{{>toc}}

The potential source effect is defined by two parameters:

• Source Potential P: the desired ideal potential rise from port 0 to port 1,
• Internal Conductance G: the internal conductance of the effect.

The potential source effect between two nodes has this footprint in the system of equations:

In the Companion Model, a potential source is exactly equal to a conductance G in parallel with a flow source w = P·G:

To help visualize how the Potential Source works, let’s say we want to create a potential rise from Ground (at port 0) to node 1 of P = 120. Let’s arbitrarily pick a conductance G = 2. In the companion model, the Source will continuously force a constant flux w = 120 * 2 = 240 into node 1. This exact same amount of flux will circulate back to Ground through the Conductor when the node 1 potential = 120, because the conductor’s flux is w = dp * G = -120 * 2.

You can create these separate Conductor and Source links to see it work for yourself — you’ll see the circulation flux of 240 in the Source link, and -240 in the Conductor link, all while constraining the node potential to 120.

The Potential Source link does the same thing internally, but the internal circulation flux described above is never actually calculated since it is abstract.

The Potential Source’s internal conductance G determines how effectively the potential source effect can control the potential rise between the nodes in the presence of extra flux to/from the rest of the network (“under load”).

Another way to visualize the Potential Source is as a resistor in series with a perfect potential rise like shown below. If we force a flux w = 10 through the potential source (as in a demand from the rest of the network, etc), then the actual node voltage will be different from the ideal source potential because of the resistive losses in the internal resistance. The example below uses the same conductance G = 2 and source potential P = 120 as above, and shows how the actual potential is 5 less than the ideal because the internal resistance dissipates dp = w / G = 5.

Real-world potential sources, such as electrical betteries and fluid fans, are never really ideal and always have some internal losses that we can model with a finite value of G. If G = 0 then the link can have no effect because both of the conductance and source companion effects vanish and the two nodes are isolated.

An ideal potential source has zero internal resistance, and therefore infinite internal conductance. We can’t actually use infinity for G because it breaks our math. But we can come close by using a very large number. A very large G causes almost no internal potential drop, and therefore maintains the link’s ideal source potential rise under all but extremely large fluxes. GUNNS places an upper limit of 1.0e15 on all conductances. This is a good value to use when modeling an ideal potential source.