Light emitting diodes are becomming very popular these days people are using them instead of other lights because they are cost effective .Because leds are connected to the AC Supply it produces ripples at the outer end which results in flickering.Any light source which is connected to AC supply results in flickers.
However flickering can be reduced with different methods one method is to use high frequency switching supplies, which helps to reduce the AC component of the main supply and reduces the flickers but not all flickers are avoidable .Remaining flickers can be removed by attenuating g the AC component at the output
My ex has epilepsy. One year we both fell into the Christmas Tree as she had a seizure and I tried to catch her. I would suggest that the makers of AC LED lighting be very careful about flickering. Flickering lights can be a trigger for epileptic seizures.
Without researching the issue I'd guess that LEDs have a sharper on/off brightness response than a heated wire does.
On another note, I've taken night shots at the drag races; the ones where I slow the exposure and follow the cars have clear car images, blurred stationary stuff (fences and guard rails for instance), and stretched out multi-pulsed images of the high wattage track lights... The first time I saw it it was a surprise.
Today LED lighting is not limited to festival lightings, they are also used in day-today life. old-fashioned incandescent or harsh fluorescent are replaced by bright and shiny new LEDs. The local supermarket, the office where you work, the train station, restaurant or department store—all of these businesses are seeing the benefits of this rapidly developing technology. The light produced by LED bulbs is brighter and warmer and more focused than ever before, and business owners love the savings to their energy bills.
This approach may well shorten the lifetime of the string of LED's. The LED's are brighter because the duty cycle has increased from about .5 too near 1, the average current has increased by a factor of 2 and device (LED) heating has certainly increased. It appears in the video that one of the LED's in the demo string of white LED's has been replaced with a red LED. Was this due to failure of the white LED operating under the more stringent conditions?
The lifespan of LED's on holiday strings (in my experience) is freakishly short. BOM cost is most definately paramount, so I think they tend to use LED "seconds" or output from questionable chinese firms. Even with good LED's, 100k hour mtbf with a well lit house will result in a measurable # of LED failures per season.
As to the full wave rectifier, I'm surprised it works at all. Certainly doubling the fequency will help with percievable flicker. But, the strings I've used seem to be set up as two anti-parallel strings (each half lights up on alternating cycles). I would suspect such a string would only light half the LED's with a full bridge. I'll have to give this a try (perhaps my assumption of anti-parallel isn't correct).
Good questions, Manning6. From your assessment, it seems the LED lights would last about half as long as they would without the flicker correction. If the flickering bothers you, this may be a reasonable trade-off.
My circuit in Gadget Freak case #230 solves the problem of flicker and stress on the LEDs through the use of an electrolytic capacitor and an energy efficient dimming method. I have built 4 such lamps, which I use every day. The circuit is cheap and simple, but not as much so as your bridge rectifier idea. My circuit also employs a bridge rectifier. My circuit is a lot less stressful to the LEDs and the eyes than a triac or PWM type of dimmer. One of my LED lamps is an LED spotlight from Lowe's, during Christmas season, that I converted to an indoor LED spotlight by replacing the wasteful Chinese electronics with my circuit. The flicker in the Chinese circuit, using a 2.2uF filter capacitor was very bad, where mine uses a 100uF capacitor to smooth out the ripple (and therefore the flicker).
My circuit could be built as an add-on device, like yours.
The answer to just about any why question in this context is going to be money. Granted, the rectifier is quite inexpensive. The margin on such products is probably quite low - adding the rectifier while staying cost-competitive won't work.
There's also the cost of getting this now-new product tested by UL so a company can affix the all-important UL label.
TJ, with the number of people who can perceive the flicker and are bothered by it, it could be a competitive advantage to reduce or eliminate it. Even if there may be additional cost, I believe many consumers would pay it.
I have encountered strings of outdoor LED lights made by Phillips that are wired in such a way that the first two LEDs at each end form the bridge rectifier. I rewored these to elimiate this feature which had thess four LEDs still flickering and dimmer than the rest.
It's surprising that design made it into production, Szyhxc. Perhaps Phillips outslourced it and then didn't run quality control on the finished product. This can happen sometimes when the design and production are both jobbed out to an original design manufacturer.
I had several issues with the Phillips strings but they did have the correct bulb spacing to replace my 240 c7 incandesents saving me $50 per season. More details seem way beyond the scope of this Gadget, unless someone is truely interested.
