J.W. Speaker Denouncing Low-CCT White Light?

[JustAnOldFashionedLEDGuy]

HPS lights are not a good comparison, they are generally terrible, make all the colors look the same, cast weird shadows, and aren't as bright as LEDs. They were providing minimal illumination.

I'm not an expert in this by any means, but I drive a variety of cars and motorcycles, and I notice than on very dark asphalt halogen headlights seem to be more effective than HID/LED.

Of course they were as bright as LEDs. They had to be, that was a legal requirement. In many cases they are much brighter due to high min/max ratio forcing higher illumination levels to reach targets, and also to account for depreciation over time and an indeterminate time to replacement. In many cases, with LED, the average lighting levels dropped.

 

[jtr1962]

But I'm sure marketing has played a big role why most automakers went from 4100-4300K to 5000K headlamps.

I don't doubt it but for me that's just tinkering around the margins. I might have liked 5000K streetlights better but I can certainly live with 4300K ones. I'd probably hate 6000K though. Way too blue for my tastes, and they make seeing worse.

To say that warm white LEDs emulate HPS is disingenuous. HPS has a very narrow spectrum, its probably the worst quality light except for LPS, even worse than MV. With LEDs, there are plenty of scenarios where one color temp can have a legit advantage over the other.

I'm not sure what type of LEDs were chosen for that particular project but if it was phosphor converted amber then they're not much better than HPS. I actually liked the old MV lamps. They had a pleasant blue-green hue which was very peaceful at night. It blended in with the moon and stars nicely. Of course, it made people look like corpses but later versions which rendered reds better fixed that somewhat.

 

[JustAnOldFashionedLEDGuy]

Don't you mean CRI, not CCT? In many urban areas, since the streetlights are also lighting the sidewalks and businesses, high CRI might be desirable as you said. On an open highway in the middle of nowhere, probably not.

Sorry, I am multi-tasking and really not giving this much attention. I fixed my post.

A lot of what you are saying JTR is correct, and the other poster (OP), is quoting studies that do not say what the poster thinks they say, and conclusions are being made from the data that frankly the data does not support. The real answer is something in the middle of course, and the answer varies by location, and just because Title 24 is doing it, does not mean it is a good idea, and looking at their justification, they are not making the changes for the right reasons.

 

[och]

Of course they were as bright as LEDs. They had to be, that was a legal requirement. In many cases they are much brighter due to high min/max rations forcing higher illumination levels to reach targets, and also do account for depreciation over time and an indeterminate time to replacement. In many cases, with LED, the average lighting levels dropped.

This is true if you're only looking at the total lumen output, but doesn't tell the whole story. LEDs are directional, so for road lighting application they deliver much more light onto the road vs HPS/HID where a lot of light is lost in the fixture. In reality, most LED street lights are much brighter than HPS they replaced.

 

[JustAnOldFashionedLEDGuy]

This is true if you're only looking at the total lumen output, but doesn't tell the whole story. LEDs are directional, so for road lighting application they deliver much more light onto the road vs HPS/HID where a lot of light is lost in the fixture. In reality, most LED street lights are much brighter than HPS they replaced.

Please don't try to speak "expertly" on things you clearly are not an expert on.

It does not matter what happens in the fixture and what I stated is not the fixture. It is what hits the road. That is the level (target) that you must meet to meet the lighting regulations which invariably were set as legal limits and tend to be followed to avoid lawsuits. HPS levels must be hit at the end of life of the buib, which may have been a 4 year target, but average replacement may have been 3 years meaning often well above required levels. LED fixtures worst case were L70, even even 5+ years ago, many were being run L80, L85, so not much degradation from new to old. As well, the HPS fixtures would often have larger mix/max, and again, depening on the standard being followed, may have forced higher average levels to ensure minimums were met. LED lights with lower degradation from new to replacement, and better min/max ratios resulted in often lower lighting levels on the road at installation, but better over time, and more even, with potentially higher minimums, but similar averages.

Conclusion: Most LED street lights were absolutely not brighter than the HPS they replaced. However, due to the higher blue levels, which is what you eye/brain uses to determine brightness, they appear as such, and when the stray light gets to mesoptic/scotopic levels, the useful/perceived brightness can be much higher.

 

[JustAnOldFashionedLEDGuy]

To me 5000K always made sense as its closest to sunlight. Many CCT preferences are cultural though.

