1
\$\begingroup\$

I need to build a 2 stage inverting charge pump in order to convert a 12v rail into -24V which is then regulated to -12V using a 7912 LDO. it is used for a audio amplifier design and I need about 100mA of output current.

The oscillator is built from a CD40106B using a 100pF cap and a 50K resistor to get the highest frequency as recommended by datasheet. only one pin creates the actual oscillator and the other pins are inverted to create the two clock signals (one is the inversion of the other) and since there are 6 gates available, I have 3 complementary clock signals and decided to create 3 pumps (each with 2 stage) and put them in parallel in order to maximize the output current. also one of the stages is 180 degrees out of phase with the others in order to reduce ripple/switching noise.

I used push pull transistors in order to boost the output current as well (the CD40106B is speced to supply a max of about 2 or 3mA!). 2N2222A and BC327-40 are the parts I chose.

I also initially did a prototype of the circuit on the breadboard and it seemed to work perfectly fine and even with a 10k load resistor, it held about -23V.

Then I put everything on a veroboard and soldered it up but it doesn't work. Even the oscillator outputs seem to be weird, before I connect them to the transistor bases, the output voltage on DC setting of multimeter is 5.6 or there abouts for the original signal and 6.something volts for the inverted signal which is expected, but after connecting that to the transistor base, it's only about 1 or 1.4v! after disconnecting the base again, the output stays at about 1.4 so I'm assuming I destroyed the CD40106B as well? I didn't use base resistors so I'm assuming that's the issue? but the transistors are rated for about 100mA base current and the CD40106B isn't supposed to supply more than a few mAs?

P.S: diodes are schottky (1N5819) and the caps are 470uF 50V electrolytic.

here is the full schematic, please help me fix this and get this charge pump to work(please ignore the 100mA current load on the output capacitor). enter image description here

and here is a simulation of the oscillating waveform created by the CD40106B: enter image description here

\$\endgroup\$
4
  • \$\begingroup\$ Are you missing gate resistors? \$\endgroup\$
    – MadHatter
    Jul 21 '19 at 18:07
  • \$\begingroup\$ these are BJTs, so there is no gate! but yes, I omitted base resistors since the CD40106B doesn't seem to be able to damage them, at least on paper! \$\endgroup\$
    – OM222O
    Jul 21 '19 at 18:45
  • \$\begingroup\$ I also noticed that the output of the CD40106B is loaded down after each stage. Q1 is goes to only about 10V, /Q goes to about 7V, so on and so forth. 1K base resistors improves it, but it still looks a bit rounded on the edges. 10K seems to fix it all together. I have to try and build a completely new prototype but I'm not quire sure why this happens. why do the transistors without high value base resistors mess up the signal output of the IC? \$\endgroup\$
    – OM222O
    Jul 21 '19 at 20:41
  • \$\begingroup\$ Because a BJT transistor has no resistance through the base to emitter. Without a base resistor to limit the current you will basically be shorting through the emitter... \$\endgroup\$
    – MadHatter
    Jul 21 '19 at 21:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.