Solve Opamp Amplifier Low-Level Tune Problem

[m718]

I've made a single opamp amplifier for a mic but it has this problem, when I play a very low
200hz tune the output of the opamp shows 6 mV and another 200hz tune at higher volume shows a 600 mV output, the problem is that if the loud tune is ON the low level tune doesn't make any difference on the output. Shouldn't the output go from 600 mV to 606 mV ?
And each tune is played by a separate speaker.

 

[Bob S]

Are you summing the two input signals through a summing junction (the negative (inverting) input of an operational amplifier? This should sum the two amplitudes. See thumbnail.
Bob S

 

[berkeman]

I've made a single opamp amplifier for a mic but it has this problem, when I play a very low
200hz tune the output of the opamp shows 6 mV and another 200hz tune at higher volume shows a 600 mV output, the problem is that if the loud tune is ON the low level tune doesn't make any difference on the output. Shouldn't the output go from 600 mV to 606 mV ?
And each tune is played by a separate speaker.

Are the two tones in phase? If not, then you will not get a different peak value in general.

 

[m718]

Are you summing the two input signals through a summing junction (the negative (inverting) input of an operational amplifier? This should sum the two amplitudes. See thumbnail.
Bob S

I have the two sounds coming from two separate speakers but I have just one mic receiving both and the mic is connected like this: + and - of OPAMP to ground thru 1K, 100K - to OUT,
dynamic mic to ground and + of OPAMP

I also tried different frequencies for the two sounds.

 

[m718]

Are the two tones in phase? If not, then you will not get a different peak value in general.

but if your speaking into a mic and there is background noise all the sounds are not in phase how come in that case all the sounds add to the amplifier output?

 

[Bob S]

The velocity of sound is about 1100 ft per second, and the wavelength of 200 Hz is about 5.5 feet. Try moving one speaker about 1 to 3 feet closer and/or further away from the mic.
Bob S

 

[m718]

The velocity of sound is about 1100 ft per second, and the wavelength of 200 Hz is about 5.5 feet. Try moving one speaker about 1 to 3 feet closer and/or further away from the mic.
Bob S

I get the same results .
could this have somthing to do with the dynamic range of the opamp I read somewhere the there is a minimum amount of signal that the op amp can amplify in the presence of a much larger signal?

 

[m718]

Are the two tones in phase? If not, then you will not get a different peak value in general.

if out of phase signals don't increase the peak voltage would I still see a power output increase
and if yes how would I measure the power output of the opamp?

 

[berkeman]

I like Bob's suggestion the best. Drive both speakers with the same tone source, and move the speakers to adjust the phasing at the mic. You should be able to see some delta in the peak amplitude as you move the quieter speaker to in-phase and out-of-phase positions.

 

[m718]

I like Bob's suggestion the best. Drive both speakers with the same tone source, and move the speakers to adjust the phasing at the mic. You should be able to see some delta in the peak amplitude as you move the quieter speaker to in-phase and out-of-phase positions.

but I need this to work without of phase signals too so if I can't get the peak voltage to change, does the power output change? or anything else to show that the small speakers is ON?

 

[berkeman]

Can you tell us a bit about your application? Maybe we can make some suggestions, if we know what you are trying to do overall, and what the nature of the signals is.

 

[uart]

but I need this to work without of phase signals too so if I can't get the peak voltage to change, does the power output change? or anything else to show that the small speakers is ON?

Out of phase signals cancel. That's just how audio signals combine, it has nothing to do with you're amplifier. It's already happened in the air before it even gets to your microphone so get over it (or state more precisely what you're trying to achieve so someone can help you).

BTW. If the signals are not the same frequency then they are are additive, but not in the simple way you're expecting. If you have a 600mV (rms) tone and a 6mV tone at another frequency then the combined rms voltage is \sqrt{(600^2 + 6^2)} which is only about 600.03mV. Note however that although this looks almost unchanged (from the original 600mV) it does however contain precisely the sum total of the power of the 600mV tone and of the 6mV tone.

 

[Bob S]

If you combine two tones of slightly different frequencies, you will get sidebands of the difference frequency and the sum frequency. The amplitude will have hills and valleys in it with the max an min being about the sum and difference of the peak amplitudes of the two frequencies.
Bob S

 

[berkeman]

If you combine two tones of slightly different frequencies, you will get sidebands of the difference frequency and the sum frequency. The amplitude will have hills and valleys in it with the max an min being about the sum and difference of the peak amplitudes of the two frequencies.
Bob S

Only if the combination mechanism is non-linear. The microphone picking up multiple audio tones is a linear mechanism.

 

[Bob S]

Only if the combination mechanism is non-linear. The microphone picking up multiple audio tones is a linear mechanism.

I seem to get a 10-Hz beat frequency in the amplitude between a 500 Hz and a 510 Hz tone (2:1 ratio).
Bob S

 

[m718]

Out of phase signals cancel. That's just how audio signals combine, it has nothing to do with you're amplifier. It's already happened in the air before it even gets to your microphone so get over it (or state more precisely what you're trying to achieve so someone can help you).

