50Hz is only relevant if you have a pure sine wave.

And it's only with a pure sine wave you can use a simple formula to move between top-to-top or rms.

The question is what you need to measure. A maximum voltage. Or the exact rms value because you need to measure a reasonably exact power? Or if the need is something else?

If you need to measure true rms (as different from guestimated rms based on the assumption that you have a perfect sine wave) then you need to measure all through the curve. You can't just measure the top voltage. And you need to figure out how much distortion you have on the curve - how much overtones you have. These overtones decides how fast you need to sample to get a good rms value. And they also decide if you need lots and lots of ADC bits to get precision enough.

But the issue here is that you have not given us any information about required precision or amount of overtones.