Forums » CircuitLab Support Forum

I don't think the graph that comes up from this sim is correct.

1) Open 2) Simulate 3) Select Time Domain 4) Click on the node between the two caps 5) Click Run Time-domain simulation 6) Puzzle over why the graph shows an oscillation

The voltage there is indeterminate, as there is no DC path from the midpoint to anywhere, and therefore no solution. Try putting a 1 megohm resistor from the midpoint to ground and things should simulate better.

It also helps if you click on "skip initial conditions", otherwise you just see a boring unchanging repeat of the initial conditions, which don't change with time.

A little bit of background to @arduinohacker's comments.

Rule 1 in simulators: every node must have a DC path to ground. It's not always clear in CircuitLab what has and has not got a DC path to ground in some of the components.

Some simulators force a Gmin value in the region of 10e-14 (equivalent to a resistor of 1/Gmin) from all nodes to ground for this reason.

In your case you have a node that has no DC path to anywhere so there is nothing to define the voltage at that node. It could be anywhere between +/- infinity and the circuit DC conditions would still be valid.

If you put a 1G Ohm resistor in parallel with each cap you'll find the node between them sits at 2.5V (i.e. half of the 5V supply). If, as @arduinohacker suggested, you put one to ground from the node then it will sit at 0V.

If you look at the plot, the "oscillation" you're seeing is not actually an oscillation in the circuit, what you're seeing is the numerical noise of the simulator. It's saying this node sits at whatever voltage and every time it calculates a new value it is the same +/- some very small error value.

If you want to see something more interesting happening in your circuit then have a look at:

See also:

https://www.circuitlab.com/docs/the-basics/#ground