Hi guys
I am not good at analog, but we have a need to connect multiple audio feeds with different signal strength. So we need an AGC circuit, unfortunately I can't find a chip to do just that so I attempt to design my own.
So I start with LTSpice with ideal parts, and here I am at, it seems to work over simulation. Can you tell me what have I done wrong, what can be improve on etc...
LTSpice simulation file is attached.
Thanks guys
Don't know what your fidelity requirements are, but if you're looking for a good quality dedicated chip for it, THAT semiconductor makes a few - you basically just need the rectifier and cap and a resistor divider to make sure the gain input is set correctly. Otherwise, they can be found looking for a compressor or compader circuit, there are a few ICs available.
I know it's possible to make them with a JFET to attenuate the input signal, but it has to be biased correctly to stay in the linear region, so there's a limited signal that it can variably attenuate.
Don't know what your fidelity requirements are, but if you're looking for a good quality dedicated chip for it, THAT semiconductor makes a few - you basically just need the rectifier and cap and a resistor divider to make sure the gain input is set correctly. Otherwise, they can be found looking for a compressor or compader circuit, there are a few ICs available.
I know it's possible to make them with a JFET to attenuate the input signal, but it has to be biased correctly to stay in the linear region, so there's a limited signal that it can variably attenuate.
I would rather use a chip if I know what chip to use, it's part of an push to talk intercom, just need to be good enough for conversation.
Is this the compressor circuit you are referring to:
https://www.electroschematics.com/9400/audio-compressor-agc/ I did try using a JFET, but I have a hard time get it to work properly. Basically I can't get a JFET to work in my circuit.
I've done this before, see the thread linked below. Unfortunately the ADC requirement was scrapped, so it never went any further.
For ICs, the NJM2783 or MAX9814 are good. The discrete design also worked well and had the advantage of being able to work with a two wire phantom power interface.
https://www.eevblog.com/forum/projects/mic-pre-amp-circuit-discrete-or-ic/msg1410745/#msg1410745
Thanks for the link the and IC suggestions, learned a few things there. I think the MAX9814 will work well for me.
I would rather use a chip if I know what chip to use, it's part of an push to talk intercom, just need to be good enough for conversation.
Is this the compressor circuit you are referring to:
https://www.electroschematics.com/9400/audio-compressor-agc/
I did try using a JFET, but I have a hard time get it to work properly. Basically I can't get a JFET to work in my circuit.
That's the idea, don't have a specific schematic in mind, but I've had similar trouble designing ones - I think you've got to pick bias values carefully since the ohmic region of the JFET is fairly narrow, and then you have to make sure the input signal is pretty small - think mic level vs. line level, because too large a signal will mean a lot of clipping when you go outside the operating area.
Anyways, for chips, the ones I were talking about are called "dynamics processors" by THAT
http://www.thatcorp.com/THAT_Dynamics_Processing_ICs.shtmlI've used the THAT2181 for some high fidelity variable amplification, and while it requires an external amp (the VCA is a current mode device) and a rectifier, the circuit isn't hard to make work and the datasheet is pretty good. Other chips in this group offer similar functions, sometimes bundled with other stuff, with lower fidelity, which should be fine for many applications.
thanks for all your help guys!!