Hi , I want to design a SMPS circuit. I want to use smth like Viper22 offline switch for control circuit and more. The problem is , there is no spice model for viper22 or uc3844. How do you guys simulate this kind of circuits ?

 

uc3844

The UC384x family has spice models. Many different companies have made it.
I think the viper22 does not have one but should not be hard to make.
---edited---
I use LTspie. It has a LT1244 which is almost the same as the UC3844. The LT124x family will work to replace the UC384x.

 

In simulation, for component selection and characterization e.g. inductors, transformers, diodes, and power switches I just use generic parts like signal generators, e.g. square, triangle, and sawtooth along with generic comparators and opamps to get the forward path and feedback loops behaving the way I need. Then I look at complex chips to finalize the design.

The goal should be understanding rather than design verification.

 

The UC384x family has spice models. Many different companies have made it.
I think the viper22 does not have one but should not be hard to make.
---edited---
I use LTspie. It has a LT1244 which is almost the same as the UC3844. The LT124x family will work to replace the US384x.

u are right thank you, but how it is not hard to make could you detail please

 

In simulation, for component selection and characterization e.g. inductors, transformers, diodes, and power switches I just use generic parts like signal generators, e.g. square, triangle, and sawtooth along with generic comparators and opamps to get the forward path and feedback loops behaving the way I need. Then I look at complex chips to finalize the design.

The goal should be understanding rather than design verification.

Yes but simulating previously can prevent huge cost sometimes you know..

 

how it is not hard to make could you detail please

A very simple Viper is here. The startup parts are not here, and the supply is not here. I would add a 12V battery to power the parts.
60khz starts the cycle by turning on the MOSFET. When the current monitoring resistor gets 0.23V across it the Latch is turned off.
When the PWM is up and running current flows in the LED, causing current to enter "FB" pin. This will reduce the current needed to turn off the MOSFET. The current into "FB" can cause the MOSFET to go to 0%, at about 1.23V on the FB pin.

 

Yes but simulating previously can prevent huge cost sometimes you know..

I do know; no surprise there.