I need a circuit to detect and filter 10 KHZ sound

[spec]

I suggest it would be much simpler and better to use a bandpass filter which does not have the problems of the 567. Also the filter gives the output signal and amplitude, so it is more flexible.

In the filter of posts #9 and #12, it is impossible to do a simulation with out specifying the opamp: loading, slew rate, input impedance, output voltage swing and output current capability all come into the equation. What opamp type is specified?

spec

 

[spec]

If you define your frequency tolerance of 10kHz, this would define the maximum Q of a simple 2nd order BPF toplogy shown. If you want a variable Q variable F and fixed gain and variable or fixed threshold, you better define it first, before we get another batch of almost solutions.

Goods specs is crucial to any design. Include everything relevant, such as say data rate and group delay distortion.

The variable Q with constant gain used a pot in the feedback loop in a topology where one R could tune the Q while another could tune f but the threshold was fixed and adjacent note rejection could be 30 dB as I recall so the Q could be between 20 and 100.mThe higher the Q the smaller your f error tolerance will be and the more stable the components must be.

... so you must give better specs for passband gain, BW, band reject levels , threshold and duration levels for triggering etc. You may also want to specify error rate for true and false detection rates as anything can go wrong, it will. This will determine your minimum SNR of the signal & Note, that a loud snap is broad band noise may trigger it.

Hy Tony,

If the OP could define the filter parameters down to the N th degree he would hardly need to be asking how to design a bandpass filter on ETO- he would be an expert and be able to design a better filter than us.

spec

 

[schmitt trigger]

Its been over three days after the OP asked his original question and has not come back.................

Anyways, to his original request:

use simple electret microphone with pair of 2N2222 transistors to detect a 10 Khz tone

the simple answer is no, you can't without additional circuitry.

 

[spec]

Its been over three days after the OP asked his original question and has not come back.................

Anyways, to his original request: the simple answer is no, you can't without additional circuitry.

You said it- that is a simple answer

Shame it is so late too!

spec

 

[Nigel Goodwin]

Navid ? Please help answer questions about accuracy of frequency and noise interference levels

Answer questions?, he's not even provided the slightest detail in his original question, we're over 20 posts now and no one has the slightest clue what he might be doing, or wanting to do

 

[spec]

The OP has given 'likes' for the three posts that are to the point and a help in this thread. He did miss post #5 10, #11 and #12 though

Now you have also started firing bullets from the side lines- for what purpose

spec

 

[MikeMl]

In the filter of posts #9 and #12, it is impossible to do a simulation with out specifying the opamp: loading, slew rate, input impedance, output voltage swing and output current capability all come into the equation. What opamp type is specified?

It is absolutely possible to do simulation without specifying a specific opamp. I just did it. And the results I show will be attainable with about 90% of the IC opamps that can be purchased from a supplier like Mouser or Digikey...

The circuit has to "work" with an idealized opamp before you worry about non-ideal behavior... The circuit I presented requires nothing special from the opamp, except GBW product, which should be at least 1e6. At 10kHz, about 90% of the commercially-available IC opamps will work just fine...

 

[spec]

It is absolutely possible to do simulation without specifying a specific opamp. I just did it. And the results I show will be attainable with about 90% of the IC opamps that can be purchased from a supplier like Mouser or Digikey...

The circuit has to "work" with an idealized opamp before you worry about non-ideal behavior... The circuit I presented requires nothing special from the opamp, except GBW product, which should be at least 1e6. At 10kHz, about 90% of the commercially-available IC opamps will work just fine...

I don't know what the relevance of a GBW product of a million has. Very few common opamps have much gain at the active region of the filter centered on 10Kz and not to mention a gain of 10 which you have chosen for some reason or a compliant slew rate. If you have done a simulation with an idealized opamp how much help is that to the OP who may like to build the circuit. Also, many common opamps are limited on output current and swing and this needs to be checked by simulation if you like or by analysis. Filters are notorious for causing overloading problems, not to mention instability in the frequency domain.

And while I am having a rant, I simply cannot see the point of posting another filter- especially when I had already posted a filter design- with such a low Q and out of band attenuation. A Butterworth response would be the last choice for this application. Although it is a nice flat pass band, gentle filter for HiFi.

spec

 

[MikeMl]

...I simply cannot see the point of posting another filter- especially when I had already posted a filter design-

That is your real problem, isn't it?

Your design did not "detect" anything, it just "filtered" it.

 

[spec]

That is your real problem, isn't it?

