Hi all,

I've got some big gaps in my knowledge of electronics, and I'm hoping you can help me. I'm designing a circuit that uses 120VAC as the input. I'd like to add an LED that lights up if there is a connection to earth. Like one of those little receptacle testers.

The only way I would know how is to put the LED between the line and the ground with a voltage dropper so as not to fry the LED. But that would mean sending current to ground deliberately, and I'd like to avoid that.

Normally I'd just buy a premade gadget that would do the thing for me, but in this case, I don't even know what to search for. I keep getting results for ground fault indicators. What I want is more like a ground happy indicator.

Is there a way to power the LED only when there is a ground connection without sending any current to ground?

 

Is there a way to power the LED only when there is a ground connection without sending any current to ground?
NO

 

By definition, you measure a connections resistance by forcing a current and then measure the resulting voltage drop.

 

Is there a way to power the LED only when there is a ground connection without sending any current to ground?

No, but you can make the current low enough that it will not be a problem or dangerous.
Small, high-brightness LEDs can give readily visible light at less than a mA of current.

 

Most LEDs will not like the reverse voltage they would encounter when used like this though you could add a diode in inverse parallel.
Perhaps a better solution would be a small neon bulb with series resistor.

 

A cheaper and more tolerant way to verify that a ground is present is to use a neon bulb pilot light. It will use a lot less power than an LED. If you are willing to become a whole lot more elaborate, you can use a voltage detector circuit to check for any voltage between the incoming power neutral and the "green wire" safety ground connection.

 

As mentioned, a Neon lamp with suitable series resistor to mimic the live conductor detector screwdriver.
They just need conduction through ones body to ground.

 

Thank you for all the responses. So there is no way to indicate a good ground without sending current to ground? Ok. Then my next step I suppose is to find or make an indicator that sends as little current to ground as possible. I'll look into the high-brightness LEDs and neon bulbs.

The idea of a voltage detector got me thinking. A voltmeter would detect voltage across the line and ground, with only a tiny bit of current (I don't know how much, but its less than 1mA according to my multimeter.) I don't need an actual Volt reading; just an indication if voltage is present. In addition, there are voltmeters that use a separate source for power than what they're measuring. I wonder if I could find something like that: a separately powered voltmeter with no meter.

 

Thank you for all the responses. So there is no way to indicate a good ground without sending current to ground?

The current with a Neon lamp is infinitely small, it has to travel through the body, as in the case of the type in #7 !

 

Thank you for all the responses. So there is no way to indicate a good ground without sending current to ground? Ok. Then my next step I suppose is to find or make an indicator that sends as little current to ground as possible. I'll look into the high-brightness LEDs and neon bulbs.

The idea of a voltage detector got me thinking. A voltmeter would detect voltage across the line and ground, with only a tiny bit of current (I don't know how much, but its less than 1mA according to my multimeter.) I don't need an actual Volt reading; just an indication if voltage is present. In addition, there are voltmeters that use a separate source for power than what they're measuring. I wonder if I could find something like that: a separately powered voltmeter with no meter.

The usual kind of multimeter has a battery for its circuitry and the display but still needs to draw some current from the voltage it is measuring in order to measure it. This may be very low, perhaps the meter will have a resistance of 10MΩ.

 

The testers that include a ground wire integrity test put a current of several amps thru the ground connection. But that is a ground connection integrity test, not a check for a ground connection.

 

I put together some information in an attempt to compare the different ways I know of to make an indicator light. An LED and resistor, an LED and capacitor, and a neon bulb and resistor. Here's what I have so far:


I found an LED that has 38mcd at 1.8V, 2mA. I think that would be bright enough for an indicator.
Also found a neon lamp that uses 120V, 0.5mA. I don't know how bright it is.

I don't know how to find the waste heat from a capacitor. I'm assuming none until I can figure out otherwise. There must be some, but searching didn't find me much on the subject.

I'm also unsure how efficient LEDs and neon lamps are. I'm assuming for now, that both produce about 60% waste heat. I know that it varies from one LED to another, but I have no idea about neon. Searching hasn't helped much here either.

So far it seems the LED/capacitor circuit and the neon lamp circuit are far better than the LED/resistor circuit both for efficiency and for current (I don't know the actual current of the LED/capacitor circuit, but I suspect it's very, very small.)

I've been reading that the LED/capacitor circuit might change it's voltage very easily due to line noise or changes in input voltage. I don't know a lot about that sort of thing. It also makes for a fairly complicated circuit with lots of components.

The neon lamp on the other hand, is very simple. I guess it's less affected by variations in the input than the capacitor. Also, it's got that cool vintage look. I'm kind of leaning in this direction.

 

One more interesting fact about the neon light is that the glow comes from a significant part of the mass of the gas, rather than from a surface on a semiconductor chip. That affects the emission calculations a bit.

 

I don't know how to find the waste heat from a capacitor.

A capacitor for this application is nearly a pure reactance and will dissipate no measurable power.

I'm also unsure how efficient LEDs and neon lamps are. I'm assuming for now, that both produce about 60% waste heat.

Perhaps a little more, but it's still so small as to be negligible for most applications.

 

I reverse-engineered an old defunct C9 LED bulb to see how they made it work in such a small package:




At the time I didn't know what a capacitive dropper was, so it took quite a while to figure it out. Eventually I came up with this:



I had to put a whole string of working ones on my little power meter to register any current. They seem to work really well. Except they're supposedly waterproof, and as you can see in the pictures, they're not. Anyway, my power meter also let me know that these guys have about 0.5 power factor. I vaguely understand that power factor happens when capacitors or inductors get involved and move the peak voltage out of phase with the peak current. Which means the apparent power (VA) is not equal to the usable power (W). In this case, only half the power is getting used to run the LEDs.

So does 0.5 Pf mean that the capacitive dropper is only half as efficient as I'm thinking? Where does the rest of the power actually go?

 

So does 0.5 Pf mean that the capacitive dropper is only half as efficient as I'm thinking? Where does the rest of the power actually go?

There is no "rest of the power".
0.5 PF means that half the volt-amps are doing work (real power), and the other half are doing no work (apparent power).

 

For really low power you could use a capacitor in series with a neon.

 

Just as a mater of interest, I use Superbright Green LED's as indicators when testing logic circuits at 5V. They are perfectly visible when running at 70uA current!

 

Just as a mater of interest, I use Superbright Green LED's as indicators when testing logic circuits at 5V. They are perfectly visible when running at 70uA current!

It is not likely that a 5 volt, or lower voltage, LED will survive in a 120 volt application.

 

It is not likely that a 5 volt, or lower voltage, LED will survive in a 120 volt application.

Naturally, you'd need some form of current limiting in series with the LED, and either another LED in anti-parallel with it, or a normal diode, to avoid reverse breakdown.