Disappointing trend in LED manufacturing

[JoakimFlorence]

So all three colors are using the same blue LED chips, and the phosphor converts blue to either red or green. What bugs me about this is that it would be more efficient (and probably a little brighter) to use native red or green LEDs instead of the phosphor-downconverted blue ones.

Theoretically, yes.
But the thing is, due to mass production, blue chips are simply cheaper than red or green chips, and also more efficient. Not that higher efficiency red or green chips do not exist, but I mean more efficient for the same price. For the same price as a blue chip, you are going to get a red or green chip that is half as efficient. What this ends up meaning is that a blue chip coated with phosphor is going to end up having about the same efficiency as red or green LED. Unless the manufacturer wants to shell out lots of money and get the most efficient red and green chips, which are probably going to be several times more expensive due to a lower volume of production than blue chips. Blue chips are used to make white LEDs, which are ubiquitous.

 

[HarryN]

Part of the reason for this transition is that the gases used to produce red / orange and amber direct emitting LEDs are extremely toxic. The companies are under a lot of pressure to remove this from their production sites.

Some companies only have GaN production capability in house. For them to build a high end, red / amber / orange emitter would be very difficult and time consuming vs just using what they are good at making (for example Cree) vs a place like Lumileds.

The PC LEDs share the same general electrical and thermal characteristics as the B / G which makes assembly easier.

Less obvious is that the people who are really good at making these R / O / A direct emitters got their PhDs quite a while back. Many are retired or nearly so. Very few western Universities have the equipment or gas handling capability to train people to produce these materials anymore.

In order to get a PhD in the US or EU, you have to do original work on a subject and it needs to be interesting enough to a government agency to fund it. Since these topics have been studied and documented so heavily in the past, it is difficult to find something novel.

This isn't a problem in the far east as they are perfectly happy to hand out a PhD for "replicating" the work of others, but in the US this is at best Masters level work.

 

[PhotonWrangler]

Part of the reason for this transition is that the gases used to produce red / orange and amber direct emitting LEDs are extremely toxic. The companies are under a lot of pressure to remove this from their production sites.

Harry, that's an angle that I hadn't considered before; thanks. It makes sense to move away from manufacturing techniques that involve toxic materials. What I didn't know was that the manufacturing of blue LEDs was less toxic.

 

[HarryN]

Harry, that's an angle that I hadn't considered before; thanks. It makes sense to move away from manufacturing techniques that involve toxic materials. What I didn't know was that the manufacturing of blue LEDs was less toxic.

It is a relative thing as far as toxicity. The ones used to make direct emitting R / O / A are extremely toxic even when you can't smell them and the ones to make B and G give you a chance to escape.

B / G use a lot of high purity NH3 / ammonia - similar to what is used in some farm applications.

I used to sell the equipment that used these gases to make LEDs. (all types).

The industry has had remarkably few incidents in spite of the very large number of users world wide. I credit HP (now Lumileds) and some related industry pioneers for setting such good examples for how it should be done in a safe manner.

 

[JoakimFlorence]

It's true production of red LEDs involve a more hazardous production process, but I suspect that may not really be the main reason for more red LEDs not being used. (Blue LEDs are based on GaN, whereas red is based on GaAsP)
Rather I suspect there has been much less recent investment-research-development in LED wavelengths other than blue. Everyone wants to focus on better blue LEDs because those are going to make better white LEDs, and there are a lot more white LEDs (and higher power ones) being produced than color LEDs. So again it comes back to scale of production. It probably would be possible to develop cleaner technology to produce red LEDs, but the equivalent resources and motivation to make that happen are just not there.

If it was simply only an issue of environmental concern or hazardous conditions, it would not be much of an issue to simply offsource production to China. Although it is also true that LED production is a very high technology thing and the efficiency of LED chips produced by Chinese companies are much lower than American, German, and Japanese companies (Cree, Osram, Nichia being the big three). (And if I were a Western company, I would not want to hand the specifics of my production process over to a Chinese company because they would steal it - something that has been a huge problem)

There may be other smaller issues as well. I'm really not sure about this but I think red LEDs might be a little more vulnerable to degradation from UV in sunlight, which could be an issue for cars.
I'm sure this is not actually a real factor taken into consideration but also colors produced via phosphor could be perceived as a little more "pleasant" than bare emitters, or a little more desirable in color. Completely monochromatic light can come across as a little "unnatural".
It can also simplify the electronics for wiring the lighting in cars when there are not two different voltages, which I am sure could be another big factor.

 

[JustAnOldFashionedLEDGuy]

Purely a matter of performance AND cost, and the two are tied together.

Phosphor pumped Green LEDs are right now more efficient in all cases. Phosphor pumped RED LEDs are more efficient are operating temperatures if not more efficient near room temperature. That is one aspect of performance.

