Problem with Varistor created with Simscape Custom Component

Hi,

I developed a varistor using Simscape language and it partially works. When the voltage across the varistor exceeds a certain negative voltage (around -1kV) the simulation breaks and results diverge and I have no idea why. On the other side, the component behaves fine when the voltage aplied is positive by I need a component that works well on both polarities.

Btw, I know that there is a Varistor Component in simscape but I realized that it has a bug and therefore I'm trying to develop one of my own.

This is the simscape code: as a Current Source

component MOV
    % Varistor Modeled as a Current Source

    nodes
        p = foundation.electrical.electrical; % :top
        n = foundation.electrical.electrical; % :bottom
    end
 annotations
    Icon = 'MOV_a.jpg';
  end

    parameters
        V1 = { 1, 'V' };   % Leakage to Normal voltage transition
        V2 = { 30, 'V' };   % Normal to Upturn voltage transitino
        R1 = {1e9,'Ohm'};  % Insulation Resistance
        R2 = {  1,'Ohm'};  % Upturn Resistance
        a  = {  30,'1'}           % powerlaw exponent
    end

    variables
        i = { 0, 'A' }; % Current
        v = {value = { 0, 'V' }, priority = priority.high}; % Voltage

    end

    branches
        i : p.i -> n.i; % Current through from node p to node n
    end

    equations
        v == p.v - n.v; % Voltage across between node p and node n

%         k  = 1/(a*R1*V1^(a-1));
%         c1 = V1/R1 - k*V1^a;
%         c2 = k*(V2^a - V1^a) + V1/R1 - V2/R2;
% 

        if (abs(v) < V1)
            i == v/R1
        elseif(abs(v) > V2)
            i == v/R2 + (1/(a*R1*V1^(a-1)))*(V2^a - V1^a) + V1/R1 - V2/R2;
        else
            i == (1/(a*R1*V1^(a-1)))*v^a + V1/R1 - (1/(a*R1*V1^(a-1)))*V1^a;
        end


    end
end

This is the circuit to test the model:

And this is the result of the simulation:

Ass you can see the MOV clamps the voltage in the negative semiperiod but at aroung 1000 V the voltage across the MOV drops (reaching a nonsensical value of 10^11 and then it remains at that value for the rest of the simulation)

I suppose the is an error in the simscape code but I have revised over and over and can't seem to find the bug, please help!

(FIND THE SIMULINK MODEL AND COMPONENT ATTACHED)

Answer 1

[Sayan]

Hi Santiago Alfonso Ospina Botero,.

I understand from your issue that you are simulating the behavior of the existing Varistor block by creating a block using the Simscape language, and the result is not as expected.

I have checked your previous question on the MATLAB answer community titled "There is a problem with the Simscape Varistor Model" where you plotted the I-V characteristics of the Varistor by developing the equations.

In both the Desmos file and the Simscape language file, I have found a common mistake.

• In the Desmos file, while developing the equation of "k" you are using the high-value resistance of leakage, which should be replaced with the low-value resistance of the upturn region.
• In the Simscape file too, you have used high-value resistance, as shown in this code snippet.

elseif(abs(v) > V2)
            i == v/R2 + (1/(a*R1*V1^(a-1)))*(V2^a - V1^a) + V1/R1 - V2/R2;%line 41
else
            i == (1/(a*R1*V1^(a-1)))*v^a + V1/R1 - (1/(a*R1*V1^(a-1)))*V1^a;%line 43
%in both the equation R1 should be replaced with upturn resistance R2 the term 1/(a*R1*V1^(a-1)

You can find the equations in the following documentation.

https://www.mathworks.com/help/releases/R2023a/sps/ref/varistor.html#:~:text=k%3D,NU

Hope this helps in resolving the issue.