# Frequency effect of capacitor on three phase

At recent days, I was studying on load affect of transformers. I realized, capacitor changes frequency of voltage at some conditions.

Here the circuit that I used

If we run the simulation, we see everything is normal, frequency is 60 Hz and stable. But if we change the resistances with capacitors, and disconnect the ground, it behaves so different as below:

If we connect ground to primary one, simulation runs for so little time and crash. The frequency is not stable as well...

If I load the system unbalanced, the circuit works but not stable again.

Primary frequency is also so different than secondary.

The all simulations is ran at least 4-5 seconds, so these are stable conditions.

What happens to this circuits? What is difference? Why does capacitor affect the frequency? We generally use capacitors as filters to fix frequency distortion in AC systems, but this distorted the frequency.

• Some type of simulation error or computational rounding error or something. Inductors capacitors and resistors cannot change the frequency of an input signal. Jan 1 at 8:22
• youtube.com/watch?v=vrZ_fMqd8k8 Jan 1 at 8:25
• Don't trust pop-up boxes to calculate your frequency, you don't know how they handle limited time records. Post plots of what the output does, so we can see the waveforms. Ls and Cs do not change the frequency in the steady state, that's as fundamental as energy conservation or thermodynamics, it just does not happen. But they do ring on transients, which can confuse 'frequency calculating' pop-up boxes in simulators that maybe work by just counting zero crossings unintelligently. Jan 1 at 8:47
• See your transformers model. If the coupling coefficient is not exactly 1, you will see logically starting oscillations in your simulation. When this coefficient decrease, the frequency of starting oscillation will change also. Jan 1 at 9:07
• It looks like you have a floating ground in your first picture. Some simulators don't handle that -- which, rightfully, should be the user's job. I don't know how the transformer is modelled, or how Multisim handles these cases, but you need a ground in order for the matrix (circuit) to be solved. If you really intend on having a "floating ground", then add a high valued resistance to ground from a node (e.g. 1meg), to provide a DC path for the solver. Jan 1 at 9:41

Coefficient K of coupling is very important.

• I thnik your results are for not-steady conditions. But the issue what I mentioned occurs even in steady conditions. But thanks for sharing your experiemen results.
– Wtow
Jan 1 at 13:19
• And all these occur when I just play with ground and loads.
– Wtow
Jan 1 at 13:21
• Sorry. Even in steady-state conditions and output ground was forgotten, nothing happens, some results. What is the model in Multisim ? Jan 1 at 15:24
• What do you mean by >What is the model in Multisim ?<
– Wtow
Jan 1 at 17:37
• I see that it is a very simplified model and one must add some components ... Google search " multisim transformer model" multisim.com/help/components/configurable-transformer/… and this ni.com/fr-be/support/documentation/supplemental/12/… Jan 1 at 17:40