This post goes way over the TL;DR line. If you want to get to my summary on buying LED Christmas lights that will last a while, go to the Conclusions section in red, below.
Source: Except where noted, images in this post are from the Gencraft.com AI with a prompt of “Christmas lights”.
Intro: The ghost of Christmas lights past.
My parents had the same sets of Christmas tree lights for my entire childhood. And then some, given that I was the youngest of four children.
I, by contrast, am getting ready to toss (recycle) yet another couple of strings of dead Christmas lights. In this case, some elderly miniature incandescent light strings that started off the season dead. Again. And for which I am finally throwing in the towel.
I am hardly alone in this. Based on what I’ve been finding on-line, many consumers simply expect to buy new lights every year or two. Locally, Fairfax County maintains dedicated bins at the landfill, solely for recycling dead Christmas lights, as distinct from other scrap electrical wire.
Do not put wire in with the household recyclables. To recycle electrical wire in Fairfax Count, VA, you need to take a trip to the County solid waste transfer station, and drop your scrap wire off at the electronics recycling station there. At that point, as I understand it, U.S. recyclers mechanically separate the copper from the insulation by grinding the wires into tiny fragments, and taking whatever pure copper separates from the insulation after grinding. This article from Recycling Today says that wires as thin as those found in telephone cable can recycled. As telephone cable is typically 22 or 24 gauge copper wire, that implies that Christmas lights (typically 22-gauge or heavier) can be recycled. The same article notes that copper recovery ranges from 88% to 98%, depending on the type and size of wire being recycled. That article also makes it clear that separating the wire by type helps optimize the recovery. In that light, it is no surprise that the County separates Christmas lights (typically with thin 22-gauge wire) from other scrap wire.
In effect, disposable Christmas lights are now normal, at least for low-end lights such as those found in big-box stores.
The main point of this post is to get a straight answer to a simple question:
How long can I expect a string of miniature LED Christmas lights to last?
Near as I can tell, the answer has almost nothing to do with what’s advertised on the box, typically the 20,000 hour average life of an individual LED. Instead, life expectancy is all about fragility: thin 22-gauge copper wires, cheesy construction, outdoor use (particularly with replaceable mini bulbs), and the owner’s willingness and ability to repair breaks.
In short, YMMV. Might last a week. Might last decades. Near as I can tell, from the standpoint of longevity, what you’re buying is the wiring. Or, more generally, the materials and construction of the light string. The LED “bulbs” themselves more-or-less come along for the ride.
1: The sockets stop working, not the bulbs. An illustration of the issue.
The well-respected New York Times Wirecutter chose a GE 150-light LED string as their best Christmas lights this year (reference). Among other bits of praise, they say (emphasis mine):
The bulbs should last for at least 10 holiday seasons, and you can replace individual bulbs if they go out (or you can just leave those; the rest of the strand will stay lit)
Buy if you follow the link in the article to the product listing on the Lowe’s website, you find a large number of complaints in the user ratings section. Almost one-quarter of purchasers gave these lights a one-star rating. Most of those low ratings were for early failure. One commenter got it spot-on (emphasis mine):
I also observe, from the Lowe’s comments, that the preponderance of individuals who hated these lights stated that they were used outside, including year-round use outside. To me, this strongly suggest that weather, and water intrusion, were the primary reasons for rapid failure of these lights. More on that below.
2: 100,000 hours just ain’t what it used to be. Why stated LED lifetime does not matter for Christmas lights.
I used to know what the expected lifetime of a light bulb meant. But I don’t any more. This, despite buying products such as Christmas lights that prominently feature the LED lifetime on the package.
And it’s a fair bet that you don’t know what it means, either, for LEDs.
Back when old-fashioned incandescent and fluorescent bulbs were the norm, the expected lifetime of a bulb was pretty simple. The manufacturer plugged in a random sample of 100 bulbs, turned them on, flicked them on-and-off on a standard schedule, and waited for half the bulbs to burn out. The time it took for half the bulbs to burn out was the expected lifetime of the bulb.
YMMV. Obviously, this was under controlled conditions. The bulbs were indoors, they weren’t shaken, the voltage was correct, they were mounted base-down (important for compact fluorescents), and so on. But, given those limits, the definition of expected bulb life was pretty straightforward.
