Comparison Test: LED drop-in headlight bulbs vs. halogen bulbs [pics inside]

What am I Doing if I'm not Going Crazy?
The original posts are quoted below. In light of learnings about the human vision system, pun intended, I'm re-amping this thread. If you *really* *really* *really* want to understand what is going on, please read what I've written. I'm not an expert, but I absorb technical information fairly well and figure out what is useful to what I wish to accomplish. You're encouraged to seek out any specifics here to validate them because I'm only human and to err is to be human.

How Applicable is this Information?
It has been nearly 2 years since the start of this thread and I've spent no small sum in buying products and taking out time to have rigor in my tests.Technology can change incredibly fast and we're looking at LED replacement headlamps that allegedly boast anywhere from 4,000 lumens to 10,000 lumens per bulb. Ideally, we're trying to recreate a filament/halogen/incandescent bulbs with a different type of tech and have a useful and safe replacement.

Is this a Product Placement/Endorsement?
I'm not allowed to curse on the forums. I dislike branding and product placement I do not have shirts with logos on them, I removed my from VW emblem as one of my first 'mods' on my B4 and I will always side with what works “best” to the applicable scenario and circumstances. The follow product is mentioned, reviewed, and 'approved' because it in my ultimately biased opinion, it works. Your milage may vary and so will you opinion.

For the TL DR crowd
Why should you trust anything that I'm saying or typing?
I'm a photographer in a large entertainment department and my specialty with photography is dealing with low light aka night and I've been shooting for ~15 years. All oh my peers warn newly hired individuals to not ask me a question unless they truly want to know the answer in full, technical detail. Despite being red-green color deficient, I've spent a great deal of my time understand color systems to where I can actually tell when I'm dealing with situation where my eyes do not see what everyone else sees, though despite this, I'm well adjusted to common photography tools and can make photographic corrections that my peers cannot due to my deep understanding of how these systems work. Specifically, in the past year alone I was graced with some free reign to work on various projects and one of which allowed me to trace back correlated color temperature and white balance set in cameras to their physics origins and the mathematics behind adjusting these. Long live Plank! Enough about me, back to [head]lights, okay?

Layout
As I've been writing this out, I've determined that I'm going to break it up into sections. I could start a new thread. We have too many of those. I have provided links to the sections of previously reserved posts and maintained the integrity of my old posts and their many inaccuracies.

Scroll down or click here to read about the terms that have a significant roll in understanding *why* and *how* the light from LED, filaments, sparks (HIDs), or lighting things on fire affect how well we see when we are using light from out headlights

Scroll down or click here to read about the the physical size and locations of LEDs and filaments and why almost every single drop-in LED headlight bulb replacement is a complete waste of moment and quite typically a safety problem for you and other drivers

Scroll down or click here to read over the review of the Amazon LED headlight bulbs that I've purchased and will be modifying to meet appropriate specifications to replace my H7 lamps in Ecode reflector based headlights

I'm looking forward to your photos. Hopefully they show good beams and not just an LED module in the shape of a headlight that makes light come out in an approximately forward direction.

I have one set that arrived, the Big COB variant, and the other two have shipped. Only one of those two has info in the tracking system and it currently says it'll be here May 20th.

Joy

That's a slow boat!

