Bass Beat Detector Help

[tufty88]

Does anyone know the dynamic range of this circuit? Or how to calculate it. The circuit attached quotes greater than 50 db. I want to compare the dymanic range of both of these circuits but not sure how to.

https://www.electro-tech-online.com/attachments/agc-amp-jpg.14022/

Cheers
Jonathan

 

[tufty88]

Sorry not very clear, the circuit I now want to know the range for is this one:

https://www.electro-tech-online.com/attachments/schem-v2-0-png.50064/.png

 

[audioguru]

The circuits reduce the dynamic range of music or speech which is not what you want from a bass beat detector circuit.

The circuit with the Jfet doing the dynamic range reduction is too simple.

The circuit with the transistors and diodes is not good.

 

[tufty88]

In my application, I need a circuit that provides a consistent output signal for a varying input sound level detected through a microphone. The signal is then filtered, peak detected, and A/D converted through comparator and timer to give a blinking LED in time to beat of music (like the circuit posted originally in this post). The rest of the circuit works great, but have tried both these AGC circuits to limited success. The next one I was going to try was your recommended one audioguru: **broken link removed**.

 

[tufty88]

What I've been doing is using a high gain on these AGC amplifiers - in the premise that for low volume sounds, the high gain will amplify the signals to suitable level, but for louder sounds the AGC kicks in and reduce it only care about frequencies below 160 Hz so gain bandwidth limit of op amps isn't really an issue, so I tried using a gain of 2 Meg. I noticed with the FET version, that it would work well, and then cut out momentarily, and repeat this- I wonder if this is something to do with the attack cutting out.

 

[tufty88]

**Sorry gain of 2000 not 2 Meg. I'm wondering now if I can use a high gain pre amp, that lets me detect both distant and near speaker sounds, but then feed it through second AGC or limiter stage, that attenuates higher range of outputs.

 

[audioguru]

A compressor/limiter reduces the dynamic range of music so all sounds including the bass beats will be at the same levels.
Then you might not be able to detect bass beats.

 

[tufty88]

Audioguru,

How about if I were to amplify a microphone signal using a pre amp, and then bass filter, before passing signal through the fast audio limiter. Might this enable detection of bass signals at a fixed level, whilst filtering higher frequencies?

 

[tufty88]

On second thoughts I can see why that would level out attenuated higher frequency signals.

Might this be more suitable?

Microphone--pre amp--fast audio limiter--steep roll off bass filter-- peak detector

Any feedback welcomed

Thanks

 

[audioguru]

I think you want to use a fast audio limiter as an automatic volume control.
It should have a fast attack time and a slow release time.

Limit the wideband audio, filter away the high frequencies then peak detect the bass beats.

 

[tufty88]

Thanks Audioguru,

I'm going to try using a narrow bass pass fliege filter. Will post the results.

 

[audioguru]

A narrow bandpass filter produces "one note bass" or in your case it will show "one note bass" like old boom-boxes.
Instead you need a fairly sharp lowpass filter.

 

[alec_t]

Even a simple low-pass filter may suffice. Try this:-


Edit: This is intended for line-level inputs ~1V peak

 

[tufty88]

Thanks guys,

I have built the fast audio limiter and it works brilliantly so far.
I'm getting a nice clean bass response using a 6th order sallen key low pass filter after audio limiter ( I've chosen 6th order as I have spare op amps in my circuit).

The problem I have now, is that my output signal I believe is too low voltage to transfer through the peak detector part of the circuit. Due to diode voltage drop. I think.

The circuit I use for peak detection is outlined here

https://www.electro-tech-online.com/custompdfs/2013/07/matt_aldrich_eeng236b.pdf

Alec,

I think your suggested circuit will sort this issue

My supply voltage is 6V, and I'm using LM324 which has max output of 4.5V, so I'll adjust gain to be a bit smaller to avoid saturation of output signal.

Will get more development done after work tomorrow and update with progress. Really appreciate the help here and guys and im learning a lot.

Jonathan

 

[audioguru]

Most peak detector circuits have the rectifying diode inside the negative feedback loop of an opamp so there is no diode voltage drop.

The lousy old LM324 quad opamp is very slow so maybe the beat is past before it can react.

 

[alec_t]

Aw, come on AG. Even the 324 can manage a couple of hundred Hz (as evidenced by the sim result in post #33) . We're talking bass beat, not super hi-fi.

 

[audioguru]

My customers heard telephone-quality voices played on a lousy old LM358 amplifier which is the dual of an LM324 quad. Many complaints due to muffled high frequencies and crossover distortion. I fixed the Korean telephone system.

But I agree that an LM324 can be used to blink an LED to the bass beat of music. Our vision will not see the crossover distortion.

 

[alec_t]

Our vision will not see the crossover distortion.

For this application I would have thought even 50% distortion would be tolerable .

 

[tufty88]

Thanks guys- I have ordered some op amps with a larger slew rate to try out effect.

Having said that, I was checking my circuit today and noticed I had in fact made a mistake in its wiring- even though I was getting what seemed to be a good output- it was not an attenuated output from the audio limiter.

I re-connected the amp up to match schematic- and now the output of my audio limiter seems to be very small at 0.25V or so. This is using a 9V battery supply.

I tried using a 13V supply and the size of this output increased dramatically- this enables me to process the signal through the filter.

Is there an obvious way to change the output limit of the fast audio limiter without increasing supply voltage?

Perhaps a MOSFET with lower VGS turn on voltage?

Cheers
Jonathan

 

[audioguru]

We do not know which opamp you used and we do not know your schematic.
Since you are using a single-polarity supply then the opamps must be biased properly. We do not know how they are biased.