Multiple variable resistors cause instability

Hi there,

I have implemented variable resistors like in the image below:

A very simple but often helpful solution.

However, If I use multiple of them, the system becomes very unstable. The voltage sources start overshooting and the systems eventually goes to inf.

I tried limiting the multiplier output with a rate limiter, which helps a but, but not when adding more variable resistors. I also tried smaller sample steps, but again this doesn't help anymore when adding more variable resistors.

If I leave out the memory block, the systems becomes stable, however Simulink can not always solve the algebraic loops, so thats also not an option.

The total system looks like this:

ML and RG are ideal voltage sources with a series resitance, and the metro in the middle is a controlled current source.

Does anyone know why my system becomes unstable?

Many thanks in advance for your help

PS: an impression of the instable voltage over the metro:

[nick]

Hi Werner

Kindly share the Simulink model to help resolve the issue more effectively.

Answer 1

[Zuber Khan]

Hi,

I understand that dealing with instability in Simulink models is an unpleasant experience. I have got a few ideas that might help stabilize the system when you are working with variable resistors and controlled sources:

• You have mentioned that removing the memory block helps but causes algebraic loops. You can play around with Simulink’s algebraic loop solver settings, or try adding a tiny delay using a discrete delay block. That can break the loop without messing up your system dynamics too much. For suggestions on how to resolve a continuous algebraic loop, please refer to the following post:

https://blogs.mathworks.com/simulink/2015/07/18/why-you-should-never-break-an-algebraic-loop-with-with-a-memory-block/

• Since you have already tried a rate limiter, make sure it’s set up to handle rapid changes that might be throwing things off. You might want to put rate limiters on the inputs to the more sensitive parts of your system.
• Kindly check if tweaking some of the parameter values, like resistances or source voltages, makes a difference. This can help you identify the potential root causes of instability.
• You can also check the Simulink Diagnostics Viewer to see if it can point out any issues, like solver errors or algebraic loops.
• You have mentioned that there is a transition, when the system is stable and it gradually becomes unstable upon adding more variable resistors. I would recommend you to inspect the output of every section in the model, rather than focusing solely on voltage values. Examine which outputs are most significantly impacted by the addition of more resistors. This approach should help you pinpoint the source of the issue more effectively.

Hopefully, one of these tips might help you get to the bottom of the instability as without the model it is impossible to arrive at exact conclusion. If you are still stuck, kindly share a simplified version of your model for more specific advice.

Regards,

Zuber