Applicable Terms

• Lumens: Basic unit of luminous flux. This as the raw output of light.
• Lux: Basic unit of illuminance. This as how much light is cast upon a surface. If you focus a 100 lumen flashlight from a narrow illumination patch to a wide illumination patch, you still have a 100 lumen light, but you decrease the lux by having the beam cover a wider area.
• CRI: The acronym stands for Color Rendering/Rendition Index. CRI is a unit-less number, though actually a percentage specifically, that represents how much of the visible spectrum is reproduced by the light source. To me, this is the most important figure. This number though is also rated using photopic vision.
• Efficacy: Expressed usually as lumens per watt or lm/w. This is how well a light source converts power into light. In a way it's how efficient a LED is, but do not confuse this word with efficiency. It is common that low CRI, say the 70s, have efficacies of 130 lm/w whereas high CRI LEDs have efficacies of 90-110 lm/w. This is also a term used to describe the sensitivity of a sensory device or organ like camera sensors and the human retina.
• Luminaire: Standardized fancy name for a lighting source, lense, and structure. A headlight is a specialized luminaire as well as flood lights used in warehouses.
• Kruithof curve: This is info from a guy that is used in the lighting industry as a "reference" as to what color of light is most pleasing. It is a guide to choosing a CCT based on what is considered aesthetic based on lux. It *does not* take into account science or the efficacies of the human vision system.
• CCT: Corrected Color Temperature. This refers to the black body temperature, which can be correlated to the spectral power distribution of the light source. It is quite literally the heate temperature of an ideal black body on the Kelvin scale. Explaining all of that covers multiple fields in chemistry, physics, and biology. Just know that it's measured in Kelvin, 'K' for short, and this correlates directly to a physical temperature in the typical range of 2,500° to 10,000°. Be careful to not confuse it with a order of magnitude designator 'k' or kilo- meaning we're multiplying the value by 1000. Values that are lower are perceived as 'warmer', such as red hot steel, and values that are higher are perceived as 'cooler', think white hot steel. Out of all of these, this is a very close second place to the most important specification of a light source and is the most often misunderstood.
• Mired: pronounced as “mirrored” but dropping the “ored” sound for “ed” like the name. It's shorthand for micro-reciprocal degree and is a odd term and unit-less number. By nature of the "micro" prefix, we're dealing with one, one-millionth of a reciprocal degree, which means you take the number one million and divide that by the degree. What degree though? The degrees in Kelvin of the CCT you have. Using mireds we're able to do mathematical manipulation of color tempertures with much greater ease because it makes it much easier to tell how far 4000 K is from 2500 K or 5000 K is from 6500 K.
• Dominant Wavelength: Our eyes evolved with the sun and there are very few light sources that replicate it. If there is a 'peak' or 'spike' in the visual spectrum from an artificial light source, it is referred to by this term. Fluorescent lights have a dominant wavelength in the green portion of the spectrum due to their use of mercury and it's emission properties.
• Spectral Power Distribution: Aside from which wavelength of light is emitted the most from a light source, how much of the rest of the visual spectrum is represented is refered to by this three word term. Even though light that is white can have a warm cast with lower CCTs or a cool cast with high CCTs, knowing what the CCT of that white light is will be determined by it's spectral power distribution. Moreover, as this term take space to write and is a pain to constantly speak, it is very commonly refered to by it's initials, SPD.
• Dynamic Range: How much range something has within it's defined system. You can think of this as the RPM range of a TDI engine, the 3 inHg to 17 inHg range of a VNT actuator, and many other things. In this context, we're referring to the dynamic range of the human retina and a camera's CMOS/CCD sensor the perception of light.
• Photopic vision: The 'mode' at which we all see during normal daytime or in a moderately lit office environment. This vision is dominated by the cone cells in your eye and this is how you see color. Photopic vision is from about 10 lux to tens of thousands lux. The most sensitive wavelength is at ~555 nm and a efficacy of ~683 lm/W. The most commonly referenced model is CIE1951 for photopic vision, but it has some flaws. A good site for the proposed CIE2006 model which corrects these details have a look into the research here. A comprehensive PDF on the topic is here, too.
• Scotopic vision: The 'mode' at which we all see during night time without artificial light. Literally, this is our night vision. This vision is dominated by the rod cells in your eye. Scotopic vision takes place between about 0.000001 lux to 0.001 lux. The most sensitive wavelength is at ~505 nm with an efficacy of ~1700 lm/W.
• Mesopic vision: This is the in between 'mode' of your eyes that in our modern society most will people mistake for night vision. Due to the developed world's broad use of artificial lights, this mode of vision is used when people think they're actually using scotopic vision, aka night vision. If you're in a city where the street lights have begun to be converted to a 'weird' cool white LED luminaire, it is because of research into mesopic vision, which currently is viewed as a simple blending of the two other modes.
• S/P Ratio: A primitive understanding to mesopic vision lead to the first crude analog that was utilized to “determine” how effective a light source was for mesopic vision. Due to the different efficacies of scotopic and photopic vision, a simple ratio between the luminaire's efficacy ratings between the two. How poor of a judge is it? It's analogous to increasing boost or fuel maps by 15% and expecting your TDI to still run correctly. The reason why this is such a poor measure is due to the fact that cones and rods in your eyes are not always functioning at the same time. If they were, then this number would mean something.
• Cones: The cells in your eyes that have 3 different types of pigments, for normal people as least, which are responsible for photopic vision. They are very insensitive to light below 0.1 lux, or the equivalent to a full moon on a clear night. Overall, their dynamic range is from about 0.01 lux to 10,000 lux.
• Rods: The cells in your eyes that have a single pigment that is more sensitive to bluer forms or white light. However, this pigment bleaches very very easily and can take 30 to 60 minutes to refresh in your retina in order to regain visual input again. They are a sensitivity that ranges from about 0.000001 lux to ~10 lux and are typically bleached by 100 lux.
• Pupillary Response: This is a fun one that gets overlooked and tells us why more doesn't always equate to better. If you start shining light too bright in a certain area of your vision, your pupils will contract and make the rest of your vision dimmer, thus harder to see. I personally experienced this when using Philips X-treme Vision, +130% efficacy lamps in my headlights. This was in conjunction with my relayed harness. I opted to switch to long life bulbs and though they emitted less light I was able to see my surroundings more easily.