BTW. If the signals are not the same frequency then they are are additive, but not in the simple way you're expecting. If you have a 600mV (rms) tone and a 6mV tone at another frequency then the combined rms voltage is \sqrt{(600^2 + 6^2)} which is only about 600.03mV. Note however that although this looks almost unchanged (from the original 600mV) it does however contain precisely the sum total of the power of the 600mV tone and of the 6mV tone.

I tried this in simulation (TINA TI) two voltages, 600 and 6 connected to one opamp the total shows 606mV. And if in rms the total comes to 600.03 would the peak be 606mV?

I made a gaussmeter and I noticed that in the presence of large signals it didn't sense smaller magnetic field changes that it would when it was not in a strong field, its a coil sensor meter, so I was wondering why it works like this.

 

[m718]

I did a new test and it doesn't even have to be 600mV, the 6 mV signal can't be seen even with a 30 mV background signal.

 

[m718]

maybe its my test equipment I will measure the output another way and see what happens

 

[uart]

I tried this in simulation (TINA TI) two voltages, 600 and 6 connected to one opamp the total shows 606mV. And if in rms the total comes to 600.03 would the peak be 606mV?

I made a gaussmeter and I noticed that in the presence of large signals it didn't sense smaller magnetic field changes that it would when it was not in a strong field, its a coil sensor meter, so I was wondering why it works like this.

Again let me summarize. Ideally 600 mV tone + 6 mV tone gives :

1. 606mV if same frequency and in phase.
2. 594mV if same freq and out of phase.
3. sqrt(600^2 + 6^2) = 600.03 if either same freq and quadrature phased or different frequencies.

 

[uart]

I did a new test and it doesn't even have to be 600mV, the 6 mV signal can't be seen even with a 30 mV background signal.

Expected is sqrt(30^2+6^2) = 30.6mV, so just how accurate is the measurement anyway?

 

[uart]

I seem to get a 10-Hz beat frequency in the amplitude between a 500 Hz and a 510 Hz tone (2:1 ratio).
Bob S

Hi Bob this has been discussed several times here recently so I'll only be brief. Technically you only get sidebands when you multiply two tones, not when you add them. Yes I know that the combination looks like there is a 10Hz component there but consider this, if you applied that waveform to a linear system with a resonance at 10Hz then I can tell for certain that it wouldn't resonate it.

Note to M718. Look carefully at Bobs wave form. He adds a "2" and a "1" but he doesn't get a "3". The RMS of two individual signals is 2/sqrt(2) and 1/sqrt(2), note however that the rms value of the resulting waveform is NOT the sum (which would be 3/sqrt(2)), the true RMS value of that waveform is actually sqrt((2/sqrt(2))^2 + (1/sqrt(2))^2) = sqrt(2.5).

 

[Bob S]

To clarify the thumbnail in my post #15, here in this thumbnail are both a sum-and-difference (+), and a product (x) modulation of a 500-Hz signal with a 510-Hz signal. The amplitudes of the two signals in both plots are the same.
Bob S

 

[m718]

I have one of those soundcard spectrum analyzers its called VA it has a db meter. My question is what is the db meter measuring, just the voltage applied to the mic input? if so is it measuring the RMS or PEAK voltage? because unlike the ac voltmeter I was using before this db meter shows accuratly even the small signals in the presence of large signals.

 

[uart]

The dB it is reporting will be 20*log10(voltage).

Can you give some numerical values of what your dB meter is reporting.

 

[m718]

The db depends on the mic input sensitivity but with loud speaker 300mV its -9.6 db with other speaker at 6 mV ON it goes up to -10.1 db

 

[m718]

The dB it is reporting will be 20*log10(voltage).

Can you give some numerical values of what your dB meter is reporting.

How would I make a db meter like this but without software and using computer sound card,
just an analog circuit with seven segment leds to show db level?

the software name is visual analyzer 8 if you need to see it.

 

[uart]

The db depends on the mic input sensitivity but with loud speaker 300mV its -9.6 db with other speaker at 6 mV ON it goes up to -10.1 db

I hope you realize that -10.1dB is actually less than -9.6dB (0.5dB less to be precise).

If -9.6dB corresponds to 300mV then -10.1dB is only 283mV.

 

[m718]

I hope you realize that -10.1dB is actually less than -9.6dB (0.5dB less to be precise).

If -9.6dB corresponds to 300mV then -10.1dB is only 283mV.

I checked again I had it wrong its the other way around and its about .3db difference not .5. So how come this voltage difference can't be seen with my multimeter? is it the meter?

 

[Bob S]

I checked again I had it wrong its the other way around and its about .3db difference not .5. So how come this voltage difference can't be seen with my multimeter? is it the meter?

The multimeter is a linear scale, while the dB meter is a logarithmic scale;
dBV= 20 Log(V) as pointed out above by uart.
The log scale is much more sensitive to small percentage changes over a large dynamic range than a linear scale.

Bob S

 

[m718]

The multimeter is a linear scale, while the dB meter is a logarithmic scale;
dBV= 20 Log(V) as pointed out above by uart.
The log scale is much more sensitive to small percentage changes over a large dynamic range than a linear scale.

Bob S

What kind of circuit do I have to connect to my opamp output to get the logarithmic scale?
I basically want the output I get on the computer soundcard db meter except without using the soundcard, an analog circuit that I can just connect to my opamp output, and 7 segment display for the db level.