I'm designing a circuit with AD633 analog multiplier in proteus. The inputs in AD633 are a square pulse pattern and a sinusoidal wave. When I run the simulation, fatal simulation error dialog box occurs and shows two warnings including 'Time step too small, trouble with node..... '(as the following image). Did anyone face similar problem while working with AD633? If did, what is the solution for this problem?

 

Yes, I have sometimes found the AD633 model somewhat problematic in LTspice also.

Try eliminating the initial bias calculation (add .uic) in the transient simulation to see if that helps.

 

Yes, I have sometimes found the AD633 model somewhat problematic in LTspice also.

Try eliminating the initial bias calculation (add .uic) in the transient simulation to see if that helps.

Actually I'm doing the simulation on Proteus 8 not LTspice. I think your suggestion if for LTspice. Isn't it?
Can you or anyone suggest what to do to make AD633 work fine without any fatal simulation error on proteus?

 

Can you screenshot the circuit and I can take a look..

 

Okay... Carl is correct.. To simulate you need to change the model to "Simple"

 

Can you screenshot the circuit and I can take a look..

It's a huge circuit. It might be difficult for you to look at it. But I'm showing you anyway (I've divided the total circuit into 3 screenshots). I'm trying to do a multiplexing project. There are two users from whom I'm taking digital pattern data and modulating them with a carrier with the help of the multiplier AD633. I don't think the problem is in the circuit. Because when I make a small test circuit with AD633 (with a pulse and a sinusoidal input), it still doesn't work and shows similar fatal simulation error.

 

Okay... Carl is correct.. To simulate you need to change the model to "Simple"

Sorry I didn't get what you said. What did you mean by "simple"? please explain

 

Actually I'm doing the simulation on Proteus 8 not LTspice.

But isn't that a Spice simulator?
It should have something equivalent to the .uic command.

 

Edit properties... Here is a screenshot... The normal spice model has been modified as Carl suggested... Clearly a fault down the line.

 

The normal spice model has been modified as Carl suggested.

I was referring to eliminating the initial conditions part of the transient simulation, which has nothing to do with the model.

 

I was referring to eliminating the initial conditions part of the transient simulation, which has nothing to do with the model.

Probably, but someone has modified the file and it works... I know didley squat about spice models... I leave that to Alec...

 

Edit properties... Here is a screenshot... The normal spice model has been modified as Carl suggested... Clearly a fault down the line.

Okay got it. Finally it's working now. Thank you.