
3
CHARGE PUMP DC-TO-DC CONVERTER
TC7662A
TC7662A-5 9/11/96
2001 Microchip Technology Inc.
DS21468
Note one of its characteristics is ESR (equivalent series
resistance). This parasitic resistance winds up in series with
the load. Thus, both voltage and power conversion effi-
ciency are compromised if a low ESR capacitor is not used.
For example, in the "Test Circuit", changing CP and CR
capacitors from typical ESR to low ESR types, the effective
converter output impedance changed from 45
to 40, an
improvement of 12%.
This applies to all types of capacitors, including film
types (polyester, polycarbonate etc.).
Some applications information suggests that the ca-
pacitor is not critical and attributes the limiting factor to the
capacitor's reactance value. Let's examine this:
where DS (duty cycle) = 50%.
Thus, ZC ≈ 1.33 at f = 12kHz, where C = 10F.
For the TC7662A, f = 12,000Hz, and a typical value of
C would be 10
F. This is a reactive impedance of ≈1.33.
If the ESR is as great as 5
, the reactive value is not as
critical as it would first appear, since the ESR would dominate.
The 5
value is typical of a general-purpose electrolytic
capacitor.
Synchronizing
The TC7662A may be synchronized by connecting pin
7 of the TC7662A through a 100k resistor in series with a
diode to a negative-going pulse source. The negative pulse
voltage can be +5V with a 5 microsecond duration going
negative to 0V.
TEST CIRCUIT
Figure 2.
Synchronization
XC =
and ZC =
,
XC
DS
APPLICATIONS INFORMATION
Theory of Operation
The TC7662A is a capacitive charge pump (some-
times called a switched-capacitor circuit), where four
MOSFET switches control the charge and discharge of a
capacitor.
The functional diagram (page 1) shows how the switch-
ing action works. SW1 and SW2 are turned on simulta-
neously, charging C1 to the supply voltage, VDD. This
assumes that the ON resistance of the MOSFETs in series
with the capacitor produce a charging time (3 time con-
stants) less than the ON time provided by the oscillator
frequency, as shown:
3 (RDS(ON) C1) <C1/(0.5 fOSC).
In the next cycle, SW1 and SW2 are turned OFF and,
after a very short interval with all switches OFF (preventing
large currents from occurring due to cross conduction),
SW3 and SW4 are turned ON. The charge in C1 is then
transferred to COUT, BUT WITH THE POLARITY IN-
VERTED. In this way, a negative voltage is derived.
An oscillator supplies pulses to a flip-flop that is fed to a
set of level shifters. These level shifters then drive each set
of switches at one-half the oscillator frequency.
The oscillator has a pin that controls the frequency of
oscillation. Pin 7 can have a capacitor added that is con-
nected to ground. This will lower the frequency of the
oscillator by adding capacitance to the internal timing ca-
pacitor of the TC7662A. (See Oscillator Frequency vs. CEXT,
page 5.)
Capacitors
In early charge pump converters, capacitors were not
considered critical due to the high RDS(ON) of the MOSFET
switches. In order to understand this, let’s look at a model of
a typical electrolytic capacitor (Figure 1).
1
2
πf C
TC7662A
1
2
3
4
8
7
5
CP
+
10
F
COSC
R L
V
(–5V)
OUT
10
F
C R
I L
IS
V
(+5V)
+
NC
6
+
TTL
Q
100 k
TO PIN 7
TC7662A
Figure 1.
Capacitor Equivalent Circuit
EPR
ESL
ESR
C