Now ponder this: There are about 9000 hours in a year. So an LED or LED fixture that claims to be rated for 100,000 hour expected life — how, exactly, did they determine that? Did they plug the thing in and wait eleven years before they marketed it? I don’t think so.
Worse, a non-defective LED typically doesn’t burn out, it just fades away. It produces less and less light, sometimes accompanied by a shift in the color of the light, as it ages. For the typical LED, if you are waiting for it to “burn out”, be prepared for an extremely long wait.
As a result, the expected lifetime of an LED fixture (or an individual LED) quoted for marketing materials is defined as the “L70” point: The point at which the light output is projected to fall below 70 percent of the original rated value. By agreement, that’s the point at which most people will perceive that the LED light fixture or light has gotten noticeably dimmer.
Long-time readers my (or may not) recall that the same 70 percent figure is also the threshold for a “just noticeable” difference in sound intensity (Post dated 3/17/2019). There, a five-decibel reduction in noise (29% reduction in noise, as perceived) was the threshold beyond which individuals would typically notice that a source of noise had diminished. I’m fairly sure that 30% reduction being “just noticeable”, for both sound and light intensity, is rooted in human physiology.
The “projected to” phrase above is how manufacturers avoid having to leave these plugged in for a decade before they can rate them and sell them. They take the decline in light output over the first few thousand hours of operation, and extrapolate that out to a 30 percent decline, based on typical observed LED performance.
Worse still, as far as I can tell, the “L70” rating for an individual LED — the lifetime quoted on the box — is based on non-defective LEDs. There’s a separate rating for the fraction of the batch that failed, in some sense (that’s part of the “B” rating for LEDs, or acceptable failure rate, described in this reference). But the separate figure on an LED’s acceptable failure rate or “B” rating is never quoted in consumer materials. You’ll never find that listed on your box of Christmas lights.
Finally, and on the positive side, most sources agree that when LEDs do “burn out” they typically short-circuit. This is good, from the perspective of Christmas light strings, because those short-circuited LEDs continue to pass electrical current along to the rest of the string. (As opposed to incandescent bulbs which, absent fail-safe “shunts” described below, create an open circuit when they burn out and so turn off the entire string on which they reside). The upshot is that if an LED in a light string burns out prematurely, the most likely consequence is that the remaining LEDs just burn a bit brighter, as the resistance of the dead LED is lost when it short-circuits.
Let me now summarize why the “20,000 hour LED” phrase commonly found on big-box LED light strings is essentially irrelevant:
- That’s a ludicrously long time. If “a Christmas season” is 12 hours/day for four weeks, that’s roughly 60 Christmas seasons. (20000/(4*7*12) = ~59.5).
- That’s not even when the typical LED burns out, that’s just when it gets noticeably dim compared to its original light output.
- That doesn’t tell you anything about the fraction of LEDs that are likely to be defective and so burn out prematurely.
- When they burn out, you can change the bulb on most (but by no means all) strings currently offered to consumers.
- If they burn out, they typically don’t extinguish the rest of the string that they are on. Typical failure mode for an LED is short-circuit, which means that current continues to flow through the dead LED and on to the rest of the string.
3: Outdoor use, “commercial-grade” light strings, and sealed sockets with non-replaceable bulbs.
For incandescent lights — including mini-incandescent light strings — the short lifetime of the bulbs means that you have to be able to replace them. Thus, traditional incandescent Christmas light strings had to have sockets and replaceable bulbs. Just like, say, a table lamp.
With LEDs, by contrast, it’s not clear that replaceable bulbs are necessary, or even that they are a good idea. That’s particularly true for light strings used outdoors, and even more so for mini-lights, where scraping any corrosion or oxidation off the contacts of a socket is challenging, to say the least.
And so, modern “commercial grade” LED holiday light strings — those marketed mainly to contractors who are paid to decorate houses and other buildings — are characterized by two things: Thicker wires (almost always), and sealed (non-replaceable) bulb sockets (usually).
Like so, from 1000 Bulbs.com.
Note this explanation of what makes LED Christmas lights commercial grade, from Christmaslightsetc.com.
That said, sealed sockets and non-removable LED bulbs do not themselves guarantee that a light set is robust. Ultimately, if you want long-lasting lights, you probably have to shop for lights marketed as commercial grade, sold to people who are in the business of doing holiday lighting.
