Error"Failed to resolve Convergence Issue"

[GoviRe]

Hello,

I am working with the Multisim13 and I when I am trying to implement the design as attached shows a window saying ""Failed to resolve Convergence Issue" as shown in attachment.

I found similar threads here 

http://forums.ni.com/t5/Circuit-Design-Suite-Multisim/how-to-solve-simulation-error/m-p/838043/highl...

http://forums.ni.com/t5/Circuit-Design-Suite-Multisim/Problem-with-simulation/td-p/2603281/highlight...

Btu were not upto the solution to this.

Thanks.

[johnsold · Accepted solution]

According to the OPA381 data sheet it is designed to work with capacitive loads up to 100 pF, in inverting amplifier configurations, and as a transimpedance amplifier which typically is used with a (very small) current input and fairly large feedback resistors.  Your circuit operates with capacitive loads of ~ 5 uF (50000x the design load), unity gain, non-inverting configuration, and with a voltage input.  Depending on how the model for the OPA381 handles these out-of-normal range values, convergence errors may not be too surprising.

The coupling network between the stages has a low pass cut off of ~1.3 Hz. The first stage input network is more complicated but also limits bandwidth to very low frequencies.

What is this circuit supposed to do?

I simulated the coupling and feedback networks (assuming ideal unity gain amplifiers). This is the output with a 1 volt input.

Lynn

[GoviRe]

Hello Lynn,

The circuit implemented is a Sallen-Key Butterworth lowpass filter with a cutoff frequency of 2.4Hz being driven by a low ressitance(OP-AMP) with a biasing and  having the phototranistor ouput to the OP-AMP.

If possible can you post the circuit you simulated having the output as shown in the figure in your post.

Thanks.

.

[johnsold]

The Sallen-Key lowpass filter has C3 connected at the R5-U2 junction, not as your image shows.

The attached file has the SPICE listing I used. Because I did only AC analysis, none of the DC sources are required. In the listing R2 is the parallel equivalent of your R2 || R3 = 50 kΩ. The op amps are modeled as linear voltage controlled voltage sources with gain of 1 (E1 and E2). R5 was omitted because no current would flow through it. In the corrected Sallen-Key circuit it will be necessary.  Resistors Rv1 and Re1 are probably not necessary but assure a DC path to ground from all nodes.  I do not have Multisim. These files are modeled with a SPICE 3f5 engine.

The file test 2 stage.txt is the listing for the circuit in your image and test 2 stage SK.txt is modified to represent the Sallen-Key circuit.

The plot below is my analysis of the Sallen-Key version.

Lynn