Very simple Audio compressor for FM transmitter

[dr pepper]

I have something very similar, only the gain control device is a depletion mode fet not a bipolar tranny, it works well, I can have the mic on my lap and still use it, totally useless in a noisy environment though.

 

[audioguru]

Thanks fort the very old schematic. I used a Jfet to replace the transistor when I made a compressor about 40 years ago.

 

[MrAl]

Hi again,

I intended to make one for the TV because of those pesky, loud commercials, but didnt get to it yet.
I got a couple of those linear opto isolators for the job, the ones with the purely resistive output element. They would be ideal for audio but i havent checked to see if the response is too slow or not yet.

 

[Willen]

Hi again,

I intended to make one for the TV because of those pesky, loud commercials, but didnt get to it yet.
I got a couple of those linear opto isolators for the job, the ones with the purely resistive output element. They would be ideal for audio but i havent checked to see if the response is too slow or not yet.

Passing audio through opto isolator? What is input and output component, which would work as audio speed?

 

[AnalogKid]

I intended to make one for the TV because of those pesky, loud commercials, but didnt get to it yet.

Your results may vary, but - most TV commercials are not louder because the peaks are larger, but because the audio is heavily processed and compressed (Audimax, Volumax, etc.). A peak-detecting compressor/limiter probably will not do much.

ak

 

[MrAl]

Your results may vary, but - most TV commercials are not louder because the peaks are larger, but because the audio is heavily processed and compressed (Audimax, Volumax, etc.). A peak-detecting compressor/limiter probably will not do much.

ak

Hi,

Yes i forgot about that aspect of it. But earlier on they were actually coming on louder.
I also notice lately in movies there are a lot of soft passages too where the actors are talking and i cant hear them because they are whispering a little. That's annoying too.

 

[MrAl]

Passing audio through opto isolator? What is input and output component, which would work as audio speed?

Hi Willen,

Well the speed of the audio wave itself is not an issue, it's a pure resistive element, a resistor, an LDR.
The LED internally shines on the LDR thus changing the resistance.

The speed of the audio can be almost anything, it's the speed of attach and decay that i was thinking of. If that is too slow it might sound like breathing in the background or maybe just take too long to response. The repeatibility also comes into question, as to the actual resistance the LDR returns to after a loud sound.

A typical part number i think is VTL5C4, but there are others, and you can make one yourself with a cadmium sulfide photo cell and an LED probably a white LED. As the LED gets brighter (ie more current) the resistance goes down.

 

[MrAl]

Hello again,

Here is a circuit i found that uses the analog opto. The op amp for this circuit should be a low noise low distortion type op amp.
The main idea is that the current drives the internal LED, and the output is a variable resistance.
I am sure you can come up with other ideas too for this kind of device.

 

[ronsimpson]

VTL5C4

I wish I had that part 40-45 years ago. It would have made my life better.
The cadmium sulfide cells I used were too slow.

 

[AnalogKid]

I also notice lately in movies there are a lot of soft passages too where the actors are talking and i cant hear them because they are whispering a little. That's annoying too.

A "drawback" of digital audio is that there is about 30 dB of extra dynamic range to play with. I'm sure whispers drew you into a sense of shared intimacy - in the mix booth.

ak

 

[ronsimpson]

In real life I only get 50% of a whisper any why. So in the movies its about the same.

 

[dr pepper]

Willen, I think whats been said is that the audio isnt passing through the opto, moreover the resistive section is being used as part of a potential divider for the audio, sort of a electronic volume control.
The led side is filtered substantially and hardly any audio will be on it, the filter is usually designed to give fast attack (to prevent sudden sounds from passing) and a slow decay (to keep the sound down in case another loud sound comes along shortly after, and so as not to distort lower frequencies).
Passing audio through an opto isnt as easy as it sounds, most have a non linear frequency response, you'd have to either compensate for this using another opto or use one designed for audio, which is a special part.

 

[AnalogKid]

Most low cost compressor circuits are like the link in post #180. A bipolar transistor, MOSFET, of JFET is the shunt leg of a voltage divider, and its effective resistance changes as a function of a rectified and filtered control voltage. A common problem is that the device's resistance also is a function of the audio signal across it. This introduces some amplitude-dependent harmonic distortion. High-end circuits use a gain cell (sometimes called a Gilbert cell) or analog multiplier to form a true VCA (voltage-controlled amplifier). The circuits are much more complex than a single transistor, but do not have any of the subsequent distortion issues.

A compromise between those two ends is the optical device mentioned in post #183. This is a light source and a photo-sensitive resistor like a cadmium sulphide cell. The resistance element goes where the transistor or FET goes. Since the resistance does not vary with the audio across it, you get the low distortion of a gain cell with the simplicity of a standard transistor limiter.

ak

 

[audioguru]

I notice that today many people who are speaking end each sentence with a growl. Either they run out of breath and do not take in more air or they don't care if their voice stops before they are finished talking.
My Sound Level project has a microphone and a bar graph. It shows that the peak levels of TV programs and commercials are the same but the average level of commercials is much higher because they have too much compression.

 

[MrAl]

I notice that today many people who are speaking end each sentence with a growl. Either they run out of breath and do not take in more air or they don't care if their voice stops before they are finished talking.
My Sound Level project has a microphone and a bar graph. It shows that the peak levels of TV programs and commercials are the same but the average level of commercials is much higher because they have too much compression.

Hi again,

Yes i noticed that too. The actors are supposed to do some speech therapy that teaches them to speak in a special way so that their words are understood more clearly, but i am not sure if they still do that these days because i hear a lot of slurred speed in movies and so i cant understand some words they are saying. It's so important for each and every word to be spoken clearly in a movie because the movie depends on almost every word in the production, unlike human to human where we can always say, "what?".

BTW there is still a gain controlled amplifier out there if anyone wants to do it that way. Was this mentioned yet in this thread? Someone could look up the part number maybe.

A high quality compressor is an interesting project because we are dealing with audio, where we like to keep the distortion down yet we still have to process the audio signal which must add at least some extra distortion.

 

[AnalogKid]

MAX9814

ak