LED pilot light powered from 230V

[tcmtech]

If you are curious about the real reverse breakdown voltage of an LED all you need to do is run your AC side capacitor through a full wave bridge and then connect your LED backward and take a DC voltage reading of that.

Just because someone could not get one to conduct in reverse at 40 volts doesn't mean that it wouldn't conduct at 45 or 60 or 100 volts which is still well below the peak voltages that typical household electricity has available.

Every type of diode has a point where it starts working like a zener diode. Some are just far higher than others hence the PRV rating numbers.

 

[ericgibbs]

A few of these, might do it?

Just to set the record straight, I did NOT give Carbonzit a bad rep...OK.!!!
If I have something to say to CZ, I will post my comments to him directly, I am sure he would appreciate it to being done that way.

With regard to the UL jurisdiction or not over the safety aspects of hobbyist projects, I would have thought with all the test work they do on electrical/electronic products it would be wise to benefit from their mass of experience.

Some of the designs posted on the web are down right dangerous.

They fact that someone has built a mains powered project with little regard to safety and claims it works OK is misleading.
It may work OK, but what should be considered, is that if an inexperienced hobbyist who is not aware of the possible hazards, copies your circuit.

Some of the features that improve the safety of project cost so little, for example in the transformer less mains power supply in this thread.
ie: a 1meg cap discharge resistor, say 0.1penny, likewise a series limiting 15R/100R resistor.

When you publish a circuit you are responsible for the safety aspects of the circuit as well as its reliability.

 

[carbonzit]

Some of the designs posted on the web are down right dangerous.

They fact that someone has built a mains powered project with little regard to safety and claims it works OK is misleading.
It may work OK, but what should be considered, is that if an inexperienced hobbyist who is not aware of the possible hazards, copies your circuit.

I must admit you make some good points there. I hadn't thought about people copying circuits they see here without checking them or thinking too much about safety.

 

[adrianbodor]

ie: a 1meg cap discharge resistor, say 0.1penny, likewise a series limiting 15R/100R resistor.

When you publish a circuit you are responsible for the safety aspects of the circuit as well as its reliability.

For me safety is the first priority. In many circuits with 230v LEDs the limiting inrush resistor is set to 1k or something equally, why do you say it needs to be lower to about 15R/100R?

Also what is the chance that after 3 years the X2 cap to blow and to have lot of smoke in the entire house?

For all the people worried about safety of this circuit you might consider the transformer alternative. Here is the same circuit I posted
but it it includes a transformer and a voltage regulator. Power consumption is about the same but cost is a bit higher for this version because of the trafo, about 4-5$ for a 6VAC 100mA encapsulated trafo, very small

**broken link removed**

Thank you.

 

[Gayan Soyza]

Energy savers Never use continuous power on LED's.They will go dim after couple of months.

What they do is they use pulse current on LED's using a driver cct to longer life LED's & better manage peaks.

Hobbiest like you & me use a capacitor supply & make a simple cct to embedded into a small AC lamp holder.

I'd eliminate using zenners on AC circuits.I have lots of scenarios after burning & shorting them.

 

[killivolt]

Just to set the record straight, I did NOT give Carbonzit a bad rep...OK.!!!

Sorry, Eric. I never thought you did. I was just using it as a quick example. (lack of judgement)

I know better, you are not bothered by such things

Safety is another matter all together.

Edit: Hey, I just noticed the MOD.

 

[carbonzit]

Also what is the chance that after 3 years the X2 cap to blow and to have lot of smoke in the entire house?

A very small probability, in my estimation.

For all the people worried about safety of this circuit you might consider the transformer alternative. Here is the same circuit I posted
but it it includes a transformer and a voltage regulator. Power consumption is about the same but cost is a bit higher for this version because of the trafo, about 4-5$ for a 6VAC 100mA encapsulated trafo, very small

Sorry, but now you've gone from bad to worse. While I agree that some protection (like a rectifier before the LED) is useful, your solution is complete overkill. Now, if one were designing this for an agency like NASA, say, you might want to use such a "gold-plated" circuit.

The voltage regulator is completely unnecessary here. LEDs can easily handle normal power-line voltage fluctuations, if the circuit is reasonably designed to start with. So you can eliminate the LM7805 as well as both capacitors adjacent to it.

I wouldn't even bother with a filter capacitor. Why use it? The flash rate (120 hZ) is far above anybody's visual threshold. If you sweep your eyes past the LED you might see a couple "dashes" of light, but who cares?

