Announcement

Collapse
No announcement yet.

Sealight LED vs. H9 Halogen

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Sealight LED vs. H9 Halogen

    First I want to say thank you to everyone active on this board. I've been lurking for a long time because I'm always interested in getting the best lighting for my vehicle. It's always been a little disappointing finding some real data on the latest LED upgrades, though. I've spent plenty of hard-earned money on Amazon LEDs and LEDs from elsewhere and found them all to be pretty lacking.

    I've also found most of the "testing" pretty lacking as well, and that's inspired me to invest some time and effort into properly testing these bulbs.

    Earlier today I received the last of my big order from Amazon. I ordered quite a few things to be able to accurately test LED bulbs and I hope to refine my testing process and procedures with time--it's just hard finding time to do this on the side and especially doing it correctly instead of just slapping some things together and taking random measurements.

    My setup is a "90mm" Hella H9 low-beam projector. I also started by testing a "Sealight" H9 LED bulb from eBay, since Sealight seems to be a popular brand. I also bought a Extech LT40 light meter after reading about how most people are using the wrong lux meter. I believe I read it on either Reddit or TacomaWorld that the other Extech lux meters didn't have the correct spectral response curve for accurately measuring LED light sources.

    Anyway, the only place I could find to test bulbs in my house conveniently was in my kitchen/living room. My testing wall is about 18 feet away--yes, not ideal, but at least it's not too close.

    I carefully aimed the lamps to "visual optical left" specification, which means the left side of the cutoff should sit 0.4 degrees below the horizontal. To help me find the horizontal and vertical, I bought a DeWalt laser plumb, and I used that in conjunction with a Bosch Blaze laser range finder along with some trigonometry (god, high school was too long ago) to find what 0.4 degrees below the horizontal translated to in terms of inches below the red laser line on my living room wall.

    I decided to wait until night, when my wife and kids were sleeping, so I could sneak into kitchen and play with the headlamp and the bulbs alone, and with the lights off. I decided to take measurements at just 0.9D, V for tonight--that's 0.9 degrees below the horizontal, and straight ahead, which I believe corresponds to about 50 meters (~150 feet) down the road.

    And there's a reason I choose 0.9D, V to start with, it's one of the points that has a legal requirement. I specifically chose not to wave my lux meter around. I hate how people wave lux meters around to take a random peak measurement--thank you evo77 for your explanations on this forum and Facebook, I learned a lot from you!



    The results at 0.9D, V were:

    Sealight: 130 lux
    Osram H9: 279 lux
    Legal requirement: at least 149 lux (at 18 feet)

    Max reading from waving lux meter around for the Sealight: 170 lux, and I found this max by setting the light meter to "max" mode--it will only remember the max value and display it--and waving it around all over the beam pattern on the wall. I found the max somewhere really pretty close to the ground--way below test point 0.9D, V. This reading of 170 is greater than 130, yes, but it's also useless--you want light as high as possible so it can travel as far as possible, not an oversaturated, blown out foreground.

    Obviously, I could stop here, but I'd be doing only slightly better than the other testing outlets. For one, it's very important to note that my adjustable power supply maxed out while trying to power the H9 bulb--the H9 bulb is rated for a nominal 65 watts, but my power supply tapped out at 12 volts and 4.931 amps, which is 59 watts. A vehicle with a functioning alternator should have no issue feeding the H9 bulb the full 65+ watts. So, in short, the Osram could have been even brighter had I ordered a beefier power supply (I couldn't find any for a reasonable price).

    I also don't like lux so I'll convert to candela at 0.9D, V:

    Sealight: 3913 cd
    Osram H9: 8398 cd
    Legal requirement: at least 4500 cd

    It's pretty clear that the Sealight got whooped, but that's pretty much expected. First, it's $45. Second, we all know PnP LEDs generally suck, doubly so for cheap Amazon/eBay options. Third, it's passively cooled and maxes out at about 13 watts (12 volts, just over 1 amp). If we assume a luminous efficacy of 100 lumens/watt, then it has about 1300 lumens. A real H9 bulb is supposed to have 2100+ lumens.

    I'm now interested in testing the premium options, like Morimoto 2Stroke 2.0 and Diode Dynamics.

    Here are the pictures:

    Please feel free to suggest ways I can improve my testing--I know I have plenty of room for improvement, and let me know if there's anything specific you would like me to test, as I find this absolutely engrossing (much more than my day job!)



    Sealight result:



    H9 result:









    Sealight Power draw:



    Approximate location of peak Sealight reading:






    Last edited by projector-head; February 15th, 2020, 03:42 AM.

  • #2
    Interesting post. Thanks for sharing your tests.

    Re-reading the table in that FMVSS 108, it seems odd to me that (4D,4R) needs to be that bright relative to many other points. I would think while it is to the side, it is close enough to the vehicle that it needs to have a maximum allowable brightness instead of minimum. No?
    Jul 2012 ROTM (3-way quad headlight) ; Sep 2015 ROTM (custom muli-lens 7" fogs)

    3-way quad wiring; foreground limiter; squirrel finder;

    Comment


    • #3
      4D,4R in that chart must be typo, it supposed to be 12,500cd max


      Observing beam quality has few different levels.
      Meeting FMVSS only telling you that beam satisfy minimum requirement, but that is far away from what is overall ideal beam pattern.

      I wrote article about one aspect of beam quality here
      https://betterautomotivelighting.com...-distribution/

      For example, measuring 0.86D V point only can tell absolute measurement of candela reading, but the gradient of intensity distribution changes how we perceive illumination impression.