I don't have a problem- you are trying to avoid the issues by ad hominem attacks.

Your design did not "detect" anything, it just ? "filtered" it.

That is exactly what I said. For all we know the OP might need an output at 10Khz that represents the magnitude of the 10Hz signal. Further more, I said that, if required, we could provide a DC proportional to the 10K Hz signal- a trivial addition.

I cannot understand your point of view!

spec

 

[AnalogKid]

A Butterworth response would be the last choice for this application.

Or the only (two-pole) choice. The filter topologies in post #4 and post #9 are identical.

ak

 

[spec]

Or the only (two-pole) choice. The filter topologies in post #4 and post #9 are identical.

ak

I am no filter expert, but it is my understanding that the topology does not define the transfer function, but the relative position of the poles does. The component values are quite different between the two filters as is the performance.

I did not state any transfer function (it is a MFBF) but the other party claimed a Butterworth response, which as an objective would not be optimum as steeper sides with some in-band ripple would be better for a bandpass filter in this application. As it happens the second filter is not a Butterworth filter anyway, according to post #10 and #12.

spec

UPDATE 2016_07-02
As the MFBF topology only has a single slope for its low and high pass sections it is not possible to have any characteristic transfer function: Bessel, Butterworth, Chebyshev, elliptic etc etc. Some of you saw this from day one.

 

[Tony Stewart]

I am no filter expert, but it is my understanding that the topology does not define the transfer function, but the relative position of the poles does. The component values are quite different between the two filters as is the performance.

I did not state any transfer function (it is a MFBF) but the other party claimed a Butterworth response, which as an objective would not be optimum as steeper sides with some in-band ripple would be better for a bandpass filter in this application. As it happens the second filter is not a Butterworth filter anyway, according to post #10 and #12.

spec

Spec "Butterworth filter" means maximally flat magnitude filter and the slope is user selectable.
There are dozens of other BPF filter types. Raised Cosine, Gaussian, Elliptical, Chebychev, Cauer, double tuned tank, .......these filters have different features like min ripple, linear phase, minimum group delay distortion, minimim InterSymbol Interference (ISI), minimum loss and steepest slope with controlled phase distortion, State Variable multiple output. single sideband, FM limit with pre-emphasis, linear phase with pre-compensation , recursive BPF , BPF with odd harmonic recursive notch, etc etc...

 

[spec]

Spec "Butterworth filter" means maximally flat magnitude filter and the slope is user selectable.
There are dozens of other BPF filter types. Raised Cosine, Gaussian, Elliptical, Chebychev, Cauer, double tuned tank, .......these filters have different features like min ripple, linear phase, minimum group delay distortion, minimim InterSymbol Interference (ISI), minimum loss and steepest slope with controlled phase distortion, State Variable multiple output. single sideband etc etc...

Thanks Tony- as I said I am no filter expert- afraid that at work anything that was slightly difficult in that area I just asked specialists to do- They made some fantastic filters which worked a treat, unlike some filter designs. One of the big issues was component value sensitivity.
I still have a copy of Williams and Taylors's 'Electronic Filter Design Handbook' which used to be the filter bible in the old days.

spec

 

[audioguru]

Did you notice the Multiple Feedback Bandpass Filter graph with dB curves in post #12 shows the very gradual highpass and lowpass slopes at only 6dB/octave (20dB/decade) which is only single-order?

 

[AnalogKid]

Why would it be anything else?

ak

 

[MikeMl]

The filter I posted is a 2-pole Butterworth Bandpass, implemented using the MFB opamp topology. Since it uses only two poles, the roll-off on either side is attributable to a single pole, hence the slope is only 6db per octave. This filter is low-Q, so is not likely to ring if the noise is pulse-like...

 

[audioguru]

The Multiple Feedback Bandpass filter is too simple "to filter other sounds below 8 khz and upper than 11 kHz". The Q of it affects only the peak bandpass, not the slopes.

 

[spec]

The Multiple Feedback Bandpass filter is too simple "to filter other sounds below 8 khz and upper than 11 kHz". The Q of it affects only the peak bandpass, not the slopes.

Yes, you are quite right- I have been doing some filter revision

A ceramic resonator should make a decent bandpass filter: https://www.murata.com/~/media/webrenewal/support/library/catalog/products/filter/cerafil/p11e.ashx

spec

 

[audioguru]

An LMF100 dual switched capacitor filter IC is being discontinued but is still available. Its two sections can be seriesed high Q bandpass filters.