Finding lowest frequency component in composite waveform

[moby]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

 

[Don Lancaster]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

Digital filtering, of course.


--
Many thanks,

Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: [email protected]

Please visit my GURU's LAIR web site at http://www.tinaja.com

 

[John Larkin]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

Possibly silly ideas:

If you pass the composite signal through a series of integrators, low
frequencies are selectively amplified 6 dB/octave per integrator
stage.

A cutoff-frequency feedback loop could be closed around a sharp-cutoff
tunable lowpass filter, switched-capacitor probably.

John

 

[Michael A. Terrell]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

Look at the signal with a spectrum analyzer that covers the
frequencies in question. All will be revealed. ;-)


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Le Chaud Lapin]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

How analog/low-powered?

Beat the input signal with step-wise sweep oscillator starting at
lowest tone and going upward. Low-pass filter the result into a
threshold detector, and stop when triggered. You could store the
inputs to your VCO, pre-amp, and comparator in ROMDAC.

-Le Chaud Lapin-

 

[Ken Smith]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

A HC4046.

Take the input and convert it to a PWM signal a some high frequency.

Apply this PWM signal to the XOR phase detector.

Use an op-amp integrator based loop filter. Make the PLL filter have a low
enough gain that a nearby strong signal won't yank the PLL up away
from the lowest frequency signal. Don't forget the zero to make the loop
stable.

Force the integrator output to 0V. Apply a small current to the input to
cause the integrator to slowly ramp the VCO input up.

When the PLL comes to the first frequency, it will lock onto it.

 

[Mac]

I have an application where I need to find the lowest frequency
component in an audio signal
(say 500Hz..8KHz) comprised of a number of unrelated tones. This will
not necessarily be the dominant amplitude component and I will need to
have an amplitude threshold below which minor components are ignored.
DSP based solutions are not ideal. Something analog and low powered
would be ideal. Suggestions ?.
cheers
M

You could use a frequency sweep and a mixer-type phase detector. Feed your
signal and the frequency sweep into the phase detector. The mixer output,
suitably filtered, would tell you the amplitude at that frequency.

--Mac

 

[Ken Smith]

A HC4046.

Take the input and convert it to a PWM signal a some high frequency.

Make that, lets say, 512 times the suspected frequency. If not, you may
have a lock onto the harmonic beats. ie: the VCO is running 512 times the
input.

Apply this PWM signal to the XOR phase detector.

Also convert the output of the divider chain to PWM at, lets say, 256
times the suspected frequency. You have to do this to avoid subharmonic
locking.