Why LEDs should not be run in Halogen reflectors

LEDs produce short wavelength light which human eyes have an extremely hard time focusing on and processing, making them appear much brighter when looking at the light even though they may not actually put out as much light as it would seem. This gives a very short intro to the issue: https://www.assemblymag.com/blogs/14-assembly-blog/post/91320-beware-the-blue-light

You can check out how a handful of leds reverse lights compare to stock here:
https://www.tacomaworld.com/threads/the-921-led-reverse-light-bulb-study.490137/

And for a serious reverse light upgrade:
https://www.tacomaworld.com/threads/gy6-35-hir-921-reverse-light-upgrade-vs-high-power-leds.474996/

 

Its too bad the Philips 35w equal to 50w EcoVantages were discontinued. Luckily I bought four of them just before they were discontinued.

 

You mean the Philips 30w were discontinued. The new Osram 35w are even brighter than the 50w. I still need to build that harness for more direct power.

 

The source of the light (LED, halogen, HID, et al) has zilch to do with any of this. The wavelength IS the color. 5k, 6k, whatever. The human-detectable visual wavelength (being the inverse of the frequency) spectrum is very limited with infrared falling at one end and ultraviolet at the other.

The source of light, whether a 5k HID or a 5k LED is indistinguishable to the human eye. Bluer visual light is harder in the eye. The source of the light makes no difference.

 

Wavelength is color at a single point in the spectrum. Wavelength is not color temperature, like 5000k or 6000k. Color temperature (which is what is referred to in kelvin) accounts for all the wavelengths in a given light source and there are many. High color temperature light definitely refracts more, but it is specifically the short wavelength light that causes the issue.

While the light source may be sourced from many different things, talking about rear facing lights on a vehicle tends to be incandescent/halogen or LED. Automotive LEDs typically have massive spikes in the blue/violet spectrum which are short wavelength.

Here is a spectometer comparison I pulled between Rigid Fog LEDs and 90w Flosser halogen fogs as an example of halogen vs LED lighting profile. No matter how you filter a halogen, it is just not going to have the ultra high intensity spike of the very short wavelength spectrum as the bulk of its output is on the other end of the spectrum, unlike most automotive LEDs that have extreme spikes in short wavelength light. And I only say most because exceptions would be made for yellow/amber LEDs which are designed with a non-white light source.

That massive spike on the left is what causes refraction and loss of ability for human eyes to focus on the light source.

 

First of all, my apologies. If you're trying to enter an argument, you should never mix units -- as I did with temperature and wavelength. What was worse is that there's no magic formula to directly convert wavelength into degK. You are correct that temperature defines the color. So now, I've lost the right to try and make my original point. But that doesn't mean I'll stop arguing.

Your chart is telling. The LED is an amazing device for creating visible light in a very narrow, high amplitude, (the spike you refer to) portion of the electromagnetic spectrum. Paraphrasing your response, having such a narrow spike closer to the ultraviolet end of the spectrum is the issue, correct? I believe (opinion) the LED is the future of automotive lighting, but since we are in gen 1.0001 of the automotive headlights, there are issues to work through. The trick it appears will be for the LED manufacturers to broaden and essentially flatten the curve near what your chart shows at around 440nm in order to keep the color but bring more eye-friendly wavelengths into the spectrum.

 

I definitely agree the LED is the future of automotive lighting and in no way meant to suggest otherwise.

What is confusing is that the high amplitude spike in the low wavelength portion of the spectrum has already been solved by LED manufactures but I have yet to see it applied to automotive. Residential and commercial LED lighting continue to push the technology envelope developing higher quality light but in automotive applications the light quality really does not seem to be deemed important. Even looking ahead at automotive lighting technology conventions the consensus seems to be the LED light quality is good enough. However, studies have been published that show short wavelength light is actually much more difficult for older peoples eyes to process, leading them to be less safe for older drivers. So maybe they will evolve sooner than later, as you mention automotive LEDs are still in v1.0.

This is an apple to oranges example of a high quality soft white interior LED light bulb compared to a the Rigid LED fog. The LED spectrum can be manipulated. While the GE LED light profile would be superior for automotive, significantly cutting down the short wavelength emission reducing refraction and glare being safer for oncoming drivers, safer for older drivers, it also has a stronger profile in long wavelength for good performance in rain/snow/fog/dust and is drastically higher CRI so you can differentiate colors for easier object identification. The drawback, it would look like a halogen. Hard to sell the expensive upgrade to a consumer when they cannot tell a significant difference and other people cannot either.

 

Heres a great review of the Philips H4 LED headlight bulb installed in a Toyota Tundra reflective headlight housing:

http://www.xenondepot.com/2016-tundra-Philips-H4-led-bulb-s/82.htm

 

I’ve tried those too.
https://www.tacomaworld.com/threads/the-ultimate-headlight-upgrade-vs-philips-led-headlights.460313/

Unfortunately the lux meter I used was not an LED grade meter, so the lux results for the Philips LEDs are way over inflated. I have far better equipment now for testing than I did back then. I also wish I spent more time than short range wall shots analyzing the asymmetrical design of the glare guard as it would appear to likely cause increased glare to oncoming traffic at distance to the left, which would only be an issue in markets where you drive on the right side of the road. Not like the left side driving in the market for which it was designed. Though the glare would be nothing like most other LEDs. It is on the to-do list to revisit as I don’t feel I spent enough time on that one.

