Design of LED light strings

[statsman]

Backstory: After Thanksgiving, I pulled out of storage several LED C6 light strings that we purchased last year and displayed through December. Not surprisingly, a couple of the strings had lights out. After watching a few YT videos, I used a non-contract voltage checker to isolate the bad bulbs. On one string, a bulb needed to be re-seated. On the other string, two bulbs needed to be replaced. Those strings are working fine at the moment.

Now to the main reason for my post (I apologize if this gets a bit long). All of the C6 light strings we own, which includes the above, all have the same five colors: Red, Blue, Yellow, Orange, and Green. Based on what I have read, the Red, Yellow, and Orange LED diodes run at 2-2.2 volts, while the Green and Blue (also White) LED diodes run at 3-3.3 volts. These are usually in the 20 mA (0.02A) range.

I understand that all of the LED lights on most strings are not wired together in the same series. Usually the string is split up into two or more series wired in parallel. Which is why when a bulb goes out on an LED light string, only a portion of the string will go out.

But what if the LED lights in a series do not total up to 120 volts? I get that the addition of one or more resistors can compensate for this. I have seen resistors attached to the LED diode lead on each individual lamp. Usually, that isn't the case. I have seen some strings with what looks like an inline tube attached to the electrical wiring that I suspect are any resistors (I don't have any string like this - just what I've seen in the older videos).

So, if the LED lights in a series do not total close to 120 volts, if there are no resistors on the individual LED diodes, and there isn't an inline tube on the electrical wires which may provide resistance, how are the strings designed to handle it? And beyond that, is there any way to test a string to determine what voltage is being passed into each series of LED lights?

For anyone who technically understands this and can explain the concept further, I would greatly appreciate it.

For everyone else, I apologize for the five minutes of your life you don't get back.

 

[ShokWaveRider]

Each LED is ~2.1v. @~20ma. at full power. So divide 110/~2.1 = roughly a 50 light string. Most LEDs can take a little more power and can run up to about 30ma., so a couple of shorted LEDs would not make that much difference.

Simply divide 110 by the number of lights to get the ~Voltage of each Light.

This is a rough calculation as LEDs are Diodes and each acts as a half wave rectifier.

 

[Fermion]

I thought this thread might be about Pixel LEDs, which my wife has gotten us into.

If you haven't heard of these, they are addressable LEDs, where each RGB LED actually has a microcontroller in it that listens to a serial data line, drives the three color channels of the LED (8 bit each for 24 bit color) and then sends the data out to the next LED. Thus the whole thing is just three wires for up to 1000 LEDs per string, all individually controllable.

The things you can do with this are pretty amazing and unlimited. Think Griswold on steroids.

 

[Sunset]

When I put out our C9 LEDs.
I had 2 strings go out and 3 strings keep working.
I couldn't figure out the issue, did try the fuse replacement.

It was the entire 2 strings , so I replaced the 2 strings with a nicer string 4" between bulbs instead of 6" for $10.

I did save the bulb covers and bulbs and fuses as I still have 3 original strings.

 

[AZ2020]

@statsman, I also fixed a few LED strings recently, so I have some familiarity with your LED questions.

While I'm not qualified to explain how the LED string works, I was able to replace the individual burned out LEDs with some that I bought online. When removing the old LEDs from the holders I found that some had resistors soldered to one leg of the LED. Not all of them, and it seemed like there were a lot of different resistor values.

@ShokWaveRider is correct in that a few burned out (shorted) LEDs will not make much of a difference. But the more that short out the higher the voltage across the remaining LEDs. Eventually they will all burn out, each one quicker than the last. If an individual LED is burned out open (instead of shorted), the string will not light. Fortunately, most LEDs burn out shorted.

The voltage across all the LEDs in the string must equal 120. There must be resistance somewhere.

 

[statsman]

Each LED is ~2.1v. @~20ma. at full power. So divide 110/~2.1 = roughly a 50 light string. Most LEDs can take a little more power and can run up to about 30ma., so a couple of shorted LEDs would not make that much difference.

Simply divide 110 by the number of lights to get the ~Voltage of each Light.

This is a rough calculation as LEDs are Diodes and each acts as a half wave rectifier.

In the case of the LED light strings in question, the LED diodes are 20mA. The red, yellow, and orange lights are ~2V, while the blue and green are ~3V.

With *this* year's purchases (Philips C6 LED), the 100 light strings are divided into two series of 50 lights each. Within those 50 lights, 30 are at 2V each, while the remaining 20 are at 3V each. This totals ~120V forward voltage. It might explain why I am not seeing any resistors anywhere. I have opened up a few bulbs, and none of them have resistors attached that I can see.

Note that I said *this* year's purchases. Last year, I purchased what I thought was the same Philips C6 LED light set. The retail boxes all say the same thing. Same descriptions, same UPC, same copyright date. It was one reason why we purchased more of the Philips lights (in addition to liking the 5 colors on this specific model - the newer model, having 6 colors with the addition of purple, have colors that are too washed out).

Anyway, the older sets are wired in three series, not two, so there's 33-34 lights in each series. This works out to about 80V (+/- a few volts) forward voltage. It is also why I think I saw some of the LED diodes with resistors soldered on the + lead. So, it appears the newer bulbs are likely not compatible with the older sets, despite being the exact same product (by UPC and description).