Post reserved for new content

Post Reserved for new content

Digital Corpus said:

So, since this thread came to an inconclusive end, I've decided to pick it back up. Looks like a few things have changed on the market of LED replacements. Since the base tech is a semiconductor, this tech can change very quickly in a short period of time.

From what I've been seeing on [fle]eBay, the most common replacement bulbs are COB or Chip on Board LEDs. I noticed a few different form factors present in the construction of the "bulb." As the LEDs have to disipate a fair amount of heat in a small area, cooling them is fairly critical. There are 3 types of cooling methods on the market. Most of the knock off or "cheap" (though they bear the same price) have a fan mounted to a flat piece of aluminum on the bottom. Then we have my 2nd choice which is a large passive heat sink on the read mount. Finally the best option is a fan on a properly finned heat sink.

As for light arrangement, there are 6 different types of setups I've seen. There is one variant that is a single sided LED. This one is harder to find and probably has the worst lighting. The "2nd Gen" variation below is an example of such. All others have at least 2 COB LEDs on opposing sides of the "bulb."

Small COB Non-offset LEDs:
http://www.ebay.com/itm/2x-1800LM-H...Parts_Accessories&hash=item43c7fda1b6&vxp=mtr

Small COB Offset LEDs like in the first post of the old thread:
http://www.ebay.com/itm/2x-H7-White...Parts_Accessories&hash=item1e89be97bb&vxp=mtr

Small COB LEDs (this is a "2nd Gen")
http://www.ebay.com/itm/one-pair-H7...Parts_Accessories&hash=item3cdaaaeb95&vxp=mtr

Big COB LEDs
http://www.ebay.com/itm/2x-Super-Br...Parts_Accessories&hash=item1c3614dc56&vxp=mtr

And a proprietary COD LED w/ a reflector:
http://www.ebay.com/itm/CREE-48W-44..._Merchandise&var=&hash=item1c3d7d3ed6&vxp=mtr


I'm going to pick up the "2nd Gen" version and the reflector version and see how they perform. Each of these sets are at least 1600 lm, allegedly, per bulb and a stock H7 bulb is rated for ~1400-1600 lm. I'll post photos in my better-than-stock, not-as-good-as-Ecode B4 headlight assemblies and see if I can get a hold of a friend who has Ecodes and try them in that set.

I'm enthusiastically curious big COB LEDs because the heat dispersion should be greater and due to the larger emission surface area, which most likely will increase the dazzle effect of the bulb. Since my high beans are pencil thin, I'm curious how the bulb will fair in that socket. I suspect this bulb should work amazingly well if you need a flood light and have a socket for an H7 bulb.

Also, all but the last LED, the one with the reflector, are rated at a white point of 6000K. In daylight, this will have a cool or slightly blue feel to the light. At night, it will be much more evident and look like a "blue bulb" version of "ultra white" stock incandescent bulbs. The last one will still have a cooler color temperature at night since most lighting is in the 2200-3100K range, but is closer to a daylight white point and this is the highest Kelvin rating I try and use for "white" LED.



I'll post back with results as soon as I get them and have time to get them out. I'll grab some against-the-garage shots and also try and do some range photos as well.

-Michael

Click to expand...