 

Just an fyi... The asymmetrical glare gaurd on the Philips H4 is identical to the one on an H4 halogen bulb (it's asymmetrical on H4 halogen bulbs as well).

I personally have not metered the Philips H4, but you'll like this review (with lux measurements and exposure locked photos):

http://www.xenondepot.com/2017-toyota-H11-led-bulb-review-s/106.htm

It's not a reflector housing -- but still an interesting review.

Steve

 

An asymmetrical H4 halogen glare guard? Here is the German made Osram Bilux 60/55w H4 bulb that is the OEM bulb used by Toyota. Note that the glare guard is symmetrical from right to left.



The Philips LED H4 while at first glance appearing symmetrical are not. Looking closer and you will see that the 2 piece glare guard is actually very slightly offset from left to right. In addition in the Z-axis, the glare guard is also not symmetrical, kicking up higher on one side. Reviewing old photos it looks like this may in an attempt to control glare better on the driver side, but I no longer have the lights to physically analyze them to confirm. Like I mentioned, I wish I spent more time testing them.

I question that review. While a projector has a light shield to neutralize the glare when using a different light sources, internally it still users a smaller reflector designed for the halogen source.

Ultra high efficiency halogen bulbs use tighter wound, more precise coils than the 'standard' ones shown in my picture above to achieve more precise focus resulting in a brighter hot spot equating to 130+% more light intensity in the hot spot for greater distance projection. Talking fractions of a millimeter deltas on the filament. A major flaw in any retrofit LED in a halogen housing is they work the exact opposite way. Rather than being smaller and more precise than those tiny wound coils, they are very bulky by comparison and way offset from center resulting in dilution or elimination of the hot spot. Placing diodes in approximately the same location is insufficient. This is clearly evident in the first set of short exposure photos in that review, where the reviewer mistakenly thinks their hot spot has gotten much larger, not realizing they have actually lost it all together.

Bi-directional LEDs also do not uniformly light the reflector, even in the internal reflector in a projector. You can see the bi-directional LEDs do not fully light the reflector my picture below with a dead spot above the bulbs. Omnidirectional halogen light that the housing is designed for will fully utilize the entire reflector or projector assembly with the precision optics for which it was designed. Current automotive LED lighting technology does not replicate the omnidirectional light required for the halogen reflector/projector.

Non-uniform LED light distribution with a dead spot above the bulb.



Both these major design flaws of LEDs retrofitted into a halogen housing result in significant losses of beam concentration, which equates to loss of distance projection, which people often talk about as loss of 'throw'. Lux is basically a measure of beam intensity at a single point in the beam. High precision optics and maximizing the reflector surface area to efficiently focus the beam concentration drive higher lux numbers. Loss of beam concentration/precision and dead spots in the reflector equate to loss of lux.

So how does this reviewer claim a 1750 lumen LED that suffers significant design handicaps for use in a halogen housing, which will result in far reduced lux readings, somehow outperforms a higher output, high precision, 2100 lumen H9 halogen for which the housing is designed?

I'd be curious if they are using the correctly calibrated testing equipment. Standard light meters will way over-inflate an LED lux measurement due to the assumption of an incandescent light spectrum profile for the lux calculation. Same mistake I made testing the Philips bulbs in the thread mentioned earlier. To read LED lux and compare to halogen lux you need a specialized multi-source LED compatible light meter, and to make it just a little more complicated it needs to be calibrated for the LED light color you are measuring. Quite a bit more expensive than a standard meter.

White light always looks 'brighter' in photos. Just look at how many victims Silverstars have claimed, people are completely convinced they are brighter, even with wall shots, when really they suffer a 20%+ loss in output over a basic halogen bulb due to the coating. I've actually mailed people light meters, taught them how to use them, just so they could prove themselves wrong first hand. Short wavelength light from LEDs also refracts at a hyper level causing surface glare back at the camera resulting in overexposure, compared to long wavelength halogen which has minimal short wavelength light, minimizing any refraction. This is why LEDs hyper illuminate street signs, and why they are terrible in rain/snow/fog/dust compared to their long wavelength counterparts. Wall shots are good for pattern, not great for output. It is a 2D slice of a 3D projection.

There are so many facets of light testing it is difficult to distill down to a single metric.

 

Are the Osram bulbs still the best for a simple upgrade ?

 

@crashnburn80

Your posts should be stickied. Enthusiast automotive groups like ours are are full of outright misinformation and wishful thinking when it comes to lighting.

 

Best is relative. The 100/90w Osrams have the highest output, but you will get better distance projection (with lower output) using a high efficiency stock wattage bulb, both Philips and Osram have released +150 bulbs.

About every single one is, except oddly for my SAE LED fog thread. Likely because it challenges products of supporting vendors for use as fog lights, though they have their uses for other purposes.

 

Is it me or can anybody else see a blue reflection that's offset from the actual headlight when looking through the rear view at a car that is running LED's?

 

The aura ring of glare? Or you referring to an artifact in the mirror?

 

It's hard to explain. So in my rear view I see the bright white of the LED in the headlight and then about a foot lower there is a blue shape. I modified this picture for reference.

 

I wish I had read all of crashnburns threads before I got opt7 LEDs for my taco and sv650s. Now Im switching over to brighter halogens with dedicated wiring for the Tacoma and will do so soon for the bike as well.

 

Thanks for the post

 

M

WOW!! That name is bringing back memories!
He answered some of my emails, way back in the early nineties, to some questions I had regarding differences between European and North American headlights standards.