Looking for some pointers towards modelling the current at each junction in this parallel circuit.

I'm using iCircuit to model it, but I can't export its data readings to my custom software.

The voltage is attached at the top-left. I move the ground connection around and connect it to each junction. Each resistor is 1 Ohm.

I'm not looking for a step-by-step or handholding, just a tip on a reasonable place to start. I understand the basics of parallel resistance. I don't understand how to build them up in a way that model the current flow in this circuit.

Thanks!

 

That should be http://icircuitapp.com/

 

Would anyone recommend nodal analysis over loop analysis here?

 

Nobody in their right mind builds a circuit like this. I saw something like this in integral calculus class. Called an infinite grid if I remember. You might get an answer from a math forum.

 

Oh awesome, knowing the name of the problem made it a lot easier. There's a ton of stuff on infinite grids. Thanks!

I'm designing the circuit so it can be cut into smaller pieces. The result has to have the same circuit pattern as the source.

As I mentioned, iCircuit models it - unfortunately I can't export it.

 

I can't believe I pulled that out of my <snip>. That was 40 years ago!
Besides, there are math wizards on this site that put me to shame.
I was just guessing because nobody else was stepping up. Glad it helped.

The point is, this is not a practical circuit. It is a math exercise.

 

I never argued it's practical haha. However, it's the only pattern I can use given the constraints. Mainly: a square subset cut from a source mesh must itself be a mesh.

Anyhow, I've interfaced Falstad's open source circuit simulator. I didn't know it initially, but this java codebase was used by the author of iCircuit.

It's also worth pointing out that it's not actually the same problem as an infinite resistive grid. The 'infinite' aspect only really comes into play for very large edge lengths. I'm dealing with 'finite' edges ranging from 10-50cm.

Anyhow, thanks again. Your replies helped me better formulate the problem.