4: So why do they even have Christmas LED mini-lights with replaceable bulbs?
I can only speculate, because nobody seems to address this on-line.
Consumer acceptance: Speaking for myself, when I bought some LED mini-lights a few years ago, I rejected any that didn’t have replaceable bulbs. My thinking was that a string with non-replaceable bulbs was made to be thrown away. Accordingly, I associated non-replaceable bulbs with cheap construction. I didn’t want to buy what had to be, in the long run, a disposable set of lights. Which, per the discussion of “commercial grade” just above, is simply dead wrong for LED light strings.
To me, it was like buying a car where you couldn’t replace the tires. I didn’t really care how long they claimed the tires would last, I just couldn’t accept the basic idea.
Insurance against defective bulbs: A likely second reason is to avoid consumer returns. If one LED in 1000 is defective (say), and if consumers won’t tolerate dead LEDs on their light strings, that’s going to result in one-in-ten 100-light strings returned.
Acceptable for indoor use. Finally, my guess is that strings with replaceable mini-LEDs don’t significantly shorten the longevity of light strings used inside. Absent water intrusion, the only degradation of the electrical contacts comes from the slow oxidation of the copper connections exposed to the air. Even in a low-voltage situation like the contacts in series-mounted LED strings, that’s something that ought to take decades to cause problems. Likely those light strings would have been tossed, for other problems, long before then.
5: Om. Ohm. Om. A brief meditation on the wiring of Christmas light strings.
Or, why your light string may look like a tangled mess, but underneath it’s straight-up math.
Source: Energy.gov. NOTE that the illustration figure of 50 lights refers to mini-incandescent lights only, not LEDs.
Mini-incandescent light string, an ohmic device.
As I was preparing to throw away two elderly non-working light strings this year, I realized that I didn’t actually understand how light strings worked. To me, they had always been just a mass of twisted green wires, with no particular rhyme or reason to them.
Plug them in, they make pretty light. Or not, as the case may be.
Moreover, I figured there must be something special going on, because (as advertised), these old incandescent mini-lights would continue to stay lit even if one bulb was burned out. (But not two.) How they could do that seemed like a completely mystery, because as far as I could tell, every socket had just two wires attached to it. How the string could stay lit with a burned-out bulb was beyond my comprehension.
Turns out, the “stays lit with one bad bulb” was a red herring. That has nothing to do with the wiring. As explained in this energy.gov web page, Christmas mini-incandescent lights contain both a filament (like a regular light bulb), and a shunt. The shunt serves to conduct electricity when the filament is broken. I assume that it’s a fairly high-resistance path for the electricity, and that’s why a light string can tolerate one dead bulb (one shunt in use), but not two. With two, the combined resistance of two shunts in the circuit must be enough to keep the filaments in the remaining bulbs from glowing noticeably.
Everything else about this old mini-incandescent light string was dictated by Ohm’s Law.
- Mini-incandescent bulbs are made to operate at 2.4 volts.
- But they plug into a 120-volt socket? How?
- That means you have to wire (120/2.4 = ) 50 of them in series, so that each bulb only “sees” a 2.4 volt differential between its two closest neighbors.
- Which is why mini-incandescent light strings always come in multiples of 50.
- Light strings with more than 50 lights are actually separate strings of 50, with those strings wired to each other in parallel (with each string seeing the full 120 volts coming out of the wall socket).
- My 150-light string had three different strands of 50 each, intertwined, so that every third bulb was on the same string.
- This meant that every part of the string (except the last two bulbs) had four wires — one hot wire running from socket-to-socket for each of the three 50-bulb-series, and a ground wire running end-to-end to complete the circuit back to the wall.
- If this had been set up so that you could plug them together end-to-end, there would have been a 5th wire, running from one end to the other to take the “hot” side of the wall plug down the socket at the end of the string. Thus, strings that can be plugged in end-to-end effectively combine both a light string, and an extension cord. That fifth wire is the “hot” wire of the extension cord.
