1996 - 2016
BSEE - KB3UON
RF Cafe began life in 1996 as "RF Tools" in an AOL screen name web space totaling 2 MB. Its primary purpose was to provide me with ready access to commonly needed formulas and reference material while performing my work as an RF system and circuit design engineer. The Internet was still largely an unknown entity at the time and not much was available in the form of WYSIWYG ...
All trademarks, copyrights, patents, and other rights of ownership to images and text used on the RF Cafe website are hereby acknowledged.
My Hobby Website:
So far, the basic flip-flop has used only PNP transistors. It could have just as easily used NPN transistors.
The functional operation would not change; only the polarities required for conduction and cutoff change. As a
technician, you may see either type of transistor used, NPN or PNP. A symbolic block diagram is sometimes used to
avoid confusion about voltage polarities.
A special kind of block diagram has been adopted as a standard symbol for the flip-flop and is shown in figures 3-25 and 3-26. The two inputs are represented by the lines on the left and the outputs by the lines on the right. INPUTS to a flip-flop are S (SET) and C (CLEAR) and OUTPUTS from a flip-flop are "1" and "0." A trigger pulse applied to the SET input causes the "1" output to be a positive or negative voltage, depending on the type of transistor. At the same time, the "0" output equals 0 volts. This condition is called the SET STATE.
Figure 3-25.—Flip-flop (SET state).
Figure 3-26.—Flip-flop (CLEAR state).
If a trigger pulse is applied to the CLEAR input, a positive or negative voltage is produced at the "0" output.
The "1" output goes to 0 volts. This condition is called the CLEAR STATE, as shown in figure 3-26.
To determine what state the flip-flop is in, you can measure either the "1" or the "0" output. Measuring 0 volts at the "1" output indicates that the flip-flop is in the CLEAR state. If the "0" output is measured, a positive or negative voltage would also indicate that the flip-flop is in the CLEAR state. Either way, only one reading is necessary.
In figure 3-27, the flip-flop is in the SET state prior to T0 (negative voltage on the "1" output). Now compare the changes in output voltage at each point in time (T0, T1, T2, and T3) with the input pulse. Studying this figure should help you understand how the flip-flop works. The positive pulse at T0 on the CLEAR input shifts the f/f to the CLEAR state (negative voltage at the "0" output). At T1 a positive pulse on the SET input drives the "1" output to the SET state. At T2 a positive pulse on the CLEAR input drives the "0" output to a CLEAR state. At T3 another positive pulse is applied to the CLEAR input. This input has no effect since the f/f is already in the CLEAR state.
Figure 3-27.—Flip-flop with trigger pulse on SET and inputs.
Some flip-flops use a third input lead, as shown in figure 3-28. This third input lead is called a TOGGLE (T) input. Every time a pulse is applied to the T input, the flip-flop will change states from whatever its state was previously. The two diodes (CR1, CR2) form a STEERING NETWORK. This steering network directs a positive input pulse to the saturated transistor, causing it to cut off. Negative pulses are blocked by the diodes. Note that if NPN transistors were used, the diodes would have to be reversed and the TOGGLE signal would have to be negative. For example, assume that Q1 is saturated, Q2 is cut off, and a positive pulse is applied the at T input. The input pulse will be directed to both transistors. The positive pulse will not affect Q2 since it is already in cutoff. Q1 however, which is conducting, will cut off and will cause Q2 to become saturated. The transistors have reversed states. A block diagram which represents a multivibrator and its outputs with only a TOGGLE input signal is shown in figure 3-29. Studying this figure should help you understand how this flip-flop works. Each TOGGLE input causes the output to change states. Figure 3-30 shows what happens when triggers are applied to all three inputs of the flip-flop shown in figure 3-28. Assume that the flip-flop in figure 3-30 is in the CLEAR state ("1" output is 0 volts, "0" output is high) prior to T0. At T0 a trigger is applied to the set input and the flip-flop changes states. Next, the CLEAR input is triggered and the flip-flop returns to the CLEAR state at T1. A TOGGLE at T2 causes the flip-flop to change state, so it is once again SET. Another TOGGLE changes the flip-flop to the CLEAR state at T3 (notice that TOGGLE triggers flip the multivibrator regardless of its previous state). Now, a SET input trigger at T4 sets the flip-flop. The CLEAR input pulse at T5 causes the circuit to CLEAR, and the CLEAR input at T6 has no effect on the flip-flop, for it is already in the CLEAR state.
