110 VAC LED circuit

Ron Czapala · 2011-08-04 08:21

I found this circuit years ago and wondered if anyone has an alternative and/or comments

I know it is not the safest circuit but it does allow you to drive an LED from 110 VAC without a transformer and with very few parts.
attachment.php?attachmentid=83733

Transformerless 110 VAC LED schematic.jpg

PJAllen · 2011-08-04 08:57

You don't need a resistor (make it all about reactance) or a bridge.

Circuits are only as safe as you are.

The cap should be rated > 200v

Phil Pilgrim (PhiPi) · 2011-08-04 09:24

I'd put two 250V caps in series in case one of them fails short. But other than that, PJ's circuit is simpler. The difference in performance is that his flickers at 60 Hz, whereas the bridge circuit flickers a 120 Hz. But even that could be mitigated by replacing the 1N4007 in PJ's circuit with second LED, oriented the same as the 1N4007. That way the forward voltage of each limits the reverse voltage the other one sees. DOING SO MAY INTRODUCE A FAILURE MODE, HOWEVER, THAT THE HIGHER-RATED 1N4007 IS THERE TO PREVENT.

NOTE: Don't be fooled by the low voltage present across the LED(s). The voltages present ANYWHERE in this circuit compared with earth ground (i.e. your body) can still be LETHAL.

-Phil

Ron Czapala · 2011-08-04 09:39

Phil Pilgrim (PhiPi) wrote: »

NOTE: Don't be fooled by the low voltage present across the LED(s). The voltages present ANYWHERE in this circuit compared with earth ground (i.e. your body) can still be LETHAL.

-Phil

My original circuit with the full wave bridge isolates the LED connections so it is probably the safest from the standpoint of contacts with the LED leads - correct?

piguy101 · 2011-08-04 09:44

I have always liked to put a large, polarized, electrolytic cap at the positive and negative of the full wave rectifier, this will smooth out the voltage yielding a higher average voltage.

Phil Pilgrim (PhiPi) · 2011-08-04 09:50

RonCzapala wrote:

My original circuit with the full wave bridge isolates the LED connections so it is probably the safest from the standpoint of contacts with the LED leads - correct?

ABSOLUTELY NOT! Rectifiers provide zero isolation. If you want isolation, use a transformer.

-Phil

Ron Czapala · 2011-08-04 10:30

Thanks for your insight!

- Ron

Ron Czapala · 2011-08-04 10:56

Since this can be done so simply and inexpensively, it makes me wonder why manufacturers of wall switches and outlets still use neon lights for lighted switches.

I have some high-end ground fault outlets and fan timers that use LEDs but they are pricey. Of course they require a neutral wire while a neon is fed thru the load.

The darn neon bulbs always end up flickering. I'd much rather have a bluish LED wall switch than the orange glow...

erco · 2011-08-04 15:12

I've read that it's bad form to use an LED as a rectifier, that it's wise to install a series rectifier such as a 1N400x.

FredBlais · 2011-08-04 22:55

PJ Allen wrote: »

You don't need a resistor (make it all about reactance) or a bridge.

Circuits are only as safe as you are.

The cap should be rated > 200v

Hi PJ,
could you give me a bit more detail about the current calculations involved in your circuit and the role of the rectifier diode?
Fred

PJAllen · 2011-08-05 06:10

FB,
in re. my calculations:
In ac circuits, capacitors and inductors have impedances (resistance in the time domain), capacitive reactance and inductive reactance.
Capacitive reactance is X_C
The formula (elementary electronics):
X_C = 1 / (2 * pi * Hz * C)
The line freq = 60 Hz
X_C = 1 / (6.2832 * 60 * 0.27uF)
X_C = 1 / 0.000102
X_C = 9803 Ω

--> 120vac / 9803 Ω = 12ma

in re. the recifier diode:
It's there to divert the current from the negative alternation (half-cycle), keeping the LED from getting "back biased" into oblivion.
As others have noted, LEDs make poor rectifiers; their PIV is in the range of... nil, practically speaking.

prof_braino · 2011-08-06 08:39

PJ Allen wrote: »

You don't need a resistor (make it all about reactance) or a bridge. Circuits are only as safe as you are.

