Audio Design Question - Equilizer, Echo and Compression system.

[especialista]

Hello, my dear friends.
Need some advice about audio design.

I'm trying to design a prototype for high gain microphone for my HamRadio stuff.

It involves the following IC's (Datasheets available here:  https://www.upload.ee/files/12308860/DataSheets.rar.html)

LA3607 - 7 Band Graphic Equilizer
SSM2166 - Microphone Preamplifier with Variable Compression and Noise Gating
PT2399 - Echo Processor

I have in my mind that the correct sequence of events would be:

Microphone - Equalizer(LA3607) - Echo(PT2399) - Audio Compressor/Gain(SSM2166)

I'm not quite sure about that sequence and would like to hear some thoughts.

Thank you.

[pickit2]

just on audio i would say Mic preAmp, echo, then Equalizer.
but as a working rig, mic preAmp & Equalzer.

[fpgaguy]

There was a rather interesting discussion on interface to  microphones from eevblog/dave jones/rode engineer - It's up on youtube maybe search eevblog #602

it's covering microphone types / directionality / and the first level amplifier considerations and not the subsequent signal chain

[rtm]

I am also a HAM-Radio operator, and this theme is close to my heart.

There are 2 possible targets with using such voice processor in HAM-Radio station:

1. Making a fancy signal for looking funny.
2. Making aggressive signal to improve readability of your signal in contests and DX-ing or in long distance communication.

Echo:
You do not need an echo in second case, as echo is an useless noise in your signal. But a lot of radio operators are using
echo in short range communications as Echo is adding a heaven style to your voice. I made such reverberator near 40 years ago for my station, but I used it for short time - it was useless. BTW it was a quite complicated device with about 30-40 chips on the board, and it is very amazing to see such device in a small chip with a price less than 1$ now ))

Compressing/Clipping:
There are different types of voice compressing. First and simple one is an automatic level control with some delay in control loop. Second type is a clipper or a limiter. It also can be simple (adding harmonics to your signal during processing) as diode limiter or complicated - making big limiting without adding harmonics. If you are planning to use it as a fancy device it will be Ok to use your SSM2166 voice compressor - it is a simple type, but if you want to improve the readability of your signal in heavy conditions, you need to use other circuits. During 70th-80th we were using High-Frequency Clipping, when a microphone signal was converted to SSB on intermediate frequency (for example 500 kHz), next it was limited there (all harmonics were gone to 1 MHz), cleaned by filter, and converted back to voice band. Also, there were other devices with effective clean limiting. I can give you a link to simple circuits of such devices if you need it. One thing about effective signal clipping: powerful part of the signal is suppressing a small part. So, after clipping in the output you will have less high frequency sounds than in the input.

Equalizer:
When you speak, your voice is not going directly to your microphone, it is going through your room, and the room has its resonances. So, the signal in microphone has not linear spectre. You need to correct room's resonances. You are using an equalizer exactly for this. Also it will be good to have a second equalizer to correct spectre after high frequency suppressing in the limiter.

So, correct universal circuit will be:

1_Microphone->2_Low_noise_amplifier->3_Equalizer1->4_Compressor->5_Clipper/Limiter->6_Equalizer2->7_Echo_processor

You need to add switchers for Compressor, Clipper/Limiter and Echo_processor here - in this way you will have a universal voice processor for all cases.