5Watter Organic Transceiver Build Phase 2: Audio Amplifier and AGC

Last time around, I built the power supply and had an absolute blast discovering why components were placed in their particular way. This time, I’ll be building up the audio amplifier section while performing research as I did in Phase 1.

In general, this section revolves around placing an LM386, the requisite components that make it function properly, and the Automatic Gain Control, or AGC, circuitry.

I understand the LM386, but my knowledge of the AGC is limited. Please take everything I say about it with a grain of salt and please correct me if I’m wrong!

Components

• J6 – a headphone jack
• U4 – LM386N audio amplifier IC
• R27 – 4.7 ohm resistor
• R30, R31 – 4.7k ohm resistor
• C50 – 33uF electrolytic capacitor
• C47 – 10uF electrolytic capacitor
• C44, C45, C46, C48 – 100nF monolithic capacitor
• C51, C59 – 1.0uF monolithic capacitor
• D9, D13 – 1N5711 Schottky diode

Circuit Explanation

The LM386 is much maligned, yet ubiquitous. As I was doing my research, I was surprised at the variation of opinions — and therefore, circuits — that exist about this IC. Wading through all this information was daunting, but I believe that I have a few takeaways after all this sorting.

To tackle this daunting task, I will rely heavily upon the datasheet provided by Texas Instruments.

Pins 1 and 8 provide a gain control for the LM386. If pins 1 and 8 are open, then the gain is set to 20dB. If a capacitor (C47, 10uF) is placed between 1 and 8 as it is in the 5Watter, then the internal 1.35kOhm resistor is bypassed, setting the audio gain to 200dB. If a resistor is placed in series with a capacitor between pins 1 and 8, then the gain can be set between to any value between 20dB and 200dB.

Practically, this means that by altering the components at C47, the gain of the audio circuit can be experimented with. I can see this being useful for either headphones or for including an integral speaker.

Some elements of the LM386 are not utilized in the 5Watter. For example, there are no resistors or capacitors between pins 1 and 5, meaning the LM386 provides no bass boost to the audio signal. Maybe if your sounds too tinny, then some bass boost could be provided yourself. However, incorporating this mod could be a mistake!

The circuit output of the LM386 dumps you out into an AGC and to the volume control and/or speaker.

---

At this time, I’m still making sense of the AGC, but I’ve gained some insight while reading the following:

• ARRL Handbook (pages 12.23-12.26)
• Experimental Methods in RF Design by Wes Hayward W7ZOI, Rick Campbell KK7B, Bob Larkin W7PUA (pages 6.15-6.22)
• Homebrew Cookbook by Eamon Skelton EI9GQ (pages 51-54)
• Building A Transceiver by Eamon Skelton EI9GQ and Elaine Edwards G4LFM (pages 72-78)
• Automatic Gain Control (AGC) in Receivers by Iulian Rosu YO3DAC/VA3IUL

I’m glad I accumulated a library before embarking on this endeavor!

Automatic Gain Controls, AGCs, come in two flavors, IF and Audio. Given that this one is located after the LM386 audio amp, I think it is an audio derived AGC system but I’m not sure. This is beneficial because signals that make it through the radio’s filtering are the only ones that will be reduced. In the IF derived system, strong adjacent signals can trigger the AGC, causing the station you’re hearing to disappear without explanation. This phenomenon is known as AGC pump.

Diodes D9 and D13 (1N5711) act as the AGC detectors. I believe that these diodes limit voltage which reduces overly strong signals. C59 (1.0uF) is for AGC attack time, or how quickly the AGC kicks in, while R31 (4.7k) is for decay time, how quickly the AGC returns volume to normal. These components are connected before mixing blocks 2 and 3, being inserted into Q6 and Q8 (MPSH10) which are IF amplifiers.

Testing

After I installed all the components, I tested the circuit. I touched the pads of R25 and R26 and sure enough, the speakers I connected hummed. It’s working! Moreover, the amp draw increased to 15.22mA which is also a good sign. Before testing the amperage this time around, I added an inline fuse. I’ve decided to go with the 5x20mm glass fuse standard as they come in small values ranging from .1A to 20A.

Researching the blocks of this radio is more difficult than I imagined. At times it feels as if I’m drinking from a firehose but that’s okay. I’m learning a great deal.