Forums » CircuitLab Support Forum

Hello,

the output of the voltahe comperator goes to 10V when I connect the feedback resistor R7.

When disconnecting the resistor, the comperator switches like it should. But I want to see the Hysteresis of the circuit.

Regards, Holger

Correction: "the Output Voltage goes to 100V..."

Changing the Type of the OP-Amp or the value of the feedback resistor (R7) makes no difference.

Hi Holger,

I think your timesteps are too long.

If you change your sinewave to 2kHz then set your sim time to 0.5ms and your time steps to 1us, it runs OK.

Note that:

i) with the resistors you've chosen, you will see only a very small range of hysteresis;

ii) A real LM741 would not work in your circuit: it requires a higher supply voltage and will not operate with inputs so close to the -ve supply rail (in your case -ve rail = ground).

BTW: in the Time Domain sim setup, if you tick "Sweep Parameter", enter "R7.R" as the Parameter then select Sweep Type = Custom and enter "10k, 1G" as the values, you get two runs, one with R7 = 10k, one with R7 = 1G (i.e. open circuit).

See also:

Hello,

great, thanks for the fast feedback. With this workaround it works fine. Anyway, it seems to be a bug.

Regards, Holger

Hi again Holger,

It is less a bug and more a feature of the way CL does its sums.

CL does not seem to dynamically adjust it's time steps the way SPICE and other simulators do (you can see this in CL from the way the trace data points in any given plot are separated by a constant time interval, which is the time step you have chosen for that sim).

Therefore if the time steps are too large, the solutions to the circuit equations may quite easily show oscillatory behaviour and/or large spikes as alternate guesses under and overshoot.

A little more info from CL:

https://www.circuitlab.com/docs/faq/#q_unrealistic_spikes

and under Time-Domain Simulation in:

https://www.circuitlab.com/docs/the-basics/