There are a lot of papers on this over the years, IES, NHTSA, Academic, that showed limited value going much over 4000K with an adequate spectrum. On high speed highways, even that is excessive since most of what you are using is central vision / photopic. I agree that cycling presents new challenges that have not often been well addressed in research. Yes, there absolutely have been studies showing more rapid detection of obstacles with higher (absolute levels) of what is effectively scotoptically active light (not necessarily pure blue), in peripheral vision and low light. My solution is using a bloody bright bicycle light on roads (but with a nice beam). I personally find little advantage of 5000K over 4000K in this situation if any. As I mix road and trail, the 4000K is superior off pavement, and I find for even typical urban obstacles like rocks, wood pieces, even critters. Enough light and the potholes are not an issue either.

 

[JustAnOldFashionedLEDGuy]

Of course of note, is that JW Speaker makes headlights, and no, headlights and streetlights are not the same. Streetlights and streetlighting systems are designed to ensure photopic levels at the road surface, and the angle is such that there rarely is a very large angle of reflection. Headlights are different, they vary between high photopic levels up close, to mesoptic and even scotopic in the distance, critical when there are not street lights. There target markets, trucks, and other work vehicles (and jeeps) are often in locations without streetlights where effectively lighting at long distances is critical. In those situations, a higher CCT is going to be superior. You need those scotopic lumens. However, given that current regulations don't recognize CCT (or CRI), or spectrum, they can't just make them brighter at 3000K, so they have 5000K. In urban environments, 4000K would be just fine, but 5000K superior outside urban. Even all these years later, head light standards are still a bit rudimentary, but I lost interest in keeping track, so others can answer more knowledgeably on the latest coming down the pipe in standards. Really, adaptive lighting should allow really useful tailoring of both intensity and CCT/CRI to suit the situation maximizing the experience for all road users.

 

[JustAnOldFashionedLEDGuy]

.... and then self driving cars, non-visual frequencies, and LIDAR will render it all moot

 

[jtr1962]

There are a lot of papers on this over the years, IES, NHTSA, Academic, that showed limited value going much over 4000K with an adequate spectrum. On high speed highways, even that is excessive since most of what you are using is central vision / photopic. I agree that cycling presents new challenges that have not often been well addressed in research. Yes, there absolutely have been studies showing more rapid detection of obstacles with higher (absolute levels) of what is effectively scotoptically active light (not necessarily pure blue), in peripheral vision and low light. My solution is using a bloody bright bicycle light on roads (but with a nice beam). I personally find little advantage of 5000K over 4000K in this situation if any. As I mix road and trail, the 4000K is superior off pavement, and I find for even typical urban obstacles like rocks, wood pieces, even critters. Enough light and the potholes are not an issue either.

Any studies on going with higher CRI? Is it possible it would be advantageous for a cyclist, even if not for a motorist?

In places where you use mostly central vision/photopic, it seems the primary thing to do is simply flood the road with as much light as practical, provided you have an adequate spectrum. This points in the direction of simply using whichever CCT gets the most lumens per watt. And not coincidentally that seems to be in the 4000K to 5000K range for most LEDs.

 

[och]

Please don't try to speak "expertly" on things you clearly are not an expert on.

I may not be an expert, but I don't try to manipulate by painting only part of the picture.

It does not matter what happens in the fixture and what I stated is not the fixture. It is what hits the road. That is the level (target) that you must meet to meet the lighting regulations which invariably were set as legal limits and tend to be followed to avoid lawsuits. HPS levels must be hit at the end of life of the buib, which may have been a 4 year target, but average replacement may have been 3 years meaning often well above required levels. LED fixtures worst case were L70, even even 5+ years ago, many were being run L80, L85, so not much degradation from new to old. As well, the HPS fixtures would often have larger mix/max, and again, depening on the standard being followed, may have forced higher average levels to ensure minimums were met. LED lights with lower degradation from new to replacement, and better min/max ratios resulted in often lower lighting levels on the road at installation, but better over time, and more even, with potentially higher minimums, but similar averages.

It matters very much what happens in the fixture. Of course they designed HPS fixtures to meet the requirement of light that must reach the road, but who is to say that new LED lights do no exceed it? Plus, many of the old HPS fixtures were very poorly maintained, and dirty glass and reflective surface inside the fixture greatly degraded light output. So in reality, LEDs are much brighter.

Conclusion: Most LED street lights were absolutely not brighter than the HPS they replaced. However, due to the higher blue levels, which is what you eye/brain uses to determine brightness, they appear as such, and when the stray light gets to mesoptic/scotopic levels, the useful/perceived brightness can be much higher.

