time step too small in ltspice, problem with diode.

Discussion in 'The Projects Forum' started by AbhimanyuSingh, Jun 22, 2011.

  1. AbhimanyuSingh

    Thread Starter New Member

    May 22, 2011
    Circuit 1,2 and 3 are very similar, except that circuit 3 has 0 dc value of the source while circuit 1 and 2 have positive and negative DC values respectively. Circuits 1 and 2 get simulated as expected and give good result. But as soon as I start circuit 3, I get a message "time step too small......trouble with mbr20100ct-instance d1"
    So, I grounded both the nodes across d1 with very high resistances. The simulation started running, but the speed was insanely less in the order of few femtoseconds, which was useless. The snapshots could be confusing, so I am also attaching the circuit files.
  2. Ron H

    AAC Fanatic!

    Apr 14, 2005
    I don't know if it can cause a "time step too small" error, but you used the generic NMOS in circuit 3, which is very small geometry relative to IRF530. I don't know if it does what you want, but the sim will run if you change the transistors to IRF530s.
  3. AbhimanyuSingh

    Thread Starter New Member

    May 22, 2011
    AAhhh..my bad...
    Thanks Ron. Thanks for pointing that out.
  4. SgtWookie


    Jul 17, 2007
    Did you realize that in the 3rd schematic, M2 is installed upside-down?
    Power MOSFETs have an intrinsic body diode; if the polarity across the drain and source is inverted, the body diode will conduct.