Hi Dieter,
This bug is fixed now in the Subversion repository. Also I made a new nightly-build that includes this fix: http://www.ida.liu.se/~pelab/modelica/OpenModelica/OMC/nightly-builds/ The correct output (check the bottom of the page): http://openmodelica.ida.liu.se:8080/cb/issue/1012?orgDitchnetTabPaneId=task-details-comments Thanks for reporting this bug! Cheers, Adrian Pop/ Adrian Pop wrote:
Hi Dieter, This seems to be a bug. I added it to the CodeBeamer system we use for project planning and development. http://openmodelica.ida.liu.se:8080/cb/issue/1012 I'll look into it. Cheers, Adrian Pop/ dieter wrote:Hi, I've got strange results with my overall model, even though the isolated classes seem to work reasonably while testing. Basically I'd like to simulate a hydraulic circuit, with a pump flow changing in time and subsequent flow resistances. I reduced the problem to the following 3 primitive classes: connector Port Real pressure; flow Real flowrate; end Port; model Pipe Port port; parameter Real a = 1; equation port.pressure = port.flowrate * a; end Pipe; model Pump Port port; equation port.flowrate = -( time - port.pressure); end Pump; //------------------------------ model Test Pump pu; Pipe pi; equation connect(pu.port,pi.port); end Test; // simulate(Test); Probably I've some severe misconceptions about coupling of the components because the model does not reproduce the analytical solution: pi.flowrate = time/(1+a) Results are basically constant (except for the time variable). There is only a step change at the last time step!? Where am I wrong? Without your help, I'm totally stuck now! (I'm using version 1.4.4 of OM.) Dieter -- Best wishes H. Dieter Wilhelm Darmstadt, Germany
