Datasheet

TPS5103
MULTIPLE MODE SYNCHRONOUS DC/DC CONTROLLER
SLVS240A SEPTEMBER 1999 REVISED MAY 2001
26
POST OFFICE BOX 655303 DALLAS, TEXAS 75265
APPLICATION INFORMATION
loop-gain compensation (continued)
There is one pole, one zero, and one integrator.
Zero
1
2 C3R4
Pole
1
2 C2R4
Integrator
1
2 ƒC3R2
The loop-gain concept is used to design a stable and fast feedback control. The loop-gain equation is derived
by the control-to-output transfer function times the compensation. The equation is shown below.
Loop gain Vod X Comp
By using a bode plot, the amplitude and the phase of this equation can be drawn with software such as MathCad.
In turn, the stability can be easily designed by adjusting the compensation perimeters. The sample bode plot
shown in Figure 45 explains the phase margin, gain margin, and the crossover frequency.
The gain is drawn as 20 log (loop-gain), and the phase is in degrees. To explain them clearer, 180 degrees is
added to the phase, so that the gain and phase share the same zero.
Where the gain curve touches the zero is the crossover frequency. The higher this frequency is, the faster the
transient response is, since the transient recovery time is 1/(crossover frequency). The phase to the zero is the
phase margin at the crossover frequency. The phase margin should be at least 60 degrees to cover all the
condition changes, such as temperature. The gain margin is the gap between the gain curve and the zero when
the phase curve touches the zero. This margin should be at least 20 dB to assure the stability over all conditions.
Phase
Margin
Phase
Gain
Crossover
Gain
Margin
54
2
44
100
96
166
f Frequency Hz
180
152
138
124
110
82
68
40
26
12
16
30
58
72
86
10 100 1 k 10 k 100 k 1 M
20 Log (Loop Gain)
180 + Phase
Figure 45. Sample Bode Plot (not the EVM)
synchronization
Some applications require switching-clock synchronization. The following two methods are used for
synchronization.
Triangle-wave synchronization