Simscape Pneumatic Orifice Problem

9 views (last 30 days)
I am new to Simulink and Simscape. I am trying to model a helium distribution system. I have been using two constant volume chambers (source and destination tanks) connected via a variable area orifice. The orifice serves as a flow control valve. This system has been working.
I now want to add a constant area orifice on the output of my source tank directly in front of the variable area orifice (my valve). When I add the second orifice the model bombs stating "Nonlinear solver: Linear Algebra error. Failed to solve using iteration matrix. Initial conditions solve failed to converge".
This seems like such a simple change. I think I have missed some fundamental concept for simscape modeling. Any advice will be greatly appreciated.
Thanks,
Phillip

Accepted Answer

Jarrod Rivituso
Jarrod Rivituso on 22 Feb 2012
I am no expert in pneumatics (I had to make sure I was spelling it right when I typed it just now) but I have a thought.
The demo models in the product generally seem to use Constant Volume Chambers to represent pipes, and I do not see too many cases of orifices being directly connected in series.
I wonder, if you add a Constant Volume Chamber between the two orifices, and then just make the volume of the chamber very small (so it represents the pipe between the orifices), does that allow the system to converge? I gave it a shot on my computer (with made-up parameters - again, I know nothing about pneumatics) and it seemed to work.
As far as "missing something fundamental", my understanding of why this problem may have occurred is that both orifice blocks do not seem to contain any state - instead they provide direct algebraic relationships between the inputs/outputs. So, as you chain them together, you end up with some fairly complex equations that Simscape has to solve. When you introduce a Constant Volume Chamber, which contains a state (the chamber pressure), Simscape can more easily solve the equations since the orifice's algebraic relationships are not chained together directly. Again, I'm not an expert so please take this interpretation with a grain of salt. I'd be very happy to have someone set me straight with a clearer interpretation :)
  1 Comment
Guy Rouleau
Guy Rouleau on 23 Feb 2012
I like the explanation Jarrod.
The "Connection Constraints" section of the "Modeling Pneumatic Systems" documentation page lists a set of things to consider when connecting pneumatic elements. In particular, it mentions:
"every node should have a volume of fluid associated with it"
http://www.mathworks.com/help/toolbox/physmod/simscape/ug/br7vtpp.html#br7vtvl-1

Sign in to comment.

More Answers (1)

Phillip
Phillip on 23 Feb 2012
Thanks Jarrod!
I added a small volumes after the fixed area orifice and the model will now run. I tried this same technidue on a larger more complicated model and it did not work. I am now rebuilding the more compllcated model one node ata a time making sure the model runs after adding each node. I'll reread the documentation page Guy mentioned. I had scanned it earlier but evidently not a very good scan.
Cheers,
Phillip

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!