I've bought quite a number of Philips LED Christmas strings over the past 6 years, and they are all wired using LEDs as rectifiers, and they are wired so half of them light up, on each half cycle of the AC. They have been a massively huge pain, from the LED failure stand point.
I buy the brand for it's color qualities, even though it takes days (*yes* days) each Christmas to get all the strings working every year. I like their white, and I like the wavelength they use for blue.
Unfortunately Philips changes the LED socket construction in some way each year, so I can't buy a new string and use the LEDs from that one.
Being here in Oregon it rains all the time in December, and on top of it all the leads on the LEDs rust/corrode/etch away. So, DigiKey gets an order for LEDs every year, on top of the ones that get blown because Phillips was too cheap to use a real rectifier. You know what, the DigiKey LEDs ended up being higher quality, they don't use the super cheap leads that are iron which then has a conductive plating over it (common in uber cheap bargin basement low-end China/Hong-Kong LEDs), which is part of why the Philips LED leads were actually rusting. The "de-plating" of the conductive coating on the Philips LED iron leads is another mess, it occurs due to water getting in the socket, add electricity, and you end up with a Middle School Earth Science level mess to say the least.
Philips must make a major killing with their huge profit margin- and maybe drive sales due to failures. Yes, these were in fact actually outdoor strings. Warranty support has been impossible to actually get, they want the original box and the original receipts.
However, last year I cleaned each socket, put many new LEDs in, and then filled each with dielectric grease before putting the LEDs back in. Guess what, all the failures I have to fix after the displays nearly quit, besides the LEDs getting blown by spikes on the AC power...
I too have several led strings that I use all year. All are on (or share) one my FWBRs. I have three strings of white that surround 3 walls of my sunroom and I have two custom made strings of Red White and Blue that sit behind two columns of glass blocks between my garage doors. I have had no problems with the white but I do tend to replace a couple blue leds each Christmas when I replace the strings with Christmas colors and use the down time to clean and repair the RW&Bs. (These are the oldest leds I have at about 6 year. I have no complaints.)
I only use Phillips where I replaced my old C7s. The Phillips strings are the only ones I have found that have the same 12" spacing and the hang on the 240 legacy staples in my house. I had heard about the rusting and coated them with grease when I first bought them on sale after the hoildays. I got through one season with them with no burnout. (I did buy a spare string but of far it is intact.)
I had no idea that seeing the flicker was so rare. The same effect is true with fluorescents, but more so. Before LCD screens, I would always change the refresh rate at any monitor I used because they would beat with the office lights. I don't see the flicker in LEDs so I must be less susceptible.
According to studies about 1 in 4,000 people are highly susceptible to flashing lights cycling in the 3 to 70 Hz range. Such obvious flickering can trigger ailments as serious as epileptic seizures. Less well known is the fact that long-term exposure to higher frequency flickering (in the 70 to 160 Hz range) can also cause malaise, headaches, and visual impairment.
Others have alluded to it, but no one has mentioned (that I saw anyway) the reason for the flicker; the cheap stings of LED's use a single diode to produce half-wave rectification. adding the bridge make it full-wave. This doubles the ripple frequency (from 60 to 120 Hz). Other's have mentioned, too, that the added brightness will probably mean a shorter life. I'd let the string run for a while with the bridge, and then check to see how warm/hot they are. If the temp rise isn't too great, the life-span effect would probably be small. Heat is definately the biggest enemy of LED's.
This can be further inmproved with a Cap and a Voltage regulator.
Output of a Full wave bridge is 1.417 X the input VAC. the string divides it over the number of bulbs but the resultant current is likely to exceed the rating.
The half wave was going to be 1/2 the VAC voltage, you are more than doubling it.
Adding a Cap on the Bridge output will smooth out ALL the ripple / flicker but raise the voltage even more. Thus the need for a voltage regulator. I'd set it at 48 VDC to avoid any hassle from the regulatory folks.