As for energy use, you can’t infer that from a simple count of bulbs. (Any time you string together 50 ohmic devices of equal electrical resistance in series, each one will “see” 2.4 volts when plugged into a 120-volt circuit.) You have to look that up, to find that each mini-incandescent consumes about 0.45 watts. Thus the entire string of 150 consumes about (150 * 0.45 =~) 68 watts.(Or thereabouts. I don’t think those bulbs are made with any great precision). Throw in a little overhead for the wires and the control box, and that string of 150 mini-incandescent lights consumed about 75 watts. (Less time off for blinking, which does, in fact, reduce the wattage consumed.)
The only thing that didn’t check out was the measured resistance of the bulbs. Again by Ohm’s law, a 0.45 watt bulb operating at 2.4 volts should have an internal resistance of about 13 ohms. (Watts = Volts x Amps implies that each bulb was passing about 0.17 amps of current, when operated at 2.4 volts and 0.45 watts. And Voltage/Current = Resistance implies that each bulb must have about 13 ohms of resistance, to pass 0.17 amps at 2.4 volts. Instead, the bulbs consistently measured about 3.5 ohms. That’s because the resistance of the filament increases dramatically as it heats up, and I was measuring them cold. And, sure enough, the measurements on this page suggest that the resistance more-than-triples as a typical incandescent light bulb filament heats up. Which would be spot-on for a cold 3.5 ohm bulb generating around 13 ohms of resistance when hot.
Back in the day, these mini-incandescent light strings seemed like a big energy savers compared to the alternative. Old-fashioned large incandescents — so-called C7 or C9 bulbs — consume an average of about 5 watts each. So a string of 25 C7 bulbs — pretty much the shortest string available — would have required about 125 watts.
But that was really an illusion. In reality, to a close approximation, all standard (non-halogen) incandescent bulbs are equally efficient. My 150-light string used about half as much electricity, and put out about half as much light, as a 25-light C7 incandescent string. Ultimately, the efficiency (if any) was from the form factor. I could cover a tree with lights, using less electricity, by using more, smaller lights (mini-incandescent) as opposed to fewer, larger lights (C7).
Finally, note that this discussion of lights in series, and strings coming in increments of 50 bulbs, only applies to mini-incandescents. What we now view as “traditional” C7 or C9 bulbs — the bigger incandescents, essentially identical to night-light bulbs — take a full 120 volts, and are wired in parallel. That is, each socket has two wires (hot and neutral) coming in and leaving, the filament of the bulb spans those two wires, and each bulb burns independently of all the others. And if one bulb goes out, all that happens is that one bulb goes out.
Mini-LED light, a non-ohmic device.
Unsurprisingly, LEDs are more efficient than incandescent lights, even when it comes to Christmas lights. Based on what I just measured for an old string of 100 mini-LEDs, they consume about 13% as much electricity as mini-incandescents, for lights that appear to be just about as bright. That’s a larger difference than you see for full-sized light bulbs, which I attribute to the high heat losses of the mini-incandescents, compared to full sized incandescents, due to the higher bulb surface area per length of filament. That said, with a modest number of lights, the total energy consumption is trivial.
If the holiday light season in my household is about two weeks, and I ran two such strings (300 mini-incandescent bulbs) 12 hours a day, that would consume 23 KWH of electricity. Or enough to electricity to drive my wife’s PHEV car for about 125 miles of errands around town. At Virginia’s generating mix, this would create about 15 pounds of C02 emissions, less than would be generate by burning a single gallon of gasoline. By contrast, three strings of elderly 100-light LEDs would clock in around 3 KWH for the Christmas season, or about 13% of the electricity use of the mini-incandescents.
The wiring of mini-LED Christmas lights is both as simple, and not as simple, as that of mini-incandescents.
To be clear, I’m only talking about Christmas lights, on wires, that plug directly into 120 volts. This excludes lights that require some sort of controller or transformer to step down the voltage. Many types of LED tapes, for instance, are mostly wired in parallel, not series, with two low-voltage wires running the length of the tape.
It is as simple, because as with the mini-incandescents, the LEDs end up wired in series. And, to make a string with lots of lights, several such series-wired strings are combined in parallel. All that is identical in concept to mini-incandescents.
But beyond that, it gets weird.
First, LEDs are non-ohmic devices. They show the same voltage drop, across the LED, no matter what voltage is applied to them.