Blue levels have nothing to do with it. I've replaced several 4100K HID wall fixtures in my shop with similar temperature LEDs, and they are much brighter while drawing far less power.

 

[JustAnOldFashionedLEDGuy]

Any studies on going with higher CRI? Is it possible it would be advantageous for a cyclist, even if not for a motorist?

In places where you use mostly central vision/photopic, it seems the primary thing to do is simply flood the road with as much light as practical, provided you have an adequate spectrum. This points in the direction of simply using whichever CCT gets the most lumens per watt. And not coincidentally that seems to be in the 4000K to 5000K range for most LEDs.

I am not aware of any studies that show high CRI is superior for roadway lighting or for headlights. I have a hard time believing a designed spectrum would not be superior to cover the need to acuity over a range of light levels at the same time. HPS was all about the most lumens possible, and it works fine for the most part on highways.

 

[alpg88]

Our parking lot was illuminated by large HPS yellow bulbs, after Sandy, management replaced them with leds, It was a huge mistake, those leds did not even compare to old lights, so much worse, it is a lot darker now than it used to be.
Now my point is not that leds are worse or better than HPS, my point is, real world situation is not as simple as led vs hps, in some cases hps can be a lot better than leds, and vise versa, the type is not as critical as actual execution of the light fixture,

 

[JustAnOldFashionedLEDGuy]

I may not be an expert, but I don't try to manipulate by painting only part of the picture.

It matters very much what happens in the fixture. Of course they designed HPS fixtures to meet the requirement of light that must reach the road, but who is to say that new LED lights do no exceed it? Plus, many of the old HPS fixtures were very poorly maintained, and dirty glass and reflective surface inside the fixture greatly degraded light output. So in reality, LEDs are much brighter.

Blue levels have nothing to do with it. I've replaced several 4100K HID wall fixtures in my shop with similar temperature LEDs, and they are much brighter while drawing far less power.

How many lighting designs for streets and parking lots did you do where you had to put your signature on a document and affix your P. Eng. stamps? How many times did you go out and actually measure lighting levels before and after? I am going to go with a big fat 0.

HPS fixtures were often maintained well as someone was in there replacing a bulb and cleaning them every 3 years, 4 rarely, and sometimes 5 with newer tech bulbs, but they didn't get much uptake. That is not exclusively the case, but was the case in the majority of major urban centers. Street lights being out is also a liability so avoided.

w.r.t your ONE OFF case of your shop, odds are you were maybe 70-75LPW on the MH at the fixture level (those smaller wall fixtures are not that efficient, and if magnetic ballast maybe even less), and probably, you didn't replace the bulbs till absolutely necessary, so they were more like 45-50, but importantly off 40% of more from when new. Then you put in a new LED fixture at 100+ LPW, perhaps with better directed optics, perhaps less uplight. The situations are not remotely the same. You are comparing a designed lighting system to some ad-hoc wall fixtures.

I not only painted the whole picture in my last post, I came up with the idea for the picture and made the paint.

 

[JustAnOldFashionedLEDGuy]

Our parking lot was illuminated by large HPS yellow bulbs, after Sandy, management replaced them with leds, It was a huge mistake, those leds did not even compare to old lights, so much worse, it is a lot darker now than it used to be.
Now my point is not that leds are worse or better than HPS, my point is, real world situation is not as simple as led vs hps, in some cases hps can be a lot better than leds, and vise versa, they type is not as critical as actual execution of the light fixture,

Problem is your management probably hired whatever electrician they normally use, he recommended some cheap lights he makes good margin on, used some stupid 2:1 "rule of thumb" to convert from HPS watts to LED watts, and well now you have a crap lighting system. When the original fixtures were put in for your parking lot, it is almost a given that an engineer or lighting designer did a proper analysis and picked specific fixtures, pole heights, and worked with pole spacing limitations to create a compliant lighting setup.
.
I take that all back if you used to have side of wall wallpacks. They are a pain in the rear when replacing. Many jurisdictions have lighting ordinances and those old wall packs are no compliant. When you replace them, you are supposed to use full cut-off enclosures (no uplight). That again takes someone who knows what they are doing to select a proper fixture.
.
Of course, I also see idiots put up wickedly bright LED fixtures, generally cheap off-shore imports, and mount them on swivel mounts and have them 30+ degrees of horizontal and the end up lighting up the neighborhood. I am currently in the process of giving someone an education about that. I suspect they are going to have to rip out $20,000 of lighting. I don't feel a bit bad though, they cut corners, and their residential neighbors are being punished for it.

