LM2596 DC-DC regulator

Discussion in 'General Electronics Chat' started by jrdoner, Jul 23, 2014.

  1. jrdoner

    Thread Starter New Member

    Feb 21, 2012
    14
    0
    I bought some DC-DC stepdown voltage regulators (LM2596), for little more than $1.00 apiece, and they seem to provide remarkably stable load regulation. My question is as follows. I know that you can't hook two non-switching step-down regulators (like the 7805) in parallel, but is there a good reason that I couldn't do that using the LM2596, if I adjust the two simultaneously for the same output. It would seem that any little imbalance between them would be compensated for by their switching rates.

    Thanks in advance for any insights.

    John Doner
     
  2. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    It is precarious to put two constant voltage supplies in parallel no matter the type. Although putting a small amount of resistance in series with each supply's output can make them behave. Select the resistance to drop .5 volts at max current output of each supply. There is no science in this, just a reasonable starting point.

    Mark
     
  3. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    If there is 0.1v difference a large current would flow.

    Maybe you can parallel them but you need to parallel their feedback inputs. I have never tried that.
     
  4. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,538
    1,251
    There are two ways to current-share two voltage regulator circuits, whether linear or switching. Both require access to the feedback loop. The first is droop share as Les described. The second is active 3rd-wirecurrent sharing - you put a current sense resistor in series with each output, condition the current signals, sum them into a shring bus, and use the difference between each individual current and the share signal to drive the voltage feedback loops. Sounds messy, is messy. Unitrode has chips for this.

    Back to droop share, it is common to OR the two voltage outputs into the common load through power diodes. The diodes provide the ballast resistance, minimize control loop cross interference. Also, since a silicon diode's Vf increases with current, the two outputs do not have to be perfectly trimmd.

    ak
     
  5. Sensacell

    Well-Known Member

    Jun 19, 2012
    1,130
    267
    There is no danger of unintended current flow- power supplies only source current, they do not sink current.

    The problem is unequal sharing of the output load current, the one with the higher setpoint will do most of the work.

    This is the reason for the resistors, helps to balance the load, but it's still not a good solution- they contribute to poor output regulation.
     
Loading...