Figure 3-28.—Flip-flop with three inputs.
Figure 3-29.—Block diagram of a flip-flop with a toggle input.
Figure 3-30.—Flip-flop with three inputs (block diagram).
Remember, a SET input will SET the flip-flop if it is in the CLEAR state, otherwise, it will not do anything; a
trigger at the CLEAR input can only CLEAR the circuit if it is SET; and a trigger applied to the TOGGLE input will
cause the bistable multivibrator to change states regardless of what state it is in.
Q7. In a bistable multivibrator, how many trigger pulses are needed to produce one complete cycle in the output?
Q8. How many stable states are there for a flip-flop?
Q9. If a voltage (positive or negative) is measured on the "1" output of a flip-flop, what state is it in?
Figure 3-31A.—RL circuit.
The secondary load impedance, R2, affects the primary impedance through reflection from secondary to primary.
If the load on the secondary is increased (R2 decreased), the load on the primary is also increased and primary
and secondary currents are increased.
T1 can be shown as an inductor and R1-R2 as a combined or equivalent series resistance (RE) since T1 has an effective inductance and any change in R1 or R2 will change the current. The equivalent circuit is shown in figure 3-31, view (B). It acts as a series RL circuit and will be discussed in those terms.
Figure 3-31B.—RL circuit.
Simple Series RL Circuit
When S1 is closed in the series RL circuit (view (B) of figure 3-31) L acts as an open at the first instant as source voltage appears across it. As current begins to flow, EL decreases and ER and I increase, all at exponential rates. Figure 3-32, view (A), shows these exponential curves. In a time equal to 5 time constants the resistor voltage and current are maximum and EL is zero. This relationship is shown in the following formula:
Figure 3-32A.—Voltage across a coil.
If S1 is closed, as shown in figure 3-31, view (B), the current will follow curve 1 as shown in figure 3-32,
view (A). The time required for the current to reach maximum depends on the size of L and RE. If RE
is small, then the RL circuit has a long time constant. If only a small portion of curve 1 (C to D of view (A)) is
used, then the current increase will have maximum change in a given time period. Further, the smaller the time
increment the more nearly linear is the current rise. A constant current increase through the coil is a key factor
in a blocking oscillator.
Blocking Oscillator Applications
A basic principle of inductance is that if the increase of current through a coil is linear; that is, the rate of current increase is constant with respect to time, then the induced voltage will be constant. This is true in both the primary and secondary of a transformer. Figure 3-32, view (B), shows the voltage waveform across the coil when the current through it increases at a constant rate. Notice that this waveform is similar in shape to the trigger pulse shown earlier in figure 3-1, view (E). By definition, a blocking oscillator is a special type of oscillator which uses inductive regenerative feedback. The output
duration and frequency of such pulses are determined by the characteristics of a transformer and its relationship to the circuit. Figure 3-33 shows a blocking oscillator. This is a simplified form used to illustrate circuit operation.
Figure 3-32B.—Voltage across a coil.
Figure 3-33.—Blocking oscillator.
When power is applied to the circuit, R1 provides forward bias and transistor Q1 conducts. Current flow through Q1 and the primary of T1 induces a voltage in L2. The phasing dots on the transformer indicate a 180-degree phase shift. As the bottom side of L1 is going negative, the bottom side of L2 is going positive. The positive voltage of L2 is coupled to the base of the transistor through C1, and Q1 conducts more. This provides more collector current and more current through L1. This action is regenerative feedback. Very rapidly, sufficient voltage is applied to saturate the base of Q1. Once the base becomes saturated, it loses control over collector current. The circuit now can be compared to a small resistor (Q1) in series with a relatively large inductor (L1), or a series RL circuit.
The operation of the circuit to this point has generated a very steep leading edge for the output pulse. Figure 3-34 shows the idealized collector and base waveforms. Once the base of Q1 (figure 3-33) becomes saturated, the current increase in L1 is determined by the time constant of L1 and the total series resistance. From T0 to T1 in figure 3-34 the current increase (not shown) is approximately linear. The voltage across L1 will be a constant value as long as the current increase through L1 is linear.