How unsafe are LED christmas light strings? In January, I was able to buy strings of 25 to 70 LEDs of various colors in various configurations.
A common configuration was similar to post #2, except there was only a single resistor and two strings of many LEDs in series biased opposite each other. .

My Kill-O-Watt meter measures 0 watts for a single string, and measured IIRC 60millamps for 5 strings. This that many strings, it was approaching usable amount of light.

I was thinking this might be an option for low power lighting, if one could get used to the weird colors.

Is this circuit too dangerous, for say, under counter lighting? Would it be any more dangerous than incandescent lights using 120v?

PJAllen · 2011-08-06 09:26

"How unsafe are xmas light strings? "
If they're UL listed then that's pretty safe, about as safe as anything gets.
[I checked the boxes of mine, many manufactures, all marked UL listed.]

Incandescent (halogen) under-counter lighting is based on 24vac (because there's no code restriction for that), there's a step-down in the mix.

You could still use the capacitive reactance approach with a 24vac step-down transformer. Try that with a 1uF (NON-polarised.)

Beau Schwabe · 2011-08-06 10:42

Here is a link to a thread I posted regarding harvesting cheap LED's from the Christmas light strings... Thread #6 has the actual schematic used in the "approved" off-the-store-shelf Christmas lights.

http://forums.parallax.com/showthread.php?116906-Cheap-Super-bright-white-LEDs

Cluso99 · 2011-08-06 11:53

There have been lots of fires caused by christmas lights here is Oz. This is often due to faulty circuits coming from China. Just because they stamp compliance on something doesn't mean anything here. Of course, in Oz we have 240Vac and it is more lethal than 110Vac.

localroger · 2011-08-06 13:37

Cluso99 wrote: »

Of course, in Oz we have 240Vac and it is more lethal than 110Vac.

My one trip to Europe so far was on business to install a product we sell stateside. Said product has a switching power supply, so compliance with British power was a non-problem; just change the plug. The power supply doesn't care about either input voltage or frequency.j

So I'm working with the British liaison tech and I start to open the enclosure with the power on, and the guy FREAKS OUT and literally rips the plug out of the wall before I can get the enclosure's back off.

ME: What's the big deal? I do this all the time.

BRITISH TECH GUY: There's a heat sink on that power supply that's hot. Very easy to touch accidentally.

ME: Well, I know that, it's bit me a few times too. No biggie.

BRITISH TECH GUY: Yeah, but we have 240 volt power here. *EVIL GRIN*

$WMc% · 2011-08-06 17:42

(BRITISH TECH GUY: Yeah, but we have 240 volt power here. *EVIL GRIN* )
'
I would have replied "Yes but its only 50 Hz, The dwell time at zero crossing is a lot longer than at 60 Hz. Plenty of time to let go !!!"

Beau Schwabe · 2011-08-06 18:00

$WMc% - Yes, but at 60Hz we have 20 more 'opportunities' to let go. :-)

$WMc% · 2011-08-06 18:27

@ Beau
'
This is true
'
Another reason to live here and not their!

prof_braino · 2011-08-07 06:49

PJ Allen wrote: »

If they're UL listed then that's pretty safe, about as safe as anything gets.
Incandescent (halogen) under-counter lighting is based on 24vac (because there's no code restriction for that), there's a step-down in the mix.

I didn't realize you were such a trusting soul. Due to past experience, I assume all markings on all products from China are only guidelines as to what they hope you think are getting, until proven otherwise.
RE: 24vac, good information but I was asking about regular 110-120v incandescent that has just chord-plugged-into-the-outlet type lights.

I have two circuits for LED's. One is the same LEDs from the string and the same single resistor as on the original string, only mounted on a PCB project board.
The other is similar to the circuit in #1 without R1and C1. There is a resistor (under 1K) in serial with the LEDs. The idea was to have enough LEDs so the voltage is as close to 110v without going over, and so a minimum value for the current limiting resistor could be used. The idea was that the least power would be dissipated as heat.

The first circuit has been operating happily since xmas. The second is my own concoction. It appears to work but I didn't build it onto a board yet, I could not determine if I missed important. This is my context for asking.

PJAllen · 2011-08-07 07:40

I didn't realize you were such a trusting soul. Due to past experience, I assume all markings on all products from China are only guidelines as to what they hope you think are getting, until proven otherwise.