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Digital Corpus said:

Here are the initial photos of our first H7 LED candidates: 4th link in my previous post, the Big COB LEDs. The are rated for 40W of power, of which I believe they will consume and 3200 lumens, though I'm not sure if this is total or per bulb.

This pair is the one that I hope will work well as my high beam due to the surface area that the light is emitted from. I expect this to cause a problem in the low beams due to the fact that the reflectors are designed to focus the light from the area of the filament instead of the 7.5 mm x 18.75 mm area. Click any photo to embiggen

Box it comes in to look pretty:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7268-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7268-hi.jpg

Here is a full body shot. The heatsinks are ill-handled as there are many dings in them, but I do not see anything that would impead their performance. Also, I'd like to note that the covering seems to basically be yellow silicone:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7275-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7275-hi.jpg

The heatsink is detachable for the simplicity of construction I assume.
[URL]http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7289-low.jpg[/URL]

Electrically, I was curious as to what made these things tick. I actually expected a giant current limiting resistor in the black aluminum chassis. Be advised, these are not weather sealed in any way.
[URL]http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7283-low.jpg[/URL]

This circuit board has a 6-pin DIP IC responsible for it is marked "5241a" hidden under the 100 uF cap. From what I can tell its a LY5241a, SM5241a, or QX5241a, which is a Chineses only IC step down converter with a max output of 2 A.
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7284-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7284-hi.jpg

Additional PCB shots. Solder joints aren't all pretty, but they are electrically sound. Oh, and those 2 pair of black ICs, that is a rectifying setup for AC power, which is strange in this use, just saying:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7286-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7286-hi.jpg

http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7287-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7287-hi.jpg

The power FET. The aluminum chassis actually presses n this guy a little bit, which has to be by design, so the chassis acts as a heatsink too.:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7288-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7288-low.jpg

My biggest complaint is that the male plug from the output of the power supply circuit is polarized, but the female plug on the actual bulb is not polarized. However, I'm assuming that the wire coloring isn't random so I'll match red to red and black to black when I plug them in.
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7282-low.jpg
http://mobrienphoto.com/Misc-Pics/TDI-Pics/Lighting%20%26%20Electrical/LED-H7-Tests/Big-COB/_OB_7282-hi.jpg

I *did not* unbolt/unscrew the COB LEDs from the aluminum housing. It looks like there is some TIM in place to help transfer heat from the LED PCB to the aluminum chassis. If these bulbs work as a replacement bulb, I may use some spare CPU TIM as replacement TIM and also put some between the bulb base and the big passive heatsink. Anyhow, this concludes the non-light portion of the review for this LED bulb and the following smilie is the 10th picture (forum limit) in this post .

Time to move some cars around and drop these into the low-beam and hi-beam sockets and see how they perform. Photos and analysis will be posted tomorrow.

Click to expand...

Digital Corpus said:

This first post is regarding my "stock" headlamps. I'm actually using some angel eye headlamps from eBay. They are a lot better than stock, but not quite as good as Ecodes, though I'm not compared the light outputs directly.

The low beams used are generic H7, not high efficacy. The Osram/Sylvania PN is 64210. High beams are Philips X-Treme Power bulbs.

Unless otherwise stated:

• All photos taken at a shutter of 1/60th, aperture f/4, ISO 800 on a Canon 7D in RAW. This exposure was chosen to avoid high beams showing up as pure white on the garage door.
• Photos processed in Lightroom 4 with white balance set to K, 4250, the logarithmic middle point between the color temperatures of the bulbs to accurately show the color difference between them.
• Only noise reduction and sharpening have been applied except for 1 photo which has been cropped.
• No tonal adjustments have been made to exaggerate luminosity differenced
• The human eye has adaptable dynamic range vs a camera's sensor which has fixed dynamic range. As such, these differences may appear greater or lesser if you conduct these same tests.
• A WhiBal G7 was used to measure, in conjunction with LR 4, the white point of the bulbs. The material used is accurate enough to use in commercial photography.