Figure 3-34.—Blocking oscillator idealized waveforms.
At time T1, L1 saturates. At this time, there is no further change in magnetic flux and no coupling from L1 to
L2. C1, which has charged during time TO to T1, will now discharge through R1 and cut off Q1. This causes
collector current to stop, and the voltage across L1 returns to 0.
The length of time between T0 and T1 (and T2 to T3 in the next cycle) is the pulse width, which depends mainly on the characteristics of the transformer and the point at which the transformer saturates. A transformer is chosen that will saturate at about 10 percent of the total circuit current. This ensures that the current increase is nearly linear. The transformer controls the pulse width because it controls the slope of collector current increase between points T0 and T1. Since TC = L ÷ R , the greater the L, the longer the TC. The longer the time constant, the slower the rate of current increase. When the rate of current increase is slow, the voltage across L1 is constant for a longer time. This primarily determines the pulse width.
From T1 to T2 (figure 3-34), transistor Q1 is held at cutoff by C1 discharging through R1 (figure
3-33). The transistor is now said to be "blocked." As C1 gradually loses its charge, the voltage on the base of Q1 returns to a forward-bias condition. At T2, the voltage on the base has become sufficiently positive to forward bias Q1, and the cycle is repeated.
The collector waveform may have an INDUCTIVE OVERSHOOT (PARASITIC OSCILLATIONS) at the end of the pulse. When Q1 cuts off, current through L1 ceases, and the magnetic field collapses, inducing a positive voltage at the collector of Q1. These oscillations are not desirable, so some means must be employed to reduce them. The transformer primary may be designed to have a high dc resistance resulting in a low Q; this resistance will decrease the amplitude of the oscillations. However, more damping may be necessary than such a low-Q transformer primary alone can achieve. If so, a DAMPING resistor can be placed in parallel with L1, as shown in figure 3-35.
Figure 3-35.—Circuit damping.
When an external resistance is placed across a tank, the formula for the Q of the tank circuit is Q = R/XL, where R is the equivalent total circuit resistance in parallel with L. You should be able to see from the equation that the Q is directly proportional to the damping resistance (R). In figure 3-35, damping resistor R2 is used to adjust the Q which reduces the amplitude of overshoot parasitic oscillations. As R2 is varied from infinity toward zero, the decreasing resistance will load the transformer to the point that pulse amplitude, pulse width, and prf are affected. If reduced enough, the oscillator will cease to function. By varying R2, varying degrees of damping can be achieved, three of which are shown in figure 3-36, view (A), view (B and view (C).
Figure 3-36A.—Waveform damping. CRITICAL DAMPING.
Figure 3-36B.—Waveform damping. UNDER DAMPING.
Figure 3-36C.—Waveform damping. OVER DAMPING.
CRITICAL DAMPING gives the most rapid transient response without overshoot. This is accomplished by adjusting
R2 to achieve a waveform as shown in figure 3-36, view (A). The resistance of R2 depends upon the Q of the
transformer. View (A) shows that oscillations, including the overshoot, are damping out.
UNDERDAMPING gives rapid transient response with overshoot caused by high or infinite resistance as shown in figure 3-36, view (B). OVERDAMPING is caused by very low resistance and gives a slow transient response. It may reduce the pulse amplitude as shown in figure 3-36, view (C).
The blocking oscillator discussed is a free-running circuit. For a fixed prf, some means of stabilizing the frequency is needed. One method is to apply external synchronization triggers (figure 3-37), view (A) and view (B). Coupling capacitor C2 feeds input synchronization (sync) triggers to the base of Q1.
Figure 3-37A.—Blocking oscillator (synchronized).
Figure 3-37B.—Blocking oscillator (synchronized).
If the trigger frequency is made slightly higher than the free-running frequency, the blocking oscillator will "lock in" at the higher frequency. For instance, assume the free-running frequency of this blocking oscillator is 2 kilohertz, with a PRT of 500 microseconds. If sync pulses with a PRT of 400 microseconds, or 2.5 kilohertz, are applied to the base, the blocking oscillator will "lock in" and run at 2.5 kilohertz. If the sync PRF is too high, however, frequency division will occur. This means that if the sync