I guess you can't trust anything anymore then because just about everything is manufactured by the ChiComs.
There's not nearly enough scrutinisation of products from China, but nobody was sticking UL listings on the melamine cat-food or the cadmium kiddie jewelry.
Do you have any sources documenting any misleading or counterfeit UL listings?

Anyway, to get back on track, Ron asked for LED circuit alternatives to his. I replied with mine and I stand by it 100%. I've had one running 24/7 in a project box for probably longer than I'd care to say and everyone's "amazed" ("...must be a real little xfmr in there...") that the box's contents are: a fuse, a diode, and a cap - and nothing has ever failed. [I replaced the LED 5 years ago, it was getting dim.]
It'll work in Angle-terre and the Antipodes, too, just use 0u1 instead [use two 0u22 (200V) in series, if you can't find a 0u1 400V.]

Ale · 2011-08-07 10:00

I have a question... those capacitors used in series are always polyester-based... why a ceramic capacitor cannot be used ? is it because they may not dissipate the heat due to the small size ?

BTW: Either 110 or ~230V can be lethal, please use appropriate precautions and safety measurements. (There are small <1W transformers that can be used to isolate the led, quite small actually!).

PJAllen · 2011-08-07 14:45

Ale,
The type of capacitor isn't important, so long as it's non-polarised and suitably rated.

BTW, wiring a "small <1W transformer" still involves connecting to the lethal voltage. There's no escaping that. If that's not your cup of tea then beg off.

BTW, owing to dubious quality or dubious country of dubious origin or otherwise, or dubious implementation, or any dubious use or misuse, a malfunction might still result - even in a loss of supposed "isolation".

Just stop already.

Humanoido · 2011-08-08 03:06

There is no valid UL in China because one cannot be sure of product verification validity. However, products made in China can ship to USA where they can undergo UL testing and have support by an American company. It's the same way that Apple has products assembled in China and shipped back to USA where the product is UL approved. There are many FAKE products in China with the FAKE UL Approved sticker. Sorting this out is a can of worms.

prof_braino · 2011-08-08 06:22

Humanoido wrote: »

There is no valid UL in China . .... Apple has products assembled in China and shipped back to USA where the product is UL approved. There are many FAKE products in China with the FAKE UL Approved sticker. Sorting this out is a can of worms.

but nobody was sticking UL listings on the melamine cat-food or the cadmium kiddie jewelry.
Do you have any sources documenting any misleading or counterfeit UL listings?

Not apart form the pattern established in the news.

http://topics.nytimes.com/top/reference/timestopics/subjects/m/melamine/index.html

In 2008, about 300,000 children in China were sickened and at least six babies died when some manufacturers added it to infant formula to make it appear more nutritious. The chemical helped cover up the fact that the milk had been diluted with water to increase the amount the milk producers were able to sell.

We can only trust the party responsible. Apple is responsible for Apple products, and may make efforts to ensure the products conform to requirements. If there is no Apple directly responsible, then there is no reason to assume there is any conformance to anything beside "immediate profit", not even "common sense".

We had a new off-source board maker, chosen for price and it was "the biggest, longest established in the area". The fiducial marks on the board did not line up properly. On inquery we found that the marks were not used for image registration, the folks put them on "because they normally appear on board." Same with UL markings, and Lead Free marking. We had not asked for UL test yet, and they used lead in manufacturing; these had been added "as decorations". Later we found out the "factory" was the front room and lawn of someones house; yet is was true that the operations was the biggest, longest established in the area. Unfortunately, the only documentation is this anecdote, so you need not accept it.

But my interpretation of the data is no claim on compliance for any product from China is to be accepted at face value, unless a local responsible party vouches for it; and even then be extremely cautious.

Back to the original question: What we are trying to find out is whether these circuits are safe compared to a standard mains incandescent light. On the MAINS side of the rectifier, there is 110V, so we know this needs to be treated accordingly. On the DC side, do we also have to assume the same, even though the voltage drop across any given LED is 3.x volts (due to the 36 or so LEDs in series)?

My interest is the most light for the least energy consumed, and the idea was the smallest current limiting resistor would disipate the smallest amount as heat.

110 VAC LED circuit

Comments