1
\$\begingroup\$

I face the "time step too small" error while simulating this circuit. I don't think it should be super hard to solve.

The schematic is for an isolated AC dimmer and consists of some optocouplers, a 555 timer, FETs, and two grounds - a normal ground and LV_GND. If I bridge these grounds together things will solve, but that's not what I'm going to build.

I've tried the gear solver etc., but they don't seem to work.

.asc file: https://file.io/zAzXYz1sMGye

enter image description here

\$\endgroup\$
2
  • 4
    \$\begingroup\$ You have a floating section of circuitry with respect to GND which SPICE doesn't like. Connect a 10meg resistor between LV_GND and GND and see if that helps. \$\endgroup\$
    – qrk
    Dec 7, 2022 at 23:58
  • \$\begingroup\$ Why did you delete the file? Why not copy-paste it here, instead? Use ``` as delimiters (before and after the pasted text). Also, What are you trying to accomplish? The two transistors short out the source to ground if they end up conducting. I doubt they will, not with those resistors for the optos. Plus, I really feel for U1 being driven the way it is. What value does V1 have? \$\endgroup\$ Dec 9, 2022 at 9:45

2 Answers 2

2
\$\begingroup\$

You are facing a problem with sharp transitions from poorly modeled components or inductors. In real life, there is also a capacitance associated with each node, which helps to make the transitions smoother.

Add an option to your LTSpice simulation in the following way: enter image description here

It will add a shunt capacitance to each node of 1fF. I tried it, and it works perfectly. For your application, a capacitance of up to 100fF will have neglected effects.

\$\endgroup\$
1
\$\begingroup\$

Too much of the circuit is floating and needs to be better referenced. V1 should be referenced to GND.

Also, Try putting a large resistor (like a 1Meg) between LV_GND and GND. Check the voltage, if it's more than a few 10mV then you might want to combine grounds for simulation purposes.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks for the reply. The circuit creates a virtual ground at the anode of D2, if I add a ground at the neut of V1 that isn't really an accurate representation of the circuit and I immagine there will be knock on effects. I tried doing this and it doesn't work anways. \$\endgroup\$ Dec 9, 2022 at 0:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.