 

[och]

How many lighting designs for streets and parking lots did you do where you had to put your signature on a document and affix your P. Eng. stamps? How many times did you go out and actually measure lighting levels before and after? I am going to go with a big fat 0.

HPS fixtures were often maintained well as someone was in there replacing a bulb and cleaning them every 3 years, 4 rarely, and sometimes 5 with newer tech bulbs, but they didn't get much uptake. That is not exclusively the case, but was the case in the majority of major urban centers. Street lights being out is also a liability so avoided.

w.r.t your ONE OFF case of your shop, odds are you were maybe 70-75LPW on the MH at the fixture level (those smaller wall fixtures are not that efficient, and if magnetic ballast maybe even less), and probably, you didn't replace the bulbs till absolutely necessary, so they were more like 45-50, but importantly off 40% of more from when new. Then you put in a new LED fixture at 100+ LPW, perhaps with better directed optics, perhaps less uplight. The situations are not remotely the same. You are comparing a designed lighting system to some ad-hoc wall fixtures.

I not only painted the whole picture in my last post, I came up with the idea for the picture and made the paint.

Except you've assumed half the picture and dismissed the other half. Especially when you claim NYC maintains light fixtures properly, and at the same time you assume I don't maintain my light fixtures right over my work benches where we do precision fabrication.

I'm not a lighting engineer, but I understand the basics, and I can see with my own eyes. Here in NYC the new LED streetlights are so much brighter than old HPS, it's not even a comparison.

 

[JustAnOldFashionedLEDGuy]

Except you've assumed half the picture and dismissed the other half. Especially when you claim NYC maintains light fixtures properly, and at the same time you assume I don't maintain my light fixtures right over my work benches where we do precision fabrication.

I'm not a lighting engineer, but I understand the basics, and I can see with my own eyes. Here in NYC the new LED streetlights are so much brighter than old HPS, it's not even a comparison.

Yes, your "feelings" and ad-hoc one off experience trumps my decades of lighting experience and significant direct experience in the topic we are discussing. Carry on. Flat Earth people trust their own eyes too ......

 

[och]

Yes, your "feelings" and ad-hoc one off experience trumps my decades of lighting experience and significant direct experience in the topic we are discussing. Carry on. Flat Earth people trust their own eyes too ......

Post some specs then.

 

[bykfixer]

You mean the earth isn't flat?

 

[cetary35]

Any studies on going with higher CRI? Is it possible it would be advantageous for a cyclist, even if not for a motorist?

In places where you use mostly central vision/photopic, it seems the primary thing to do is simply flood the road with as much light as practical, provided you have an adequate spectrum. This points in the direction of simply using whichever CCT gets the most lumens per watt. And not coincidentally that seems to be in the 4000K to 5000K range for most LEDs.

To quote Alaric,

Night driving conditions don't use photopic or scotopic vision, rather mesopic vision.
Blue light may let you see some motion in the periphery but that's about it. Remember, blue light just doesn't work well with our optical system.

That higher-CCT light going off the edges of the roadway may be disadvantageous when you look directly into that illuminated field, and I'm not sure that's worth any advantage you MIGHT get seeing some defocused motion off to the side.

 

[cetary35]

These are the relative spectrums in the UMTRI study. To assign any relevance between this study and what would occur with vastly different LED spectrums which may be tailored either indicates total lack of understanding of this topic, or not having read this in enough detail to realize the conclusions being made in this thread w.r.t. the UMTRI study are deeply flawed.

Guess what? UMTRI conducted another study with LED! Here are their findings.

we
predicted that headlamps using LEDs with the chromaticities examined in that study would lead
to more discomfort glare than the HID headlamps, and substantially more discomfort than
tungsten-halogen headlamps. Our recommendation was that, to minimize discomfort glare,
manufacturers should keep the blue content of LED headlamps as low as practicable, given other
considerations.

Ultimately, the same properties are at play. The more blue a light source has, the more glare it will produce. Even though that chart showed a spectrum normalized for a black body, it still is relevant. The chart you linked also illustrates that even providing perfect CRI and SPD while increasing CCT will still increase glare.

But hey, @jtr1962, now that we're throwing out any study since it wasn't done with LEDs, I assume we're also going to be throwing out the two studies you linked in the beginning with high CCT fluorescents?