An S-Meter and AGC circuit

Don ND6T has recently installed a 20 dB RF AGC modification in the BITX40.

He has not installed it in the uBITX yet but intends to do so soon.   It’s a simple circuit and replaces the S meter circuitry, too.

Most BITX automatic gain control schemes use the audio output to apply control of the input of the audio power amplifier. This depends upon the volume control setting and introduces considerable distortion on high level signals. By using a signal source before the control then we can use the constant fixed gain of the receiver as a good indicator of signal strength and still adjust the speaker or headphone levels for the best comfort.

This simple project uses a single stage amplifier to tap into the audio at the input of the volume control, rectify it to a DC level, filter it, and use it to control a MOSFET as a shunt across the receive RF path. This project assumes that you have already installed the RF gain control described here (for the BITX40 or here (for the uBITX) and bridges across it at the control potentiometer.

R3 and C3 are used to not only filter out the audio component, but to form a “fast attack, slow release” control signal. That means that, when a strong signal appears, RF gain will be quickly reduced but will take a second or so to restore to full gain after the signal stops. This avoids “pumping” during a single sideband transmission but is fast to react to very loud signals.

Nearly any general purpose NPN bipolar junction transistor will work as Q1 as long as it has a beta of more than 100. 2N2222, 2N3904, etc. will work quite well. Q2 can be a 2N7000 or a BS170. None of the component values are critical. The 5 volt supply makes it easy to use any part with more than a working limit of just 6 volts and the current drain is low enough to be negligible.

I have found that most 2N7002 transistors will yield at least 20 dB of RF attenuation across the HF spectrum at 50 ohm impedance. Attenuation begins around 1.4 volt bias on the gate referenced to the source and provides maximum action at around the 3.7 volt level. Very effective for a single, simple, and inexpensive device. A great first step.

A more fulsome article with construction details (using surface mount components) is posted on nd6t.com.

This simple circuit led to a discussion on the BITX20 IO Group list, started by Jerry KE7ER, about the BAP64Q pin diode attenuator.   This gives 60dB of dynamic range:  https://groups.io/g/BITX20/message/32066

Attenuating back in RF gets around the limited dynamic range that Henning points out in the first post of that thread.    Note that the control voltage is inverted with respect to the 2n7002 FET, higher voltages give less attenuation.   You could get a slightly lower noise figure for the receiver if the attenuator was inserted at a later stage of the RF chain.

Jerry observes, “The BAP64Q is relatively expensive at $0.50 single piece,
the frugal among us will note it’s down at $0.20 if you buy a few thousand.
Mouser and Digikey both stock it, Mouser points you to the wrong BAP64* datasheet.   There are other similar small signal pin diode attenuators out there from other manufacturers.”

Reference

BITx40 – Essential Mods

Fantastic BITx40 build from Glenn VK3YY

Jerry KE7ER notes that the schematics on the hfsignals.com website for the Bitx40 are final but not necessarily correct.  He then goes on to talk about some of the variances and essential hacks.

This website is dedicated to the uBITx.   However, so that everybody who also has a BITx40 doesn’t miss out entirely on the occasional “gold nugget”,  some interesting articles about the BITx40 may appear from time to time on this website.   Jerry’s comments on the variation between the published circuit diagram and the actual board currently in production meets the “gold nugget” standard.

Differences between the BITx40 Board as manufactured  and the manufacturer’s Circuit Diagram

  • The two modulator diodes D15,D16 plus balancing pot R106 got replaced by a BAT54s dual schottky:  https://groups.io/g/BITX20/message/33385
  • All those BC849’s?  Have always been MMBT3904’s.
  • L5 and C103 of the BFO are missing, the BFO is at just about the right frequency if the crystals are matched without any adjustment.  Usually.
  • L4 in the VFO is not stuffed on the board, meaning the analog VFO doesn’t oscillate.  The VFO comes instead from the si5351 PLL chip on the Raduino through the Bitx40 connector labeled “DDS1” in the schematic.
  • T6-4 is tied to C142 and R144, U3-3 is tied only to the 12v rail labeled “TX”, schematic does not show that clearly.
  • RV1-2 is not connected to U3, the schematic drawing is just a little tight there.

    Otherwise, Jerry thinks that the schematics on HF Signals are correct, reflecting what is currently being shipped.   Nothing has changed in Bitx40v3 boards shipped since Dec of 2016.

Changes that should ideally be made to the BITx40 Board

He then goes on to note that he thinks there are a few things that perhaps should change:

  • D7 and C130 should be across the coil of relay K1,  D8 and C164 are correctly across the coil of K2
  • R141 is currently a 1/4 Watt 10 ohm resistor, should be 1/2 Watt as have been reports of it burning out.
  • The surface mount MMBT3904 at Q13 could be replaced with a through-hole version of the same part, the 2n3904, as it must dissipate a fair bit of power during transmit.  Even with the radio off but hooked up to an antenna, a nearby QRO transmitter or electrical storm can blow Q13.

Essential mods

And finally Jerry suggests the following as essential mods for all purchasers:

Note that he suggested adding an incandescent lamp, which can be quite effective:
https://groups.io/g/BITX20/message/21901