I am planning on building a simple AC power monitor as shown in the attached figure. The 500 kΩ resistors are to limit the current out to the test terminals in case a hapless human (me) accidentally touched them (the 0.1 ohm shunt will be wired on the low side, but of course I could unknowingly plug it into an outlet with hot and neutral reversed). I chose 250 μA as a current limit because it is half the typical recommended maximum leakage current for AC operation and is well below the human current threshold. I could protect against bad polarity by wiring up three neon bulbs with current limiting resistors in the usual AC outlet tester way. If I can find a 25 W resistor, this should let me test to about 15 A.
To be a more safe device, the monitor lines could come through some transformers. But I don't feel this is necessary. I'd like to get the collective wisdom of the folks here about whether the pictured design is considered a reasonably safe design. It's going to go into a metal junction box and the jacks will be flush banana jacks. The input side might be connected to a DPST toggle switch.
Here's the reason why I think it's safe. The resistors would have to fail in a short to expose the user to a potentially hazardous voltage (and you'd have to stick a screwdriver down into the flush banana jacks). And, for a properly wired unit plugged into an outlet with the proper polarity, there's only one resistor short that would be hazardous -- the one connected to the hot on the voltage monitor. If it was plugged into an outlet with hot and neutral reversed, then three lines are potentially hazardous one of those three resistor shorts.
I'll use 1/4 W carbon film resistors that I have on hand.
Under what conditions can a resistor fail as a short? I've never heard of that happening, as long as the resistor was good to begin with. To be even more conservative, each resistor could be replaced by two 500 kΩ resistors in series; this greatly increases the reliability because two resistors in series both have to fail as a short.
This thing will get used infrequently and mostly indoors. My usual connections will be to the two channels of a scope so I can look at the current, voltage, and power waveforms.
To be a more safe device, the monitor lines could come through some transformers. But I don't feel this is necessary. I'd like to get the collective wisdom of the folks here about whether the pictured design is considered a reasonably safe design. It's going to go into a metal junction box and the jacks will be flush banana jacks. The input side might be connected to a DPST toggle switch.
Here's the reason why I think it's safe. The resistors would have to fail in a short to expose the user to a potentially hazardous voltage (and you'd have to stick a screwdriver down into the flush banana jacks). And, for a properly wired unit plugged into an outlet with the proper polarity, there's only one resistor short that would be hazardous -- the one connected to the hot on the voltage monitor. If it was plugged into an outlet with hot and neutral reversed, then three lines are potentially hazardous one of those three resistor shorts.
I'll use 1/4 W carbon film resistors that I have on hand.
Under what conditions can a resistor fail as a short? I've never heard of that happening, as long as the resistor was good to begin with. To be even more conservative, each resistor could be replaced by two 500 kΩ resistors in series; this greatly increases the reliability because two resistors in series both have to fail as a short.
This thing will get used infrequently and mostly indoors. My usual connections will be to the two channels of a scope so I can look at the current, voltage, and power waveforms.
Attachments
-
142.4 KB Views: 44