Second aspect is thermal stability. Phosphor pumped LEDs are very stable w.r.t. output over temperature, both amount and wavelength. Direct RED LEDs are not very stable in output or wavelength.

If you want to make a multi-color bulb or fixture with consistent output without temperature feedback and the best efficiency in all cases (which defines your thermal envelope), then phosphor pumped is much better. Your product will be cheaper and perform better.

As well, some width to the emission can improve color gamut, another plus on the performance side.

 

[jtr1962]

It can also simplify the electronics for wiring the lighting in cars when there are not two different voltages, which I am sure could be another big factor.

That's irrelevant because properly designed LED lighting uses a constant current driver, not constant voltage. The only concession to voltage might be that an LED setup for a car using red LEDs will use 6 in series and a constant current driver, versus 4 blue LEDs in series. The idea is to get the total native voltage of the LEDs close to the supply voltage, in order to get maximum efficiency from the driver. The driver simply ensures the LEDs in the string get the same current whether the car battery is low or fully charged.

 

[Monocrom]

Companies will do anything to save a buck. Especially when it comes to making cheap products.

 

[kaichu dento]

Still trendy to ignore a whole slew of information based posts and call out the 'evil companies' because it's easier than discussing things one doesn't understand.

 

[JoakimFlorence]

Companies will do anything to save a buck. Especially when it comes to making cheap products.

I think this is mainly what's going on.

But I think the fault is mainly the consumer (in this case local governments and city councils). In many of these situations they will just go with the cheapest option, which puts a lot of pressure on companies to have to cut costs as much as possible.
The main hurdle to conversion to LED street lighting was initial cost.
And of course (even more in the U.S. than in Europe) most of these LED streetlights are assembled by companies in China and are not "real" mainstream brands. (China is notorious for cutting corners and making products that will fall apart a short time after purchase)

 

[JustAnOldFashionedLEDGuy]

That's irrelevant because properly designed LED lighting uses a constant current driver, not constant voltage. The only concession to voltage might be that an LED setup for a car using red LEDs will use 6 in series and a constant current driver, versus 4 blue LEDs in series. The idea is to get the total native voltage of the LEDs close to the supply voltage, in order to get maximum efficiency from the driver. The driver simply ensures the LEDs in the string get the same current whether the car battery is low or fully charged.

Resistors is a perfectly appropriate method for the application. A change in brightness for most lights in a car w.r.t. battery voltage is totally acceptable. The change is likely less than with an incandescent bulb.

 

[JustAnOldFashionedLEDGuy]

I think this is mainly what's going on.

But I think the fault is mainly the consumer (in this case local governments and city councils). In many of these situations they will just go with the cheapest option, which puts a lot of pressure on companies to have to cut costs as much as possible.
The main hurdle to conversion to LED street lighting was initial cost.
And of course (even more in the U.S. than in Europe) most of these LED streetlights are assembled by companies in China and are not "real" mainstream brands. (China is notorious for cutting corners and making products that will fall apart a short time after purchase)

NO! Most of the street lights in the US are assembled in North America. Street lights are one area where the penetration of Chinese products is relatively low in lighting due to the high cost of failures. Most installed street lights ARE mainstream brands.

Why do people feel the need to post as facts something they know very little about?

Europe has every bit as much Chinese lighting product as the U.S., including a lot more poorly engineered power supplies.

 

[JustAnOldFashionedLEDGuy]

Still trendy to ignore a whole slew of information based posts and call out the 'evil companies' because it's easier than discussing things one doesn't understand.

I always assume it is a show off / ego thing.

 

[JoakimFlorence]

Resistors is a perfectly appropriate method for the application.

This was discussed in another thread. The short answer is that resistors are not really a satisfactory or practical method for limiting current in LED lighting. The problem is that either the efficiency losses are too high, or that small variations in input voltage can lead to large and intolerable variations in the amount of current going through the LED. You will only find resistors in very small indicator lights where the power level is so small efficiency losses do not matter.