Unless you like using as many components in your circuits as possible ...

 

[tcmtech]

When you publish a circuit you are responsible for the safety aspects of the circuit as well as its reliability.

I fully disagree. I consider the internet an all the writings on it to be of the same general caliber as those you find on the wall of public mens toilet stall at a bar.

The end user of what ever advice that they found there and applied is responsible for their well being and not any one else.

With that I freely give and encourage this advise to anyone who thinks that I am responsible for their safety.

Take a rock of reasonable size, one you can easily throw over a power line, and add a length of small gage wire slightly longer than the top power line wire is distance wise from the ground and tie the free end of the wire that is opposite of the rock around your neck and then toss the rock over the power line.

If you are 100% certain I or other people are responsible for your actions do the experiment and get back to me.

 

[peterlonz]

Carbonzit
Your ultra-simple solution is appealing.
Unfortunately I can't quite grasp what you have done without a circuit diagram.
I assume the cap is 120 VAC rated?
If the cap is in series, do you mean it's blocking current?
Can you say more about the LED you selected, view angle, MCD output, forward current & voltage etc?
Also what is this costing in watts dissipated?
I ask because recently I used a 12 mcd red LED as a "DC output available" indicator, the equipment was for outdoor use & I quickly discovered that during the day the LED was as useless as the neon mains power on indicator.
Both far too dull to be easily observable from 5 metres distance except at night of course.
I have some other products, for example a AA/AAA battery charger, that also has quite dim red LED indicators - this is damned annoying - I constantly have to approach & align my body to be sure I can check the status.
Is it worth noting that this "AC mains power LED indication" has been addressed by most of the power strip manufacturers. I happen to have a 6 way strip with on/off 10 amp rocker, overload tripping/reset, bright LED indication, & limited spike protection ............... all for $7.50
Does anyone know if there is any commonality to these designs, I think we can assume they are safe.

Peter O

 

[peterlonz]

tcmtech,
Well I hope your safety advice is given tongue in cheek.
Yes the internet at times is a sewer.
It's also a wonderful source of information & inspiration.
Everyone, in the most general of circumstances, has to decide the relative value & accuracy of what they read.
Some folk are better at this than others, that's life.
But, & here's the thing, on a site like this, mostly everyone will assume that the material posted is "reliable".
That does not mean the poster is absolutely responsible for either safety or functionality.
But most, even mere hobbyists with very limited circuit analysis skills, will (reasonably) expect that some care has been exercised to present circuits free from defects.
In effect there is a duality of care required of both the user & the poster, at least that's IMHO & I think it happens to pass the common sense test.
At times we all become engaged in activity outside our level of competence, so spotting potential problems will not always be easy, hence the desirability of warnings, if for example a circuit is essentially either a prototype or is untested.

 

[carbonzit]

Carbonzit
Your ultra-simple solution is appealing.
Unfortunately I can't quite grasp what you have done without a circuit diagram.

Here it is:

**broken link removed**

I assume the cap is 120 VAC rated?

Yep. It's a nice fat mylar or polyester cap.

If the cap is in series, do you mean it's blocking current?

It's acting as a fairly high AC impedance (low frequency, fairly low µF value).

Can you say more about the LED you selected, view angle, MCD output, forward current & voltage etc?

Heh; view angle? MCD output? forward current? what's that?

I just grabbed something out of my LED collection (mostly stripped from discarded electronics left out on the street).

Also what is this costing in watts dissipated?

I should calculate it. The LED consumes practically nothing (~3V @ ~20mA = 0.06W. However, the series resistor gets pretty warm, and is consuming most of the power here. I'm thinking of using a smaller (value) cap and a lower value resistor next time. Let Xc take care of things here ...

 

[colin55]

What is discharging the capacitor?

The reverse voltage and current through the LED will destroy it.

 

[carbonzit]

What is discharging the capacitor?

Nothing, obviously.

The reverse voltage and current through the LED will destroy it.

No.

Look, for the umpteenth time: it works. It's been working for two weeks now, shows no signs of failing.

It is not UL approved.
It is not optimized.
It is not suggested for commercial use.

But it does work.

 

[ericgibbs]

Nothing, obviously.



No.

Look, for the umpteenth time: it works. It's been working for two weeks now, shows no signs of failing.

