False readings on digi meter

[dielectricunion]

I recently did a little toubleshooting on a 3 way switched wall sconce that stopped working.

What I'm confused about is... at the box for the light, I tested the switch leg to see if it had voltage. I read 73v to neutral and 121v to ground.

Someone said this is due to capacitive coupling and I think another said the low impedance of digital meters will allow false voltage readings.

I'm still trying to wrap my brain around theory and I'm pretty green to the field.

Any advice to help me understand this phenomenon?

 

[dfmischler]

Sometimes digital meters give confusing readings because of their high input impedance (e.g. 10M ohms). A capacitively coupled voltage can seem to indicate a problem when no real problem exists. The difficulty is that if you only have a high impedance meter then you have no way to tell if the reading you got indicates a serious problem (i.e. lethal current) or nothing to worry about (i.e. microamps). Old electromechanical meters and related testers (e.g. wiggy) didn't have such high impedance so they clamped these phantom voltages and showed nothing. There are meters out there (e.g. Fluke 117) that include a 3K ohm or so shunt resistor to eliminate those phantom voltages; if you set the meter for low impedance and still see the voltage then there is probably something to worry about.

This is only a problem on AC circuits, since a capacitor will not pass DC current.

120V / 10M ohms = 12uA so a 25uA current at 120V will show up pretty well on a 10M ohm voltmeter.

120V / 3K ohms = 40mA so a 25uA current at 120V will not show at all on a 3K ohm voltmeter.

 

[Little Bill]

You said you were testing at the switch which was a 3-way. I'm willing to bet you do not have a neutral at the box. You may have mistaken a white traveler or white anything for the neutral. The voltage you were reading sounds very much like when there is an open neutral. But in your case no neutral at the switch. The 120V to ground pretty much assures no neutral at the box but confirms it is bonded at the panel.

Edit: I just reread the OP and saw the testing was at the light. Most of what I said is still true though. But instead of no neutral present, it sounds like you have lost the neutral to the light. Or you are testing the wrong wire. You need to check to see where the neutral feed to the light comes from and check the neutral there.

 

[dielectricunion]

You said you were testing at the switch which was a 3-way. I'm willing to bet you do not have a neutral at the box. You may have mistaken a white traveler or white anything for the neutral. The voltage you were reading sounds very much like when there is an open neutral. But in your case no neutral at the switch. The 120V to ground pretty much assures no neutral at the box but confirms it is bonded at the panel.

Edit: I just reread the OP and saw the testing was at the light. Most of what I said is still true though. But instead of no neutral present, it sounds like you have lost the neutral to the light. Or you are testing the wrong wire. You need to check to see where the neutral feed to the light comes from and check the neutral there.

You're right, I had a broken neutral coming from the first switch (box where power is fed) to the light box.