Multi-voltage detector using LED as an indicator

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Considering building a voltage detector using an LED as an indicator. Of course, if I use it to test 120 VAC then I need a fairly large resistor. But that would eliminate using it to test for 12 VAC, unless I make a selectable range detector. But then I run the risk of setting it for 12 VAC and what happens to the LED if I touch it to 120 VAC?! Keep in mind I'm not only considering just these two voltages I want to use a single LED and a single circuit to detect AC voltages from - oh, lets say 5 volts on up.

I recently read an article about current mirrors. Would that work? I mean set up a current that is correct for the LED then on the mirrored side, regardless of the voltage, the LED won't see more than its required current. Then, conceivably, I could ground one lead and with the other, poke it at a transformer or other device and detect whether there's live voltage there or not. It wouldn't be used to determine the voltage, just to indicate a live line.

Would this work?
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Thanks Brushrat, I know that. That's the reason for the current mirror - to PREVENT accidental destruction of the LED.
 

Alec_t

Joined Sep 17, 2013
14,280
Here's a suggestion.
This will sense AC or DC from 3.3V up to at least 120V.
The FET needs to be a logic-level NMOS type if low voltage is to be detected. Standby current is zero, so no switch is needed.
VoltageDetector.PNG
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Thank you Alec. Quick question: The "IN" - that's a "Contact" probe, right? I mean not like the NCVD's you can buy at Home Depot where it gets "Near" the live wire and lights up and chirps like a nervous canary. Right?
 

Alec_t

Joined Sep 17, 2013
14,280
Yes, a contact probe. You also need a circuit ground connection.
I don't think it's possible to get a NCVD (as distinct from a current sensor) to work with DC voltages, particularly low ones (but I'd like to be proved wrong).
 

ian field

Joined Oct 27, 2012
6,536
Considering building a voltage detector using an LED as an indicator. Of course, if I use it to test 120 VAC then I need a fairly large resistor. But that would eliminate using it to test for 12 VAC, unless I make a selectable range detector. But then I run the risk of setting it for 12 VAC and what happens to the LED if I touch it to 120 VAC?! Keep in mind I'm not only considering just these two voltages I want to use a single LED and a single circuit to detect AC voltages from - oh, lets say 5 volts on up.

I recently read an article about current mirrors. Would that work? I mean set up a current that is correct for the LED then on the mirrored side, regardless of the voltage, the LED won't see more than its required current. Then, conceivably, I could ground one lead and with the other, poke it at a transformer or other device and detect whether there's live voltage there or not. It wouldn't be used to determine the voltage, just to indicate a live line.

Would this work?
Steinel produced a range of continuity/voltage testers that could handle 4 - 440V. They had 2 LEDs so they could indicate AC or DC with polarity.

They were based on a low thermal inertia film coated ceramic tube PTC thermistor. No idea if those can be ordered off the shelf, but you could look at very bottom of the Polyfuse range.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
I like the idea of two LED's to indicate presence of AC. I'm trying to figure out how to modify the circuit Alec posted (which is appreciated). So far I'm not confident my drawing will work. At this point I'm open to any other ideas, as well as the notion of my original idea of using a current mirror circuit.
 

Alec_t

Joined Sep 17, 2013
14,280
Here's a 2-LED mod for detecting DC or AC from ~2V up to ~230VRMS. One LED lights for positive voltage, the other one lights for negative voltage. Both light for AC. Powered by a couple of coin cells, which should last for yonks.
VoltageDetector2.PNG
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Thanks Alec; I was trying to modify the first schematic you shared but never considered a double power source. Is this something you've drawn up on LTSpice? I tried to download a sim but don't know what went wrong. I haven't used a sim in years, maybe 10 years. I really need to get back to using them.

Nevertheless, thanks for the schematic. I will add this to my list of things to build.
 

Alec_t

Joined Sep 17, 2013
14,280
Do you have LTspice installed? That should open it.
It's also readable with any text editor (e.g. Notepad), but won't make much sense unless you're familiar with Spice.
 

GopherT

Joined Nov 23, 2012
8,009
Here's a suggestion.
This will sense AC or DC from 3.3V up to at least 120V.
The FET needs to be a logic-level NMOS type if low voltage is to be detected. Standby current is zero, so no switch is needed.
View attachment 107604
Here's a 2-LED mod for detecting DC or AC from ~2V up to ~230VRMS. One LED lights for positive voltage, the other one lights for negative voltage. Both light for AC. Powered by a couple of coin cells, which should last for yonks.
View attachment 107641
The first one is questionable since no current flows into the gate of the MOSFET and the LEDs are powered by the 9V battery but the second one has some electrons from Mains power going through LEDs. Does that violate site ToS? :eek:
- Mains powered LEDs
- Transformerless Power Supplies.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
I don't know if it's a violation or not. I DO know the intention is not to have mains powering a circuit. And since this is merely a "Voltage Present" tester it will only touch ANY live source, be it ±5 VDC or AC up-to 230 VAC. Not likely I'll be poking around in any kind of circuitry with that much potential. I MAY from time to time touch 120 mains but that's not the intention. It's meant for a quick and dirty verification of the presence of potential. To determine value (should I desire) I can O-Scope it or use my DMM.

But if you're concerned I can always consider this to be a taboo topic and not ask again. However, when I was working on my amplifier (bridge rectifier shorted) I was checking mains as well as secondary outputs which were in excess of 96 VAC. Pretty close to dangerous if mishandled.

Oh well. At least I got what I wanted. Now it's lunch time. MMMMmmmmm? ? ? I KNOW! PIZZA!
 

GopherT

Joined Nov 23, 2012
8,009
I don't know if it's a violation or not. I DO know the intention is not to have mains powering a circuit. And since this is merely a "Voltage Present" tester it will only touch ANY live source, be it ±5 VDC or AC up-to 230 VAC. Not likely I'll be poking around in any kind of circuitry with that much potential. I MAY from time to time touch 120 mains but that's not the intention. It's meant for a quick and dirty verification of the presence of potential. To determine value (should I desire) I can O-Scope it or use my DMM.

But if you're concerned I can always consider this to be a taboo topic and not ask again. However, when I was working on my amplifier (bridge rectifier shorted) I was checking mains as well as secondary outputs which were in excess of 96 VAC. Pretty close to dangerous if mishandled.

Oh well. At least I got what I wanted. Now it's lunch time. MMMMmmmmm? ? ? I KNOW! PIZZA!
Lunch is long past. I don't know how you guys in Mountain and Pacific Time Zone manage keep up.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
While all this has been very interesting I think it's beginning to get a bit over complicated. My original idea was to use a current mirror circuit so that no matter what voltage I put across an LED, the current never exceeds its max current. All I want it to do is confirm the presence of potential. Could be five volts, could be 120 volts, or anywhere in between. Can be AC or DC.

Later on this morning I'll be in my lab and I'll post a drawing of what I think might work.
 
Top