Problem with controlled pneumatic pressure source

Hi,

I have been developing a model of a pneumatic distribution system. During the initial development I used a constant volume pneumatic chamber behind a valve as the pressure source for the system. I now want to replace the constant volume pneumatic chamber with a model of a compressor. It appears to me that the controlled pneumatic pressure source is a reasonable component to use as a first cut at the compressor. However, when I substitute the pressure source for the chamber I get the following:

Nonlinear solver: Linear Algebra error. Failed to solve using iteration matrix. Initial conditions solve failed to converge. Problems possible for transient initialization, as well as stepsize control for transient solve, due to equations of one or more components: 'manifoldTest/Controlled Pneumatic Pressure Source' 'manifoldTest/Pneumatic Atmospheric Reference' 'manifoldTest/4500 PSI Valve/4500 PSI Solenoid/Pneumatic Piston Chamber'

Problems possible for transient initialization, as well as stepsize control for transient solve, due to equations of one or more components: 'manifoldTest/Controlled Pneumatic Pressure Source' 'manifoldTest/Pneumatic Atmospheric Reference' 'manifoldTest/4500 PSI Valve/4500 PSI Solenoid/Pneumatic Piston Chamber'

I have been careful to include a volume of fluid with all nodes. The 4500 PSI valve mentioned in the error is an electrical solenoid with dampening that is opposed by a pneumatic piston. When the solenoid is extended a variable area orifice is opened to the area of the connecting pipe. This valve works with no errors when connected with the pneumatic chamber pressure source. It seems a little strange to me that the error is reported twice for the same three components. Any advice on how to address this issue will be greatly appreciated.

Thanks,

Phillip

Answer 1

[Guy Rouleau]

It is difficult to give an accurate answer without seeing the model.

Is it possible that, at one node, you have a volume with an initial pressure that is different from the pressure specified by the pressure source?

From what you mention, it seems that a constant volume pneumatic chamber behind a valve works, but the pressure source does not. With a valve, you can have 2 initial pressure on each side. This might be what allows the simulation to starts. Having a pressure source and a volume on the the same node might not be a good idea, because you must ensure that their specified pressure at t=0 are identical.

I hope this helps.

Answer 2

[Phillip]

Guy,

I appreciate your response. The valve has the pneumatic piston that isolates it from the volume. http://lindelowkoon.com/componentTestModelControlledSource.mdl and http://lindelowkoon.com/componentTestModelTankSource.mdl are examples of a simplified system that works with the tank and does not work with the controlled pressure source. Essentially the simplified system has a tank that is isolated by a check valve and it feeds a regulator with relief valve into a second tank. This is a system that is pressurized from a compressor and then operated from the tank behind the check valve.

Thanks,

Phillip

[Guy Rouleau]

I cannot confirm because all the blocks in the model are from a library not available, but I would bet on placing an orifice between the pressure source and the valve.