Another possibility, if this is a 3 wire string, the AC may be split between 2 sets of opposing polarity. If this is the case, rewiring them to series will reduce the need for regulation.
ramjet, the peak output of a bridge rectifier is equal to the peak-to-peak input, less the diode drops. The apparent increase does not appear without the output capacitor. Likewise for a transformer driven full wave rectifier. So the LED strings are already set to handle the voltage. Where the difference occurs, and why this device eliminates the 60Hz ripple, is that the strings in question only use one half of the waveform, so that the string is only lighting on the positive peaks, which gives the flicker. The bridge rectifier supplies positive peaks for both half cycles, thus raising the flicker to 120 Hz, which the eye does not follow.
There does appear a problem at this point because the string system is now dissipating twice as much power, since the energy per second has doubled. That is the reason to consider lowering the voltage input to the string.
Ramjet wrote: "The half wave was going to be 1/2 the VAC voltage, you are more than doubling it."
You've got that wrong. A scope trace of half-wave recitified AC across the LED string will look like a bunch of humps from 0 to peak voltage with gaps between them of the same width as the humps.
A trace of full-wave rectified AC across the LED strings will look like a bunch of humps of the same amplitude as the half-wave humps but instead of gaps there will be a continuous string of humps--the gaps will be filled in by humps.
The peak voltage will be the same between half-wave and full-wave cases.
The peak current will be the same between half-wave and full-wave cases.
The power dissipated in the LEDs will be twice as great in the full-wave case because the duty cycle is doubled.
As noted, as long as the LEDs don't get too warm this is probably not a significant life limiter.
The LED flicker drives me crazy, so I'm going to have to make some of these up for the strings that I recently purchased after Christmas. Maybe the light quality will be improved enough to use them inside.
This is a clever idea. I agree with others that the typical holiday LED strings are very poor quality and may fail when the duty cycle is doubled. I would suggest safer construction since lethal volages are involved. Mositure is often a consideration during the holidays too. Electrical tape is subject to deterioration over time and damage due to sharp edges. Even though LED strings are fuse protected, the described rectifier and wiring are not. I would suggest buiding this in a grounded water proof enclosure and provide fuse protection.
The main problem with running the cheap Xmas LED lights on a full wave is that 1/2 of the set is wired for the positive half cycle and the other half of the set is wired for the negative half cycle. So if you run a set on a full wave you light up only half of all the LED's. Reverse it and you light up the other half of the LED's. You could rewire all the sets for one polarity, but that's a time consuming PITA. Also, as others have mentioned you are doubling the average forward current of the LED's, which may or may not cause the life to become too short for your needs.
My solution was far more complex. I reduce the AC line slightly with a 60 Hz power transformer, rectify, filter, then chop with a 1KHz MOSFET full bridge. The 60 Hz flicker becomes 1KHz and no one complains.
Marty48, I agree that this should be housed in a waterproof box. I also recommend strain relief on the wires. One good pull on the device as presented could put bare wires on your ladder or wet bushes. And as someone mentioned, it should have a warning label that it is for LEDs only.
I like the idea, which is mostly a discovery of the fact that the light designers cut costs to the bone by eliminating a rectifier and accepting a 60 Hz rather than 120 Hz flicker rate. In the early days of AC power and movies it was determined that a 60 Hz flicker was just on the threshold of annoyance. Old-style TVs and incandescent bulbs smooth it out somewhat due to the persistence of the phosphor and thermal inertia of the filament, which LEDs don't have.
My concern, which should have at least been given a mention in the article, is that someone might inadvertently plug some device other than an LED light string into the rectified outlet. I'm sure that, as presented, this would never receive any safety agency approval. The problem is that if any device that utilizes an input transformer or induction motor is plugged into the rectified source, the transformer or motor will look like a near-short to DC and will at best blow the device's internal fuse, if one is provided, and at worst, will overheat the transformer or wiring (including this adapter with its small-gauge wire) and start a fire.
If nothing else, the homeowner should affix a prominent red tag to the socket, reading "LED lights only!" as a reminder to anyone who might inadvertently use it for something else.
Speaking of TV & movies - film projectors typically have a frame rate of 24 fps, which is chopped to give a 48Hz repetition. This is right at the eye's threshold when the brightness is low, but unacceptable for higher brightness. Europeans have had to live with 50Hz refresh rates on CRT TVs for a long time - very annoying - even with the phosphor persistence.
Does anyone know if half-wave LED strings are used in Europe, with their 50Hz power? That would be even worse than 60Hz.
In the early days of AC power, the frequency was 25Hz, which would make annoying flicker even with incandescent lights.