Second, different colors of LEDs are made from different materials and show different voltage drops. I think that this, along with cost, is why fixed-color LED lights typically don’t use colored LEDs, but instead put a colored plastic cover over a white LED.
But mostly, LED strings appear to come in increments of 50, just like mini-incandescents. But there’s no good reason for them to do so, other than convention. The voltage drop across a typical white LED is somewhere around 3.3 volts. The “natural” number of LEDs to wire in series, for 120 volt circuits, is therefore somewhere around (120/3.3 =~) 36. In fact, while my string of 100 LED mini-lights looks like it’s 100 LEDs wired in series, that cannot possibly be true — the combined voltage drop of 100 white LEDs far exceeds the available 120 volts. Instead, if I look carefully, socket by socket, two sockets (about one-third of the way down the string, and two-thirds the way) have three wires going into them. What looks like 100 LEDs in series is actually three strings of 33/34 LEDs in series, with the three strings themselves wired together in parallel.
Thus, while my 150-light mini-incandescent was three strings of 50 bulbs wired in series, my 100-light mini-led is three strings of about 33 LEDs wired in series. In both cases, those separate strings are then connected in parallel, as in the diagram from energy.gov shown earlier in this posting.
This is why you can routinely find comments that “half the lights on the string went out”. Something happened to one of those series strings, while the other strings stayed intact.
Current-limiting resistor? Typically, LED circuits will also require a “current-limiting resistor”, which is just an ordinary electrical resistor whose purpose in the circuit is to limit the current flowing through the LEDs. LEDs typically don’t want to see any more than about 20 milliamps. Here, the total device consumes 6 watts, each of three LED strings consumes about 2 watts, and therefore the current passing through those LEDs is about (2 watts/120 volts =~) 0.016 amps, or about 16 milliamps. Whether the manufacturer manages that just by including the right number of LEDs in the string, or whether it requires a small current-limiting resistor somewhere, I cannot tell. If they can stack up just the right number, with just the right voltage drop (“forward voltage), they would not need one (see this reference for the current-limiting resistor calculation).
But the bottom line is that, like the mini-incandescent string, the mini-LED string is completely easy to understand. Ultimately, the determinants of the LED string boil down to simple arithmetic.
Modern C7 and C9 strings: None of the above.
By contrast, modern C7 and C9 lights all operate on 120 volts, and all modern light strings for these bulbs are wired in parallel. That is, two 120-volt wires run the length of the string, and each bulb bridges those wires. If the bulbs are incandescent, they are designed so that the hot filament has a high enough resistance to limit the current so that the power draw is only about 5 to 7 watts. If they are LEDs, each bulb will have its own built-in current-limiting resistor, which serves to throttle down the current (drop the voltage) so that you don’t instantly fry the LEDs that are directly connected to 120V wires.
Unlike series-connected incandescent or LED lights, parallel-connected lights can contain any arbitrary number of bulbs (subject to not exceeding the amperage rating of the wires), and each light is completely independent of all others on the string.
Conclusion: My takeaways for robust LED Christmas lights.
1: If you are using mini-LED lights outside, you’re probably better off getting strings with non-replaceable bulbs, so-called sealed units. The reason is that, when exposed to weather, the sockets themselves fail far faster than the LED bulbs typically will. A failed socket can break the series connection and so turn off the entire string of LEDs (or a good chunk of it), until you fix it. And you may need some basic electrical skills to do that.
2: If you are using mini-LED lights inside, I don’t think that sealed, non-replaceable bulbs are totally necessary. Instead, replaceable bulbs give you insurance against getting a string with a defective LED in it.
3: In theory, one or more burnt-out LEDs in a string of mini-lights should not be a problem. They should, in theory, burn out by short-circuiting. All that does is reduce the overall resistance (voltage drop) of the string of LEDs, allowing more current to flow through the remaining bulbs. They will then burn somewhat brighter, and will reach their (far-away) failure point somewhat faster because of it. But a large number of burnt-out LEDs, in a series-wired LED string, may allow enough current to pass that you might get rapid failure of the string.
4: Similarly, the easiest way to fix a broken socket is simply to remove it from the string and splice the wires together. The remaining LEDs will then burn slightly more brightly, and have a slightly reduced lifetime.
