How can there be voltage but not power in a circuit when no breaker turns it off?

Question

Second-floor outlet indicates voltage with meter but it does not power a lightbulb. We have repeatedly turned off every circuit breaker in the house and none of them has resulted in the voltage going away.

We believe that this is the only outlet remaining on an old 1920s era circuit but that is based on a previous attempt to identify which circuit it was and now we can't get the voltage to stop.

Where do we go next to comfortably get rid of this thing?

Unfortunately I don't have a meter, just a pen indicator. I assume that it is some sort of phantom voltage as mentioned in comments but if it were, it should go away when we turn off whatever close-by circuit is causing it, right? Yet it never goes away.

Based on some comments, I borrowed a meter from a neighbor and it shows not more than 1 volt. I will try shutting down the whole breaker and testing it then. And then I will disconnect (rather than just turn off) the circuit that I think it's on and see if that takes care of it.

Answer 1

This doesn't answer your question directly, but it will explain why you see what you see with your multimeter.

A multimeter commonly has a 10 Megaohm input. This is very large; at 110VAC the current through the multimeter will be 0.01mA, or 10µA. The power will be 1.2mW.

This is usually seen as a good thing; the instrument should not influence what it measures, and in electronics a 10MOhm input is fine, because you cmmonly have pull-up or pull-down resistors to avoid scenarios where you have stray voltages.

However, we don't have that in house wiring - because we don't care about stray voltages; we care about having power. It's not important if there's 80V and no current available. But it'll throw off any measurements.

The solution is to add a resistor in parallel to the multimeter. This will effectively lower the input resistance of the meter, so it allows you to measure true voltage, shunting away any stray voltages.

As the power grid is built to supply large currents, the resistor can be quite small; passing half an amp is usually no problem in this scenario - but it should at least pull tens of milliamps.

The cheapest resistor available is probably a light bulb. They are built for the voltage, safe to use, and cheap. So simply add a light bulb in parallel to the meter, and measure again. This will show you what's actually present on the wire. Stray voltages will be dissipated in the bulb.

Indicator pens is even worse; they draw no current, but inductively or capacitively couple to whatever circuit they are close to, and amplify that signal by a factor of several thousand. Thus, they can pick up a very weak signal and amplify it. Thus, if it shows no power, it's probably true, but if they show power, it doesn't really tell you anything. Overall, they're next to useless.