I'm trying to simulate a zero-crossing detector using Multsim Live ideal comparator, however it appears there is built-in hysteresis. I want output to change exactly when 0v is crossed. It should not matter if I use inverting or noninverting input as the 0v reference. I just want to demo ideal behavior. But it does not work. There's hundreds of mV hysteresis, not mentioned in spec's of this ideal device. How can this be fixed, or am I doing something wrong?
Solved! Go to Solution.
Here is the simple circuit, where I offset by 2.5 volts to get away from ground, but no matter what I tried, the output does not flip when inverting input = noninverting input. So what's happening?
it looks like if you let Multisim to choose the transient timestep automatically, for a rectangular wave it chooses just a few points which gives the strange hysteresis error.
As a workaround, try setting the maximum timestep manually to some sensible value, e.g. Period/1000. That will make the hysteresis go away or at least it will be much less noticable. I tried a transient simulation for 1ms with 1e-5 and 1e-6 max timesteps. At 1e-5 the hysteresis is still visible but at 1e-6 looks gone.