Three in Software

Drawer ANSI/AIM Code 128 in Software Three

Low-Cost Variable-Bias VGA, Reverse Gain Control (Fig 3129)

1 Choose RC to drop half the VCC when the transistor is at its desired gain: Vcc 2 RC = IC

AVC 05 V (LOW GAIN) TO 5 V (HIGH GAIN)

2 Choose an RB of 10 k 3 Find the voltage required at the AGC (AVC) port of the amplifier that causes the base current (IB) to create a collector current for the desired full gain The transistor s characteristic curves in its data sheet will contain the information on IB versus IC, while the current gain graphs will show the IC versus hFE: AVC = (I B 10K ) + 07 4 Choose the limits of AVC voltages that will supply the required range of gains by substituting the desired IB in the above equation with the minimum and maximum IB related gain values 5 Design the transistor s input/output matching networks for the desired bias at maximum gain and for the frequency of operation 6 The gain of the circuit is now controlled completely by the AGC voltage at the transistor s base But since gain is managed by altering the IC through this AGC at the transistor s base, then the input and output impedances, as well as the stage s stability, will also vary Distortion and gain compression may also occur with strong input signals

MMIC VGA (Fig 3130)

1 Most MMIC gain is moderately affected by a change in Id By looking at the device s Id versus S21(dB) curves, this susceptibility can readily be seen, and offers an easy way to operate some MMIC s as variable gain amplifiers Gain variations of 5 to 15 dB are possible, depending on the MMIC and the frequency, by varying Id through an AGC circuit

2 Utilizing a MMIC as a VGA should be adopted only for low-level signals, since the P1dB will decrease along with the Id and gain of the MMIC The exact value of the overall circuit gain variations obtained will differ with input frequency 3 Design the MMIC stage as described under Sec 373, and then permit the AGC circuit to control the MMIC s Vdd

To prevent one stage s DC biasing from affecting the next stage, a method of coupling an AC signal into or out of an amplifier, while blocking DC, must be employed The actual type of coupling for discrete circuits, sometimes combined with impedance matching, will depend on the sort of signal being amplified: DC, (low frequency) AC, RF, or wideband amplification To prevent unwanted amplifier oscillations and/or RF contamination of the system power supply, decoupling components, such as capacitors and chokes, that are capable of shunting the entire spectrum of RF and AC frequencies should be used for all active devices

Heavy decoupling of a power supply is not only required in order to prevent RF energy from entering the supply, but also to provide a reservoir of energy that may be instantly used by any active circuit on the PCB, especially high current stages such as a PA This requirement can be better appreciated if one understands that when a 24-GHz power amplifier is in operation, it is actually drawing DC current directly from the power supply at a frequency of 24 GHz This alternating current draw must be filtered before it can adversely affect other stages, as well as to provide instant current smoothly to the PA stage Thus, the decoupling capacitors themselves will act as a vital charge supply for the PA, as well as function as an RF bandstop filter for any conducted EMI Decoupling networks must not only be comprised of RF ceramic capacitors that are capable of attenuating higher frequencies, but also of the electrolytic type, which possess a much higher capacitance value for the filtering of extremely low frequencies In other words, so that we may send all frequencies to ground that may attempt to enter, or to exit, the DC power supply, the supply must be strongly RF isolated (decoupled) from the PA by a wide frequency band of various shunt capacitors of different values Series inductors and resistors help in this filtering action as well, and can be used in concert with the decoupling capacitors To function as desired, decoupling capacitors must not be used near their parallel (high impedance) resonant mode, while decoupling inductors must not be used near their series (low impedance) resonant modes If they are, then they may literally disappear from the decoupling circuit, and become completely ineffective Inductors are far from perfect components, and possess parasitic capacitances So when an amplifier must function properly across a very wide band of frequencies, two decoupling RFCs may be required; a low inductance coil that works at very high