What is wrong with this charge pump circuit ?

Discussion in 'General Electronics Chat' started by IcedFruits, Nov 17, 2014.

  1. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    hi, i was trying to simulate charge pump with 555 in ltspice, the first one with direct 555 o/p to pump capacitor works just fine as it should, but the 2nd one with added npn transistor (for bigger current handling purposes) fails to do the job.

    checked, while not connected to pump capacitor C1, clocks are at same phase at both 555 o/p & Q1 emitter. but whenever C1 is connected, emitter clock disappears & steady ~4.5v appears, also charge pump fails, with only ~5v around C4.

    i have included both ltspice files (working & not working) in the attached zip file. please run the simulation if needed and kindly help. really confused.....

    [​IMG]
     
    Last edited: Nov 17, 2014
  2. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    For starters, R5 is too big by about three orders of magnitude.
     
  3. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    checked with 20k or 1k values @ R5, they did not work.
     
  4. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    I don't see how the output can sink enough current to charge C1; I think you should replace Q1 with a push-pull driver stage.

    Take a look at the schematic for the 555 and you'll see why driving the charge pump directly from the 555 output works.
     
    Last edited: Nov 17, 2014
    IcedFruits likes this.
  5. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    amazing, sir ! u're a genious ! that worked ! i think i can tune the rest now, still suggestions are very much appreciated. :)

    [​IMG]
     
  6. cmartinez

    AAC Fanatic!

    Jan 17, 2007
    3,555
    2,517
    After looking at your post, I think that you might find this thread that I started a few weeks ago very interesting.
    Go directly to this post if you want to download the latest version of the circuit.
     
    IcedFruits likes this.
  7. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    Splendid:)
     
  8. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    thanks for the link, will look ! my goal was just to driver a mosfet in a 5 volt circuit though !
     
  9. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    Be advised that a real 555 will be marginal at a Vcc=5V. The LTSpice behavioral model is too simplistic to model this correctly. Get the 555 model that is on the LTSpice Yahoo group if you want a more realistic sim of a 555 powered from only 5V.
     
    IcedFruits likes this.
  10. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    thanks ! i will check the yahoo group model. also i think i have some CMOS TS555 chips, those work from as low as 2Volts, so i guess it wont be much of a problem. but thanks for advice ! :)
     
  11. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    The alternative linCMOS versions are claimed to work down to a 2V supply.
     
  12. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    You could also look at redesigning the output stage so that it can get closer to the supply rails and swing more voltage.
     
  13. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    u mean this ? that's the closest C1 gets to supply rails, and R4 remains to control current of C1.

    [​IMG]
     
  14. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    At the moment, you are losing a quarter the the supply voltage across the emitters of Q1 and Q2. If you swap Q1 and Q2 around (configure as common emitter rather than emitter follower), the output will be able to swing closer to the supply rails but you have to be careful because during switching transitions both transistors will be on at the same time which will cause short current spikes on the supply lines. You must also make sure that the base drive to the transistors is able to get close enough to the supply lines to ensure that each transistor can be turned fully off.

    The current spikes can be dealt with by adding a large decoupling capacitor across the supply lines to the transistors and by increasing the values of R5 and R7 a little. A common emitter stage after the output of the 555 should provide suitable base drive for Q1 and Q2.
     
  15. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    but with common emitter @ rails, probably due to opposite on effects to 555 clock, the voltage swing gets worse...... @ the meeting point of both collectors.

    also, in this configuration, both transistors don't get on at the same time either. can u please explain with schematic ?

    [​IMG]
     
    Last edited: Nov 19, 2014
  16. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,516
    1,246
    Nowhere do you say what your output current requirement is. A bipolar 555 can source/sink over 100 mA. If that's all you need, you can drive C1 directly without the two transistors for much better performance.

    ak
     
  17. crutschow

    Expert

    Mar 14, 2008
    12,990
    3,226
    You need a separate 1k resistor from the output of the 555 to each transistor base. Otherwise both transistors will be on at the same time.
    Also you don't need R5, R6, and R7.
    And why is the resistance of V1 so high?
     
    Last edited: Nov 19, 2014
  18. IcedFruits

    Thread Starter Member

    Jan 15, 2014
    84
    2
    originally only ~25-30mA. that requirement can already be met by 555 itself, just doing to the discussion for knowledge increasing purpose, to drive bigger transistors if want in future. optimization of the transistor configuration is the objective now.
    small signal BC r just for illustration purpose. pls ignore their numbers.

    they are pnp & npn..... anyways i will check with separate base transistors for any improvements. V1 is just some old rechargeable AA cells i have they don't give much high currents.
     
  19. crutschow

    Expert

    Mar 14, 2008
    12,990
    3,226
    That's obvious from the schematic. Did I post something to make you think that I didn't understand that?
     
  20. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    As I said earlier, you need a common emitter stage between the output of the 555 and Q1 &Q2. It is essential that Q1 and Q2 are driven from an output that can swing close enough to the supply rails to ensure that each transistor can switch fully off while the other is fully on. I don't think that the 555 output can do this.
     
Loading...