Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

1.4 Vdc to 300 Vcap

  1. Jul 19, 2004 #1


    User Avatar
    Homework Helper

    New to power EE.

    I'm trying to charge up a cap (120 μF) to a high voltage (300 V). I have a tiny transformer that seems to step up the voltage several hundred times (in the 10's of kHz frequency range at least). I am using two NPN bjts (2N4123) to provide a crude clock signal to the base of a PNP (MPSA56) in order to periodically pull down and then block the current through the primary of the transformer (at 14 kHz). When I probe the secondary, I see 100's of volts across the probe impedance. However, when I put a 100 kΩ across the secondary, the voltage drops dramatically. My suspicion is that the transformer is operating very inefficiently. Furthermore, when I do connect a diode and cap in series across the secondary, the voltage charges up on the cap at a very slow rate (a few sec per V). Any suggestions? Are the transistors critical? Do I have to use a MOSFET to operate the primary? Am I running it at too high frequency? Is the duty cycle critical?

    I decided to use the PNP because I figured the base could be pulled down more robustly than up. I am figuring that the current should be about 100 mA through the primary at about 1 V, so my goal is to see something like 100 μA at the secondary.
    Last edited: Jul 19, 2004
  2. jcsd
  3. Jul 20, 2004 #2
    What you are trying to do, if I understand correctly, is use a 1V source and a transformer to create a 300V potential across a capacitor.

    If you look up the schematic for a boost converter, you may find this to be a viable solution.

    If you use a 1:100 transformer to step up the voltage to 100V, then regulate it to 300V, you end up with your desired 300V across your output capacitor. The duty cycle of the converter will be nominally about 67%.

    However, if this is a one-time charge, it may not be worth the effort to design a converter.

    As for your current mehtod:

    I think the way you are switching the primary is OK. The reason your voltage sag is occurring may be due to conservation of power, but not necessarily the transofrmer efficiency. At 300V, a 100kohm resistor will need 3mA on the secondary, and if you have a 1:300 transformer, that translates to 900mA on the primary - is your source folding back?

    Gotta run...........
  4. Jul 20, 2004 #3


    User Avatar
    Homework Helper

    Thanks for the response! I was beginning to wonder if this question was just too utterly foolish. To tell you the truth, I was a bit embarassed to even ask it, as I was an EE by trade once upon a time.

    It is not so much an issue of effort on this project, it is the whole point of this project to find the simplest solution (in terms of components). I have access to decent (though not state-of-the-art) equipment, so tuning and tweaking the circuit is essentially trivial (as long as the analysis remains in the 100's of V and kHz range). My basic goals are:
    - robust design (so that I could feasibly reproduce the circuit without any need to tune or tweak)
    - absolutely minimal components, with the limitation of nothing more sophisticated than the transistors (however, with my 1.4 V source, transistors are probably the best option anyway).

    My interest in this little side project was spawned by the discovery of the use of such a circuit to produce the required burst at 300 V in a flash camera with a simple AA battery as a source. I was immediately fascinated. So far, my biggest hurdle in trying to reproduce this from scratch is the excessively slow charge time of the cap. In the camera, this only takes a few seconds. At the rate my circuit charges the cap, I estimate that it would take a good half hour - completely unacceptable.

    The circuit in the camera has some wiring that doesn't make sense to me.

    Do you happen to know of any transistors with the designations:
    They have three pins and are shaped like that characteristic half-cylinder of a transistor.

    These are connected together in (what seems to me to be) a strange way along with a 3904 (I know what that one is, of course). Both sides of the transformer seem to be controlled directly by this strange configuration. Does that sound familiar?

    I understand this issue, but I do not believe it to be the case. The voltage isn't just dropping down a 10's of dB, it is dropping down by at least 100 dB (the signal essentially vanishes on the scope, but if I jiggle the probe, I can see the ghost of a ripple past).

    I have no idea what this means (so something tells me that it is probably my problem).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook