37 79 Left 2 WINDOW 3 -60 48 Left 2 SYMATTR InstName L4 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=100 SYMBOL ind2 -48 112 R180 WINDOW 0 -48 76 Left 2 WINDOW 3 -63 48 Left 2 SYMATTR InstName L5 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=100 SYMBOL ind2 -160 176 M180 WINDOW 0 -45 70 Left 2 WINDOW 3 -60 40 Left 2 SYMATTR InstName L6 SYMATTR Value 100m SYMATTR Type ind SYMATTR SpiceLine Rser=100 SYMBOL voltage 608 -288 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value SINE(0 .1 10) SYMATTR InstName V1 SYMBOL voltage -272 64 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(-1 1 0 10n 10n 500u 1000u) SYMATTR InstName V2 SYMBOL res 240 -192 R0 WINDOW 0 -45 38 Left 2 WINDOW 3 -64 67 Left 2 SYMATTR InstName R1 SYMATTR Value 100 SYMBOL ind2 368 -224 M270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L7 SYMATTR Value .01 SYMATTR Type ind SYMATTR SpiceLine Rser=10 TEXT -264 224 Left 2 !.tran 1s TEXT -272 16 Left 2 !K1 L4 L5 L6 1 TEXT 496 -336 Left 2 !K2 L1 L2 L3 1 TEXT -288 -240 Left 2 ;DOUBLE BALANCED MIXER TEXT -288 -200 Left 2 ;John Fields, 08 December 2013

That would work if the ADC633 was happy with some common-mode voltage. Some chips with differential inputs really want them to be _differential_, without a lot of common-mode voltage on them.

I have no clue if the ADC633 is such -- that's what data sheets are for.

--
Tim Wescott 
Control system and signal processing consulting 
www.wescottdesign.com

Thanks, John

I will have a close look at this after work.

Mark Harris

OK. I will spend some practical learning time at the workbench on this.

Lots of options to choose from, thanks to the generous responses so far.

Mark Harris