id	summary	reporter	owner	description	type	status	priority	milestone	component	version	resolution	keywords	cc
1990	Symbolic Jacobian fails for functions	Christian Schubert	Willi Braun	"I tried the new Symbolic Jacobian feature on a msl32 model that previously failed with a nonlinear solver error.

Hoowever, if I run:
{{{
omc Modelica.Fluid.Examples.Tanks.TanksWithOverflow.mos +d=NLSanalyticJacobian
}}}
I get a long list of errors which in essence reads:
{{{
Error: Interner Fehler BackendDAEOptimize.derive failed: lowerTank.portDensities[1] = Modelica.Fluid.Vessels.OpenTank__OMC__36.Medium.density(Modelica.Fluid.Vessels.OpenTank__OMC__36.Medium.setState_phX(lowerTank.vessel_ps_static[1], if (-upperTank.ports[2].m_flow) > 0.0 then pipe.port_b.h_outflow else lowerTank.medium.h, {}, 0))
}}}

I presume it is because the module generating the symbolic jacobian tries to differentiate a function. 

I suggest, taht there should be an appropriate error message for the time being. 
In the long run however, it would be beneficial if it would be possible to mix symbolic and numerical jacobians, so that elements or columns may be derived numerically, if no analytical solution exists."	defect	closed	normal	1.13.0	Backend	trunk	worksforme	symbolic jacobian functions	Jens Frenkel