"Stock" Light Patterns and Luminosity
H1 Fog lights:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Halogen/_OB_7295-low.jpg

H7 Halogen Low Beams:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Halogen/_OB_7297-low.jpg

Low Beams + Fogs:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Halogen/_OB_7301-low.jpg

High Beams:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Halogen/_OB_7298-low.jpg

High Beams + Low Beams:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Halogen/_OB_7300-low.jpg

Yes, the driver's side is lower. I did this intentionally to help avoid dazzling opposing drivers during my road trip to Death Valley for astrophotography when there was a new moon. Doing this review made me measure out the setup and I use at the house too:

• The garage door is 24', 6" from the headlamps.
• Each horizontal section of the door is 21" tall.
• The center of the headlamp is 26" from the ground.
• The car rests at a declination of 6.5˚ at this distance which is +1˚ above level. This means my center "kick" in the beam pattern should be at 31" from the floor if the lights were level.
• Ecode headlights should be angled to drop 3" at a distance of 25' so that puts me at a height of 28" to meet the spec.

Looks like I need to realign my headlights after measuring out the details...

Click to expand...

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Digital Corpus said:

Color temperature of light is a very tricky topic. Generally speaking, it follows black body radiation as a guideline. As such this is where color "temperatures" come from because they are directly linked to the temperature of the reference substance in degrees on the Kelvin scale. I say guideline because even though light from an LED appears white, it rarely covers the full spectrum our eyes are used to from the sun.

As a reference, daylight is about 5500K. Here is SoCal, I find that 5000K is more common which is probably due to SMOG, which makes our sunsets oh so wonderful. Any number higher that this will begin to look cool or blue(r). Any number lower than this will look warm or orange(r).

My biggest complaint about LED lights is their color temperature, though this is mostly a result of the process technology that has to be used. Despite being advertised as 6000K, I measured the LED light to be mostly 6300K.

Stock lights balanced to 3000K and then to "neutral" 4250K:
http://mobrienphoto.com/Misc-Pics/T...ical/LED-H7-Tests/Halogen/Stock-White-low.jpg

Big COB LEDs balanced to 6300K and then to 4250K:
http://mobrienphoto.com/Misc-Pics/T...al/LED-H7-Tests/Big-COB/Big-COB-White-low.jpg

Big COB in the Low Beam sockets:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7302-low.jpg

Low Beam + Fogs:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7304-low.jpg

High Beam + Low Beam:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7303-low.jpg

Big COBs in as High Beams (misaligned):
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7308-low.jpg

Low Beams + Big COB High Beams:
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7307-low.jpg

I noticed that the "H7" base is a friction fit and the bulb can be rotated.

After adjustting both High Beams by rotating the bulbs
(FYI, the shot is cropped because I didn't realize I bumped my camera)
http://mobrienphoto.com/Misc-Pics/T...ctrical/LED-H7-Tests/Big-COB/_OB_7312-low.jpg

The light output is too high and simply too diffuse/unfocused, even for high beams.

Light is a funny thing. If you have at least a glancing familiarity of any color computer monitor or TV, you'll know that we use 3 colors to represent white when they are combined: red, green, and blue. Though not definitively red, the majority of the light on our roads is tungsten based and with a color temperature of 2700K to 3300K, "red" is the strongest component. Here is that the funny thing, you can increase green and blue and the overall brightness of that light doesn't really change, generally speaking.

So, if you add in 1600 lumens of bluish light to 1600 lumens of orangish light, your eyes will still only see about that 1600 lumens (this is an oversimplification). However, if you add 1600 lumens or orangish light to 1600 lumens of orangish light, you'll get that 3200 lumens of light. This means that when you use "blue bulb" halogens, or HID lights above about 4500K, or even these LED lights and they are they have the stock light output, you'll have difficulty seeing on the road because things will not appear brighter. However, since using Halogen bulbs closely matches the regular street lights, you'll get added visibility.

Anyhow, with that being said, these bulbs are basically crap. Unless you have some special use for them in non-standard housings, or want an overly focused high beam to behave as a flood light, they have no value. Especially because of their color temperature. I await to see how the others will fair.

Click to expand...

You rock. How much money are you investing in this research project?

Waiting with bated breath

Yep, those look like crap.

VeeDubTDI said:

You rock. How much money are you investing in this research project?

Click to expand...

It would be over $200, but I'll be returning this current set of LED bulbs. The others are from China so we'll see how quickly I'd get my money back if I return them.

Thanks for doing this, looking forward to the Gen 2 especially in Ecodes.