If there aren't any resistors with the newer sets, then bulb replacement could be a lot easier. LED diodes can be purchased with the same specifications: 20mA, 3mm diameter LED, 2V (red, yellow, orange) or 3V (blue, green). But I may be making a big assumption here, and it's part of what I am trying to get clarified and learn.

 

[ERD50]

In the case of the LED light strings in question, the LED diodes are 20mA. The red, yellow, and orange lights are ~2V, while the blue and green are ~3V.

With *this* year's purchases (Philips C6 LED), the 100 light strings are divided into two series of 50 lights each. Within those 50 lights, 30 are at 2V each, while the remaining 20 are at 3V each. This totals ~120V forward voltage. It might explain why I am not seeing any resistors anywhere. I have opened up a few bulbs, and none of them have resistors attached that I can see. ...

I'd expect a resistor or capcitor somewhere in the line, it might be molded right into a case somewhere. You may need to measure from the plug to the first or last LED in a string to find it.

The reason I expect a resistor or capacitor is that the ~ 20mA current @ 120V with the 30@2V and 20 @3V (adding up to 120V) you mention is at the 120V RMS value of the AC line. The 120V RMS will have peaks of ~ 170V (1.414 * 120V). That would also mean ~ 2.8V pk and 4.2V pk on those LEDS (assuming the voltage was divided evenly, but I think the 3V LEDs would see more of it due to the non-linear curves, but close enough).

If you look at this chart, they show an LED @ 20mA would have a 2.7V drop, and @ 3V the current is ~ 135mA. The chart stops at 3.1V and 200mA. I'd imagine that 4.2V, even for the lower % of the curve, would blow the LED, that current is going to really jump.

And this data sheet for a 3mm red LED list absolute max current @ 155mA

https://cdn-shop.adafruit.com/datasheets/WP7113SRD-D.pdf

...

Note that I said *this* year's purchases. Last year, I purchased what I thought was the same Philips C6 LED light set. The retail boxes all say the same thing. Same descriptions, same UPC, same copyright date. It was one reason why we purchased more of the Philips lights ....

I sometimes try to buy the same product later, with the hopes that it matches (for LEDs, I want the same color temperature/look for 'warm white'). But the mfg can vary these, or they just use whatever stock is on-hand or cheap at that time.

-ERD50

 

[statsman]

I sometimes try to buy the same product later, with the hopes that it matches (for LEDs, I want the same color temperature/look for 'warm white'). But the mfg can vary these, or they just use whatever stock is on-hand or cheap at that time.

Yeah. The fine print on the retail box indicates "Product may vary from image shown due to continual improvements." I'll keep a couple of last year's strings as spares. I'll do the same with this year's strings. I should also note they changed how the bulb bases lock into the bulb holders. That alone made them incompatible with each other.

DW and I really do like the colors with this set, and the design switch from three series to two series for the 100 C6 lights hasn't changed their look. Many multi-color LED lights have washed out colors. Blues that look like turquoise or cyan. Reds edging toward pink. Greens that look more like lime. The sets we have definitely have richer colors. They are a match to the C9 strings we have.

Thank you for the technical information. I understand some of it. I will try your suggestion to measure the string to see if I can determine where there might be a resistor. I'm pretty certain they aren't molded into a case, although I guess they could be buried inside the plug. The old strings had resistors soldered directly to the LED diode leads. I think they moved away from that as that is where I noticed the physical bulb failures with last year's strings. The leads on the LED diodes are not very durable.

 

[ERD50]

... Many multi-color LED lights have washed out colors. Blues that look like turquoise or cyan. Reds edging toward pink. Greens that look more like lime. The sets we have definitely have richer colors. They are a match to the C9 strings we have.
... .

For more than anyone should ever want to know about these lights, do a search on "youtube technology connections holiday lights".

As he points out, what mfgs really should be doing is using all white LEDS, and providing little translucent plastic caps to provide the color the customer wants.

The red/blue/green/yellow LEDs are monochromatic, they produce a very narrow spectrum of just that color. If you put a translucent cover/paint on it, it will only make it dimmer - you can't shift the color, there is no other color there.

The white LEDs are typically UV LED which excites a white phosphor (similar to fluorescent tubes), with maybe a blue LED to balance the color. The phosphor provides a wide-spectrum white light that *can* be filtered.

I prefer the 'warmer' look of traditional incandescent with cover/paint. The regular color LED
are too stark for my tastes, but they look pretty cool if you want that modern bright look. Personal preference, or really just different looks for different applications.

... . The old strings had resistors soldered directly to the LED diode leads. I think they moved away from that as that is where I noticed the physical bulb failures with last year's strings. The leads on the LED diodes are not very durable.

Soldering a resistor on every LED just doesn't sound like a good solution for mass production, and as you note, would be a reliability concern. There are better ways to do it (an R integrated into the LED itself, or one series R or C for the string), and I'm not surprised they went away from individual Rs soldered to the LED.

-ERD50

 

[statsman]

For more than anyone should ever want to know about these lights, do a search on "youtube technology connections holiday lights".

As he points out, what mfgs really should be doing is using all white LEDS, and providing little translucent plastic caps to provide the color the customer wants.

I watched a few of his videos about this. Paints, dyes, Sharpies, nail polish. He certainly is on a mission.

I do get his point. I'm not a fan of harsh light. Bright can be fine for outdoor lighting, especially when having to counter a street light nearby. I just don't like it when the colors aren't close to what you expect them to be. Indoors, warm white is easier to manage.