It is not UL approved.
It is not optimized.
It is not suggested for commercial use.

But it does work.

Look at this link, note this paragraph.
http://www.designledlighting.com/led-polarity

Most LEDs have low reverse breakdown voltage ratings, so they will also be damaged by an applied reverse voltage of more than a few volts. Since some manufacturers don’t follow the indicator standards above, if possible the data sheet should be consulted before hooking up the LED, or the LED may be tested in series with a resistor on a sufficiently low voltage supply to avoid the reverse breakdown. If it is desired to drive the LED directly from an AC supply of more than the reverse breakdown voltage then it may be protected by placing a diode (or another LED) in inverse parallel.

 

[carbonzit]

Well, if you didn't like my earlier circuit, you're gonna hate this one:

**broken link removed**

After building my earlier circuit (up there in post #51), it occurred to me that this could be made even simpler, in the quest for the minimal circuit that would light a LED from line voltage. Why bother with a resistor? All it really does is waste energy in the form of heat. Let good old Xc do the work here!

So after rooting around in my parts boxes, I found a nice big old X2 cap and a sacrificial LED, pulled from an phone answering machine ca. 1980s (green).

Again, it works.

DISCLAIMER:

• This device is not UL approved.
  
• Use of this circuit is not endorsed by any regulatory agency.
  
• If you decide to build this circuit, take appropriate precautions. Use only OSHA/NIOSH-approved work methods and materials. Wear adequate eye protection. Use a class II particulate inhalation filter.
  
• Build this at your own risk.
  
• Not recommended for use in commercial electronic devices.

etc., etc., etc.

I'm going to build this into a little enclosure and run it to see how long it'll last.

So now I'm ready to hear all the responses telling me why this cannot possibly work.

Except that it does ... I should add one more item to my disclaimer:

• This device in no way violates any known laws of physics.

 

[KMoffett]

Had to try it. Variac and 36v transformer (cranked to 50V) 100Ω resistor to monitor circuit current. CH2 across the secondary and CH1 across the resistor. First display was the same for several red and green LEDs that glowed dimly. The second display was for a white LED the glowed brightly. But, for all the yellow LEDs I tried, they pulsed dimly once and the capacitor charged to Vpeak. Reversing the LED resulted in a dim pulse and the capacitor charged in the other direction. Just like a half-wave rectifier and capacitor with no load.

Ken

 

[carbonzit]

Interesting, and thanks for doing that little experiment.

Questions:

1. The obvious one: why do some LEDs light (even brightly) and others not? Is it possible some contain other components (zeners, other diodes)?

2. What's with all that noise on the top waveform? (first picture)

My explanation for what goes on here:

In both my circuits, we have, essentially, an AC voltage divider, with most of the resistance (reactance) in the capacitor. The voltage divides, with most of it across the capacitor, some across the resistor, if there is one, and the remainder (very little) across the LED.

Even though the LED only conducts on half-cycles, it still admits a pulsating wave. So apparently, the voltage-division action still works, at least for some LEDs (but as shown here not for others). Why exactly, I'm not sure. It's not "leakage" through the capacitor, as someone here suggested.

 

[KMoffett]

1. The obvious one: why do some LEDs light (even brightly) and others not? Is it possible some contain other components (zeners, other diodes)?

I was using junk box LEDs for the reds, greens, and yellows. The white was new and might have been a high brightness one.

The interesting thing was that all the LED's, except the yellows were conducting in the reverse direction. And with about the same current level as in the forward direction, enough to discharge the capacitor each half cycle.

Ken

 

[peterlonz]

Fascinating

I certainly don't really know what's going on here.
I'm guessing the reason the LED's perform differently is simply that they have significantly different properties. When you test using junk box components I guess this is an inevitable problem.
I'm not sure I can see a downside: the cap could blow, the resistor could blow, or the LED fail, but so what - any of that can occur in a low voltage circuit.
Nice to see the energy absorbing resistor out of the equation.
Now what about us poor schmuks with 230/250VAC supplies, can you experimenters suggest the cap value for a typical spec LED: 5000mcd, 20mA(DC), 2.8V(DC) indicator.
BTW why not in an electronic project if mains is the power source, anything to be rid of the lousy neon?

 

[KMoffett]

Tried newer red, green, and yellow LEDs. Reds and yellows were the same as before. Reds were brighter though. The new greens behaved like the yellows...no reverse current.

Ken