5: Not discussed here, but by far the easiest way to find which LED socket (or possibly, which LED bulb) is the culprit in a dark string of LEDs is to use a non-contact voltage probe that is sensitive to low (12 volt) AC voltages. You can buy one on Amazon for under $10. That would allow you to find dead bulbs as shown in this Instructables. Plug the string in, separate the wires going to the sockets from the other wires in the string, move the voltage sensor down those socket-connected wires until it stops beeping. That’s where the break is. (There is also a device sold specifically for Christmas lights which provides the same functionality at about three times the price.) That will work on all but the last three or four bulbs in each series-connected string of LEDs. (They won’t work on the last few because the remaining voltage in the wires connecting the LEDs falls below 12 volts, the limit of sensitivity of most such non-contact voltage-sensing devices).
6: Light strings marketed as “commercial grade” and sold to business that install Christmas lights professionally are made to a higher standard. For mini-LED lights, that means 20-gauge wire instead of the lighter 22-gauge found on “homeowner grade” sets, and typically means sealed, non-replaceable LED bulbs instead of replaceable LEDs. When marketed for outdoor use, those sets also typically include waterproof (“coaxial”) power connectors instead of standard plugs and sockets. Those coaxial connectors are often proprietary, and typically you can’t mix-and-match sets across brands if you go that route.
7: But if you want go up a notch in robustness, you probably want to get C7 or C9 (big-bulb) light strings with replaceable bulbs. That’s because these strings typically use much heavier wire (18-gauge — like lamp cord — is common), and replaceable bulbs mean that any defective or burned-out LED bulbs can be swapped for new ones indefinitely.
Opting for those larger C7 bulbs comes at an energy cost (or decorating cost, take your pick). Each bulb uses about 0.6 watts. Where my 100-light string of LED mini-lights used just six watts, that same energy use would get you just ten C7 bulbs. Each bulb would be correspondingly brighter, but you’re still going to need more than ten bulbs to cover a Christmas tree.
8: It’s debatable whether or not you want to take it to the final step and opt for a “commercial grade” C7 LED light string. You probably get more rugged and robust construction with those. But, typically, those are only sold as sealed, weatherproof units with non-replaceable bulbs. Each non-replaceable bulb contains multiple LEDs. Hard failure of any one of those LEDs means you’ll be cutting and patching the string to make it work again, compared to simply replacing the bulb on a string with replaceable bulbs.
Maybe if you intend to use them outside, you’d want the sealed construction. My intended use is indoors, so I think that’s a minor drawback for me.
Finally, if you aren’t going to repair your light strings, and you use them outside, you need to keep your expectations in check. Read this posting — from a firm that installs tens of thousands of light strings outside, every year — to see what you can realistically expect from top-quality LED lights used outside.
Addendum: If you think modern electric Christmas lights are a pain …
Check out oldChristmastreelights.com. There you can see how electric Christmas tree lights have morphed over the century since they were first introduced.
That website is where I figured out that my parent’s lights were fluted C6 15-volt incandescent lamps wired in series (per oldChristmastreelights.com). Light strings of that design were last made in the late 1950s, and the bulbs remained commercially available until the 1970s. These days, the only source of replacement bulbs for those lights would be Ebay or Etsy or similar sellers of used goods.
The entire string of lights would go out when any one bulb burned out.
My dad taught us the efficient ritual of finding the dead bulb by starting at one end with a new bulb and displacing the bulbs, one by one, down the string. Which, weirdly enough, I ended up teaching the same ritual to my daughter, because mini-incandescents are also wired in series. We had to go through that ritual whenever a light string accumulated two dead bulbs or one bulb in which the shunt malfunctioned. I guess the more things change, the more they stay the same.
But that particular ritual is just awkward for modern mini-incandescents, because the bulbs are so hard to remove from their tiny little sockets. At some point, I just didn’t have the hand strength (or the patience) to work my way down the string.
And so, ultimately, what killed my more-than-decade-old mini-incandescent strings wasn’t some un-fixable flaw. It wasn’t a broken wire or a dead controller box. What killed them was the difficulty of maintaining them, coupled with their inherent inefficiency, the recyclability of their copper wire, and the cheapness of LED replacements.
Typical Christmas light designs have been in flux ever since electric Christmas lights were invented. This year, it was just time to move on.
