ripple on constant current load

Discussion in 'General Electronics Chat' started by bug13, May 29, 2013.

  1. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    Hi

    I am working on a constant current load, here are some waveform of my breadboard circuit, I can't seem to get the ripple further down to 1-2 mV.

    The ripple of my circuit is 10mV, and that translate to 50mA ripple, which I think is too much.

    So is that acceptable, or I have a lot more work to do to improve me circuit.

    Thanks

    Datasheet:
    Op amp - OP07C
    MOSFET - IRF540N

    This is the ripple with average on - 8 average
    [​IMG]

    This is the ripple without average enable.
    [​IMG]

    here is my circuit: (my LTSpice simulation is also attached)
    [​IMG]
     
    Last edited: May 29, 2013
  2. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Have you looked closely at the power supplies for ripple?

    I remember working on a HP power supply in a metrology lab, it had ripple I thought was unacceptable. Then I noticed the ripple corresponded with a AM radio someone was playing in the background. Them being tuned to the same station was coincidence, but it turned out I was looking at RF interference.
     
  3. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    Any error/ripple at the reference input (opamp positive input) will be amplified in your system. Can you post a schematic of your current real world circuit. As many details are missing in your simulation setup :)
     
  4. Gary_P

    New Member

    Apr 4, 2013
    7
    0
    When a constant current load is performing correctly all ripple from the power supply will be visible across the load resistor (assuming that you are using a resistor for load).

    To measure the current you need to do either an inline current measurement of the CCS load or a differential voltage measurement of the voltage across the load resistor. You can also measure the voltage across the sense resistor as it is ground referenced.

    I have a power supply load tester that consists of an adjustable CCS DC load with AC load capabilities on my webpage. Look in the Swenson regulator section.

    www.pimmlabs.com
     
  5. crutschow

    Expert

    Mar 14, 2008
    13,042
    3,243
    How you deriving the constant current reference (V3)? That needs to be derived from a stable voltage reference source.
     
  6. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    Yes, it looks like the majority of the noise is from my SMPS, as long as I replace the 14V rail with a 9V cell battery, the noise reduce down to ~1mV

    my 14V is the Vload and also the VCC of the op amp, and the source of to the LM7805, which serve as the Vref for the current control.
     
  7. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    here is my real word circuit, the L is just to simulate the wire to the MOSFET

    [​IMG]
     
    Last edited: May 29, 2013
  8. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    Arrh, that sounds like a good idea, why didn't I think of that?

    I have looked at your circuit, but I think I need to learn more before I can understand of your circuit. (or look at your circuit more closely) :)
     
  9. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    It's from a LM7805 -> 10K pot, so I am guess LM7805 is not good enough?
     
  10. bug13

    Thread Starter Well-Known Member

    Feb 13, 2012
    1,208
    38
    Now I find something weird, I got some noise as soon as I connect my probe to the circuit, (probe at Rsense), the ripple is about 1mV to 2mV, always at the middle of the time scale.

    This show up in my scope even my circuit is not power up, where is that from? something wrong with my scope? but when I connect the probe direct to its ground, I don't see that.

    the probe in use is a x1 probe.

    it looks like this
    [​IMG]
     
    Last edited: May 29, 2013
Loading...