Subject: Stocks and Flows

Posted by Niall Palfreyman on 3/6/2004
In Reply To:Stocks and Flows Posted by Bill Ellis on 3/4/2004

Message:

I'm sorry, Bill - I wrote this post spontaneously out of the delight I was feeling at discovering new insights from working with the SD formalism, and my descriptions were a little vague.

In process engineering one makes the distinction between effort variables and flow variables: an effort variable is (modulo an inertia
factor) essentially the derivative of a flow variable, and a flow variable can be understood in the standard SD sense. So, for example, for a mass on a spring, the force on the mass would be the effort, and the velocity of the mass would be the flow. Force accumulates as momentum, and velocity accumulates as a position coordinate x, which I somewhat sloppily referred to simply as "position". The reason I recounted this experience was because someone in this discussion thread (was it Alfred?) requested examples where one flow accumulates to dictate the value of another flow, which happened to be exactly what I was working on at the moment I read that post.

My delight arose from noticing that the description of oscillations in a spring and oscillations in a radio circuit at first seemed to me to be in disagreement with each other. Conventionally I think of Ohm's law as saying that voltage causes current, whereas in a radio circuit it is rather the case that current causes voltage. SD forced me to think through the situation in a careful causal way which revealed that it is in fact the case that voltage does _not_ directly cause current in any circuit. Rather, voltage in _any_ circuit accumulates as magnetic flux, which in turn dictates the current. Then Ohm's law comes in and says that this current is modulated by a resistance to produce a voltage drop around the circuit which contributes to the total voltage in a circuit.

What particularly appealed to me here was first the discovery that magnetic flux is the electrical analog of mechanical momentum, and secondly the discovery that the inductance equation V=L*di/dt is nothing other than Newton's second law F=m*dv/dt for electric circuits. Neat, huh?

Best wishes,
Niall.