Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
09/19/04 04:17
Read: times


 
#77662 - RE: WEOT: Charge Pumps/DC-DC converters
Responding to: ???'s previous message
Hallo Prahlad,

well, today it's very easy to get a DC/DC converter to do the job. But I remember those times, when these smart black boxes weren't available. At those days the 'charge pumpe' methode was widely used to fabricate either higher supply voltages than being present on the card, or even to fabricate supply voltages of other polarity.

In the following I want to demonstrate how these charge pumps worked. Please make no mistake, it's not my intention to recommend the use of these circuits, because modern DC/DC converters show much better performance! It's my intention to show how our fathers produced bipolar supply voltages by the help of low-cost and everywhere available parts.

A chip that was designed to support these charge pump methode is ICL7660. But even with so simple chips like TDA2003 or NE555 charge pump methode can be implemented. Have a look at the following circuit that produces about +8.6V and -7.2V from +5V supply voltage:




Have a look at the upper NE555: Chip works as astable multivibrator providing a square wave at its output of about 3kHz. When output is low, electrolytic capacitor next to it becomes charged to 5V - 0.7V = 4.3V. When output goes high, then positive electrode of this capacitor is 4.3V higher then supply voltage and shows a potential of 4.3V + 5V = 9.3V. As consequence, second electrolytic capacitor is charged up by this voltage and positive electrode of this capacitor shows a potential of 9.3V - 0.7V = 8.6V.

Performance of second NE555, which provides a negative voltage of about -7.2V is very similar. But here polarities of diodes are choosen in such a way, that voltage swings at involved electrolytic capacitors can be used to fabricate a negative supply voltage.

Of course, the actual output voltage highly depends on actual load and this can be a disadvantage. But even fluctuations of several 100mV will not present a problem in most cases. Assume for instance, that supply voltage changes by 1V. If now the operational amplifier which is supplied by this voltage shows a power supply rejection ratio (PSRR) of 80dB, then offset voltage change will be less than 1V / 10000 = 100 µV. In combination with an ADC/DAC signal voltage range of 2.5V, the according error would represent less than 1LSB of a 14bit ADC/DAC. So, a 12bit ADC/DAC would not suffer from this error.

Another disadvantage of this charge pump methode is, that a certain amount of ripple will be present superimposed to supply voltages. But a simple RC low pass filter of 100Ohm + 100µF in the supply lines of used operational amplifier would sufficiently suppress this ripple. Corner frequency of this RC filter must be much lower than switching frequency of charge pump, here 3kHz. The above filter yields a corner frequency of fc = 1 / (2 x pi x R x C) = 16Hz.

So, with a bit care these old fashioned charge pumps can even be used in high precision applications.

Where is now the advantage of modern DC/DC converters?
First, most of them will have some regulation of output voltages. Mostly, only a simple LC filter per output line is needed to suppress high frequency ripple down to an acceptable value. But the biggest advantage is, that they use such a high switching frequency, that simple and high reliable ceramic capacitors can be used! Especially if ambient temperatures are increased, a switcher containing only ceramic capacitors, like multilayer X7Rs will show a tremendous advantage in terms of service life and reliability compared to a design that uses electrolytic capacitors...

Kai

List of 23 messages in thread
TopicAuthorDate
WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
   RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
      RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
   RE: Module pricing            01/01/70 00:00      
   RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
   RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
   RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
      RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
         RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
            RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
               RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
                  RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
         RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
            RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
               RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
               RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
                  Mail Order prices- Prahlad/George            01/01/70 00:00      
                     RE: Mail Order prices- Prahlad/George            01/01/70 00:00      
                        Exotic is fun - Kai            01/01/70 00:00      
                           RE: Exotic is fun - Kai            01/01/70 00:00      
                  RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
                     RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      
   RE: WEOT: Charge Pumps/DC-DC converters            01/01/70 00:00      

Back to Subject List