Need help measuring 120VAC main

Question

I built a circuit board to measure 120VAC using a Raspberry PI. I have a bridge rectifier and voltage divider. All using surface mount 1206 parts. Everything works well. When I put a voltmeter on the test point it shows 1/100th the line voltage in DC. However, when I actually look at it on a scope, I see that the ripple is 0 to 1.2v, essentially the full rectified voltage. The filter isn't doing hardly anything.

I'm trying to keep is all as small as possible so I'd like to stick with 1206 package for the filtering capacitors especially since the PCB has already been manufactured using these. I did use a through hole for the inductor it has 5mm lead spacing. Can someone recommend what capacitor and inductor values to use to get a cleaner DC(ish) voltage. The voltage divider is before the filtering so using low voltage capacitors are not a problem.

I was using 1000pF on C1 and 2200pF on C2 and 100mH for the Inductor. Below is the layout:

>-------- L1 ---------->

| |

C1 C2

| |

>----------------------->

Any help would be GREATLY appreciated...

michaelkellett · Accepted Answer

First I must give you the quick lecture on mains safety: USE AN ISOLATING TRANSFORMER !!! If your circuit is connected to the mains supply it is not safe. Now to filtering some safe 60Hz AC to measure it: You don't really need a bridge rectifier - a single diode will do. Use LTSpice to simulate this kind of stuff - it's free and easy to use. This is about what you've got but I'm running AC into it just to see the response of the filter - as you can see its not really working at the right sort of frequency at all - you need a filter which lets through almost nothing at 60 or 120 Hz (half or full wave rectifier) The second circuit is a simplification of yours (but remember you shouldn't build this - if the 1M resistor fails you DIE !!) I think this is what you are seeing on your design the problem is that there is almost no filtering going on. This third go is with some better component values and some simplification: Now it works - its taking a while to settle and there is some ripple. If you were to use a small transformer to give you a nice safe 10V to work with you could have this design where we can reduce some impedances without risk of death ! You can get a faster response at the expense of ripple by reducing the capacitor values or the resistor values. The input impedance of the ADC must be high (> 100k). If you use a transformer you may live to improve it. Have fun, use simulation. MK

Need help measuring 120VAC main

Top Replies