Guys, the AC voltage sense module Samo found is just a transformer with an op amp voltage follower; it does NOT do RMS voltage conversion for you, you only get to sample the varying AC voltage with the Arduino. RMS voltage calculation isn't terribly hard but if you need the answer fast and in real time with some precision, you will want to use a timer, and an interrupt routine. I just did this for a DIY sine inverter project. My RMS code is intentionally not interrupt driven; my main timer1 routine does very critical interrupt driven timing for the inverter so can't have any other interrupt events going. The RMS voltage routine samples a half wave when signaled by the timer routine that the start of the wave is beginning. The routine watches the timer counter in a tight loop to wait for 2 ticks of the timer to know when to sample the A/D. After the 65 half wave (8.88 msec) samples, adding each sample squared to a sum, the result is divided by 65 and the square root taken. That is the RMS voltage. In my case, I rectified the AC from a small transformer, and scaled it so that the voltage peak is about 3.75V for sampling by the 10 bit, 5V A/D so that I get better resolution. I believe the unit Samo found centers the AC on 2.5V, so you'd have to subtract 512 from each sample before squaring and adding to the sum of the squares.
I don't think real time RMS calculation is really in the domain of novice programmers. Fortunately there are a couple other alternatives.
There are IC's that do analog RMS voltage conversion for you, though they are a bit spendy (I think about $10). They are slow, you only get the answer delayed by about 100 milliseconds, but are very accurate and the output is a continuous DC voltage representing the average AC voltage of the last 0.1 seconds. I've used one, and they can be handy.
The difficulty in real time RMS voltage calculation or cost of the analog conversion chip (and delay) is why most AVR's don't bother. They rectify the AC, scale it with a resistor divider, slightly average it with a very small capacitor, and call it the DC representation of the AC voltage. The RMS voltage of a true sine wave is thus computed from the peak voltage (peak/1.4=RMS). The averaging cap is especially needed with an ST head to filter off the spikes on the waveform peaks. It works acceptably for most sine-ish waveforms. It's what my own AVR does, because it has no processor and I needed a fast simple answer to allow fast correction for Lister flicker reduction.