Good historical points. There was also 16-2/3 Hz power at one time. The low frequencies were generally used for streetcars. I dont know if this was because they wanted slow-speed induction motors for direct drive, or they just wanted to use thicker iron. Our current 60 Hz grid is largely a relic of poor quality transformer iron in the early days. There was serious discussion in the 1950s of raising the grid frequency to 400 Hz, since modern iron could handle that with no great loss, and it would of course make all the transformers and motors MUCH smaller and cheaper, but replacing the installed equipment would be prohibitively expensive. It's interesting that we're gradually returning to that with the proliferation of VFD motor drives and universal switching power suppllies. It may be that at some point in the future we really will either convert the grid to a higher frequency or to DC, with solid-state converters at the point of service to handle legacy equipment.
One quick note about refresh rates. Commercial movies do run at 24 frames per second, but their duty cycle is very high, so the "flicker" is less noticable. However, the larger the screen (peripheral vision "sees" flicker better) and brighter the movie is, the more noticable it becomes. Flicker doesn't bother me much (although I do notice it), but I find IMAX movies frustrating because I really WANT to enjoy them, but the flicker is just too much.
LED Christmas suffer a double whammy because not only do they have a 60 Hz flash rate, but the duty cycle is awful; for two reasons. The first is, as noted in the post, the BEST duty cycle to expect is 50%. The second is: you don't come anywhere near 50% because the LED's don't turn on until the very peak of the cycle. You may be lucky if you get 10-20% duty cycle.
I likc the idea of a high frequency, current regulated, square wave inverter. Perhaps in time for the next holiday?
Or add a 10-25mfd lytic to the output of the bridge rect. Not so big as to toast the leds but big enough to extend the duty cycle a bit to reduce flicker. Or go with a choke input L-C filter to give you fairly nice DC=RMS output. Better than a switcher and won't fry radio reception nearby. Yes, I still like AM radio.
I agree with D Sherman. In my household I am the only one who deals with electrical Christmas decorations so I was not concerned. Using the slavaged plug from a scrapped string of mini lights will give 3 amp protection for most erroneously pluged in devices. Plugging in a string of incandescent mini lights will not be a problem. I have done this and they work OK.
The key word here is "cheap". You can spend more and get commercial quality LED strings that do include a bridge rectifier. I have seen these used on trees at the local mall. With the bridge rectifier, the flicker is much less noticable because it is at 120Hz and because the duty cycle is much closer to 100%. But I can still see the flicker. Your eyes are more sensitive to flicker in your peripheral vision, so look to the side and you will become more aware of it. Flicker seems to be much less of a problem with fluorescent lights because of the phosphor persistence.
I have been able to demonstrate the flicker by taking pictures of the bulbs with my digital camera while panning the camera. I can then easily see the on/off pattern of the streaks of light. Half-wave lights have less than a 50% duty cycle. The full-wave lights look like about 80% or 90%.
The high power LED fixtures I've seen at the local grocery store and gas station don't have any visible flicker, because they apparently use filtered DC and high frequency switching.
You can also see flicker on the LED tail-lights on many cars if you scan your eyes across them.
Next question - how to get rid of the flicker on your neighbors' outdoor lights.
I believe it's because the tail-lights are lit with 100% duty cycle for the higher brightness stop-light mode. In tail-light mode, they run on a low duty cycle pulse to reduce the brightness. It's at a low enough frequency that the strobe effect can be seen. But probably higher than 120Hz.
I recall reading that when the first Tesla hydro plants were built at Niagra Falls, they produced 25Hz. When the river bank eroded, the power plant on the US side fell into the river. The new plant was built for 60Hz.
Last I heard, the Canadian side was still running at 25Hz, but it's used only to supply power to electric arc furnaces for aluminum refining.
Do these light strings (unmodified) only operate on one half-cycle of AC? Is there a series diode to protect from the negative swing, or do they just rely on the reverse voltage blocking of the LEDs in series?
Another thing I've never understood: why don't people use a series capacitor to limit current in AC-powered LEDs? It would eliminate power dissipation of a resistor. Vulnerability to transients, maybe?
The final showdown is under way in our first-ever Gadget Freak of the Year contest. Who will win an all-expenses-paid trip to the Pacific Design & Manufacturing Show? It's up to you, dear readers, to tell us.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.