Think of it this way. There is a certain amount of voltage. Wired into a string there is a series of resistors and LEDs. Each LED or resistor "sucks up" a certain amount of voltage. Although of course the LEDs each have resistance, they cannot be relied upon to provide that resistance, because their resistance value suddenly drops with any voltage a tiny bit above their voltage rating. Although a certain voltage might be fed to the LEDs, transient voltage spikes are normal and common from any power supply.
The more of that voltage is being sucked up by the LEDs, the more the circuit is going to behave like just an LED, rather than a normal resistor.
If only 10% of the power was being sucked up by the resistor, that would mean that if there was a transient voltage spike 50% above normal, almost 15 times more current would be allowed through the circuit than normal.
This is intolerable for two reasons. First high power LED emitters are expensive and it might not be very economical to be able to increase current rating of the emitters by a factor of 15. But second of all, if you run an LED on only a small fraction of the current it is rated for, there will typically be a wavelength shift, and in a white LED this could alter the tint of the light, giving a noticeable slight greenish tint, although this could be adjusted for with manufacturing specifications. Even this would not be pragmatic because it would be likely there might be occasional voltage surges in the electric grid. Even though they might only last a fraction of a second, it would result in a burn out of all the LED lamps in the city. For example, static electricity or from a lightning strike. It takes very little energy to cause a burn out of LEDs. You can burn out an LED with a thousand times less energy than it normally uses per second, if enough energy gets concentrated into a very short instant of time. Remember, wattage is energy per second of time. If you were to concentrate the energy of one watt over one second into only a hundredth of a second, that would be 100 Watts, enough to burn out a 10 Watt LED. One millionth of a Watt imparted through the circuit through a static discharge is easily enough to burn out an LED. (Remember again, a Watt is not an actual unit of energy. One Watt over one entire second equals one Joule of energy. A Watt is just the rate of energy flow)

 

[Dave_H]

I agree in general, for high-power LED lighting resistive current limiting is usually not practical. It is generally done with switching constant-current output drivers. Linear constant-current regulators will work albeit not for street lighting. To cope with line variations they would need to be over-engineered and result in poor efficiency.

One case of linear regulation is (typical) 9W LED A19 bulb. If sufficient LEDs are stacked to get vf= 130-140v total (for 120vac line) the full-wave rectified line applied through a 60mA high-voltage regulator will do a reasonable job. One 100W equivalent bulb I opened uses two such chips (CYT1000A) in parallel.

Dave

 

[JustAnOldFashionedLEDGuy]

This was discussed in another thread. The short answer is that resistors are not really a satisfactory or practical method for limiting current in LED lighting. The problem is that either the efficiency losses are too high, or that small variations in input voltage can lead to large and intolerable variations in the amount of current going through the LED. You will only find resistors in very small indicator lights where the power level is so small efficiency losses do not matter.

And you have designed how many professional LED products that have gone into mass production and for automotive? I know the answer to that question is 0.

Let me state again. RESISTORS are a PERFECTLY ACCEPTABLE method of setting current in many applications and automotive is definitely one of those applications since we have a reasonably narrow range of voltages and Vf is relatively well controlled on the LED side now. For many applications, highly accurate LED output is not needed, only meeting some minimum under some conditions. With the exception of forward lighting, resistors are quite common in automotive including for tail lights because they are cheap, reliable, and they do the job.

There are actually automotive lighting and general lighting experts in these threads. You are not one of them. Your high school lessons were not relevant 10 years ago, and they still are not for the majority of people here. This forum is not for you to try to look smart.

 

[JustAnOldFashionedLEDGuy]

Dave_H, linear regulators for interior lighting had a go there for a few years, and even for exterior single voltage applications. The issue was not efficiency, but line rate flicker. Line rate flicker (and flicker in general) has become more of an issue the last 3-5 years. It is not acceptable for commercial lighting, and even large retailer put restrictions on their branded products. You can add capacitance to linear regulation solutions and still hit power factor and THD requirements, but your cost and size is now at the level of a switching supply. With residential, you already needed some added circuitry to keep dimmers happy, so cost savings of the linear quickly disappear. Low cost off-shore lighting is still seen with linear regulation solutions and high flicker.

 

[KITROBASKIN]

Is this quote from Joakim accurate?

Remember, wattage is energy per second of time. If you were to concentrate the energy of one watt over one second into only a hundredth of a second, that would be 100 Watts, enough to burn out a 10 Watt LED. One millionth of a Watt imparted through the circuit through a static discharge is easily enough to burn out an LED. (Remember again, a Watt is not an actual unit of energy. One Watt over one entire second equals one Joule of energy. A Watt is just the rate of energy flow)

 

[radellaf]

I wish the automotive LED engineers would quit using very low frequency PWM for tail lights. Personally would love to have had that outlawed from the beginning. Very distracting to me. I guess not to others.

As for the PC red and green (and yellow and orange), I've seen that on fairy light strings and... am not a huge fan. It's OK. The colors fade (yellow to white mainly) too fast. But as those are made absolutely as cheap as possible, I ain't gonna complain.

Otherwise I mostly see the RGB chip LEDs with all 3 in one package, whether 4-pin analog, or 5V with serial digital control. Ever seen a PC version of those?

 

[PhotonWrangler]

I wish the automotive LED engineers would quit using very low frequency PWM for tail lights. Personally would love to have had that outlawed from the beginning. Very distracting to me. I guess not to others.

It's very distracting to me also. If I'm sitting behind someone with low frequency PWM brake lights at a stop light, any lateral motion of my eyes creates lots of flicker. This practice needs to stop.