Scope of 'ground' discipline

Hi

In this paragraph

"3.6.4 Ground declaration
Each ground declaration is associated with an already declared net of
continuous discipline. The node associated
with the net will be the global reference node in the circuit. The net
must be assigned a continuous
discipline to be declared ground."

ground declarations have to be associated with continuous disciplines.
It doesn't explicitly state whether the association need be done at the
same scope as the ground declaration or not. In the example below (from
6.2.2), the association for ground in module sigmadelta is done in the
instantation C2 of comparator, which to me implies that the association
need not be at the same scope.

Do you think that it should be explicitly stated in the LRM that the
scope of the association can be different? Alternatively, if the scope
of the association should be the same as the gound declaration, there is
an error in the example below.

module comparator(cout, inp, inm);
  output cout;
  input inp, inm;
  electrical cout, inp, inm;
  parameter real td = 1n, tr = 1n, tf = 1n;
  real vcout;

  analog begin
    @(cross(V(inp) - V(inm), 0))
      vcout = ((V(inp) > V(inm)) ? 1 : 0);
    V(cout) <+ transition(vcout, td, tr, tf);
  end
endmodule

module integrator(out, in);
  output out;
  input in;
  electrical in, out;
  parameter real gain = 1.0;
  parameter real ic = 0.0;

  analog begin
    V(out) <+ gain*idt(V(in), ic);
  end
endmodule

module sigmadelta(out, aref, in);
  output out;
  input aref, in;
  ground gnd;
  comparator C1(.cout(aa0), .inp(in), .inm(aa2));
  integrator #(1.0) I1(.out(aa1), .in(aa0));
  comparator C2(out, aa1, gnd);
  d2a #(.width(1)) D1(aa2, aref, out); // a D/A converter
endmodule

Regards
Paul Floyd

-- 
Dr Paul Floyd
Mentor Graphics Corporation
-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.