According to my tracking April 23rd is the soonest I'll get the reflector variation and May 6th is the soonest I'll get the 2nd Gen variation.

Reflector variants showed up today.

BTW, I'm free for you to try them in my golf late tonight after work or sometime Saturday.

PS - Subscribed

Any updates ??

I have the photos of the 2nd round of lights and have had the 3rd set for about 3 weeks. My apologies for not getting to testing them. I have another project of mine, D-DAQ, that I've been working on and was about to begin prototyping when a hardware design contest came along and I decided to dedicate my free time to that endeavor. I have not forgot about these and I'm sorry for not updating this thread. I'll see what I can do to resolve that over the next few days.

No worries. Appreciate the effort for testing these for us.

Its not like other drivers are on the roads. These bright lights aren't that bad in cars. But SUVs and trucks with these powerful illegal bulbs are such a pain. Starting to see commercial diesel trucks with illegal bright lights.

Beam pattern trumps brightness of the filiment

I second beam pattern. Converted my '93 Toyota pickup from the factory sealed beans to CiBee Euro H4's and I effectively run brighter bulbs with nearly all the light hitting the pavement. There's a sharp cutoff in the lens/pattern that directs the light to the ground on low beam and has a direct & overwhelming spot/flood combo on high beam that turns night into day when needed.


Sent from my iPhone.

Bumpage. Updates?

I have a set of Chinese CREE LED bulbs I put in. The beam pattern was somewhat better than you tested, but still not that great. They are 4300K I believe. Anyway after one "bulb" took a crap I put the halogen H7 back in and it is noticably better (halogen > LED).

But knowing it would be a long turn around time and doubtful of Chinese quality I bought two sets. The second set of bulbs are still in the high beam location, for which they do admirably well. And with the amount of time I can actually use high beams in this crowded world, the instant on/off of the LED high beam is more useful as oncoming drivers recognise it faster. Before the LEDs I had some people just not get the hint.

And now I have a set of HID with proper BiXeon projectors retrofited in place. The LED high beam is mainly just there for flash to pass/signal other drivers. Works well because I can flash the LED without flashing the HID bulb every time. Also has the nice side effect of making my high beams look yellow when off (the CREE LED is yellow when off)

I am still waiting for good LED headlamps to come out that don't cost more than my whole car.

Jason

I tested the 2nd pair, but not the third. My time has been fleeting because of the project I noted in post 19. I'm the R&D & prototyping has my free time consumed right now. With the additional roles I do at work due to the scenario, I tend to be pretty exhausted once getting home too.

I'm sorry for no updates, but many things in my personal life are currently on hold.

I eventually gave up on my H7 test as well.

http://forums.tdiclub.com/showthread...Led+light+bulb

I still have the bulbs, but I've not had time. There is also a daq electronics project I've been working on that has been put on hold. A few weeks ago I had to pull the transmission for the release lever spring that broke and I ended up damaging a seal on an output flange. I have vacation starting on Sunday which I'm using to fix a couple things, including said seal, flange, outer CV joint, turbo oil drain line, turbo exhaust housing, exhaust fabrication, headliner and sunroof rehab, intermittent problems with a new hard drive in my server, data migration amongst 3 OSes, editing some family photos with our newest pets, etc etc etc. Eventually, since I still have the bulbs, I'll be getting back to this.

ANd I think you may mean this link: http://forums.tdiclub.com/showthread.php?t=396634

thanks for the fix.

So, due to some research into some "fundamentals" regarding light with respect to both photography and human perception has caused me to rethink a few things. I unfortunately have had too many things on my plate still, the least of which was an accident where someone else hit me and I'm waiting for the insurance right now, but I still have this project sitting in the back of my mind begging for completion.

Though my original focus was beam pattern, and it's largely an important topic, CRI, CCT aka white point or white balance, scotopic/photopoc ratio, and spectral power distribution come into play. The latter two are apart of a public works movement in many municipalities to reduce power consumption and increase visibility under mesopic lighting conditions.

In short, once life calms the f&%# down, I'll be come at this project is a much much more thorough manner.

There is also an Amazon LED replacement that I want to pick up because I know what LED it uses and I'm curious as to the results of swapping out a different variant would be. Anyhow, despite the similarities in the topic of this thread and what I wish to cover in addition, I'll probably start a new thread when the time comes.