      It is explained in above linked article.

      LED bulb will not truly replace filament light source, simply because of LED to be surface lighting device.
      However, effort towards achieving better result with LED bulb can be concluded as “ how accurately can mock filament behavior being surface lighting device

      Which comes down to
      1. How thin( distance between one LED emitting surface to another ) if can get
      2. How accurately contain emitting area size into filament size.

      Not to mention LED must be same distance position from bulb base.


      In pictures, first degree problem is LED position to be off. Being too short, this essentially means bulb is not inserted all the way effect.

      Cross section of projector reflector bowl is ellipsoid, it will result in focus point off set after cut off shield, will cause vertical scatter.

      Reflector optics vertical cross section is paraboloid, result in fail to parallel beam focus, scatter vertically.

      Even if positioned rather accurately, then size of the emitting surface impact beam pattern and lumen density.

      LED bulb have lot of aspect that need to be worked on to be able to replace halogen bulb

      Comment


      • #4
        Originally posted by satrya View Post
        Interesting post. Thanks for sharing your tests.

        Re-reading the table in that FMVSS 108, it seems odd to me that (4D,4R) needs to be that bright relative to many other points. I would think while it is to the side, it is close enough to the vehicle that it needs to have a maximum allowable brightness instead of minimum. No?
        Yes, that's a typo. It's supposed to read ≤12,500 at 4D,4R instead. Good catch! 4D on a average height crossover would translate to about 40 feet ahead of the car.

        You should go work at Bosch, because that picture was from the 8th edition of Bosch's Automotive Handbook, and my 10th edition (physical copy) makes the same typo!





        Originally posted by upashi View Post
        4D,4R in that chart must be typo, it supposed to be 12,500cd max


        Observing beam quality has few different levels.
        Meeting FMVSS only telling you that beam satisfy minimum requirement, but that is far away from what is overall ideal beam pattern.
        100% agree.

        I wrote article about one aspect of beam quality here
        https://betterautomotivelighting.com...-distribution/

        For example, measuring 0.86D V point only can tell absolute measurement of candela reading, but the gradient of intensity distribution changes how we perceive illumination impression.
        Very true, a few points don't tell the whole picture. I've been thinking of making my own low-resolution isocandela diagrams by moving the lux meter around on a grid on the wall, and taking the lux reading at each grid. Of course I'll know the exact coordinates of each box on the grid, and I could probably stick all the numbers into a program and make it spit out a color-coded graph/pseudo-isocandela diagram so we can get a better feel for the quality of illumination and distribution of light.

        But before I do anything like this, I would want to have a much longer testing distance. I'd rather present no data than present crappy data.

        18 feet is likely too close and as a result, the "resolution" of my pseudo-isocandela scan would likely be pretty crappy. I would want to be at least 25 feet away from any sort of wall, well, the longer the better, because then I can minimize the cosine error and get better resolution since I can't shrink the size of my photosensor (lux meter). I'd have to measure the diameter of my photosensor and sit down and do some calculations using the area of the photosensor and measuring distance. I think I would want measurements at 100s of points in the beam pattern. I read a study in which they used a real, lab-grade goniophotometer and the beam pattern of various car headlamps was contained within ~1500 points, while the gonio could take measurements at nearly 70,000 points despite the measuring distance being only 60 or 70 feet. Obviously that's some pretty high-resolution data capture!

        Cross section of projector reflector bowl is ellipsoid, it will result in focus point off set after cut off shield, will cause vertical scatter.

        Reflector optics vertical cross section is paraboloid, result in fail to parallel beam focus, scatter vertically.

        Even if positioned rather accurately, then size of the emitting surface impact beam pattern and lumen density.

        LED bulb have lot of aspect that need to be worked on to be able to replace halogen bulb
        Great explanation!
        Last edited by projector-head; February 15th, 2020, 03:56 PM.

        Comment


        • #5
          Originally posted by upashi View Post
          4D,4R in that chart must be typo, it supposed to be 12,500cd max
          Originally posted by projector-head View Post
          Yes, that's a typo. It's supposed to read ≤12,500 at 4D,4R instead. Good catch! 4D on a average height crossover would translate to about 40 feet ahead of the car.
          That makes more sense. A low beam with at least 12.5k cd there may look quite distracting


          Originally posted by projector-head View Post
          You should go work at Bosch, because that picture was from the 8th edition of Bosch's Automotive Handbook, and my 10th edition (physical copy) makes the same typo!
          I guess their internal design doesn't look up the table in that document; happens when information is repeated in several documents but the active document may have been corrected long ago.
          Jul 2012 ROTM (3-way quad headlight) ; Sep 2015 ROTM (custom muli-lens 7" fogs)

          3-way quad wiring; foreground limiter; squirrel finder;

          Comment


          • #6
            Originally posted by satrya View Post
            That makes more sense. A low beam with at least 12.5k cd there may look quite distracting
            Perhaps. I think it would depend on the relative distribution of light. Like if there's a lot of foreground illumination already, then 12.5k cd there might not be a giant deal since it would "blend" in.

            I think that's what upashi was saying--12.5k cd at that single point can't tell us much. If it was just 12.5k cd concentrated in a very small area at 4D, 4R, like pointing a flashlight out your passenger window at that point, then yes, that would be pretty annoying.

            For example, the Truck Lite LED headlights have a "X" or "V" or "crosshair" shaped hotspot, which doesn't take away from the fact that they are good lamps, but it is visually annoying.



            And it actually looks like this on the road:



            That literal hotspot on the road might drive some people crazy, though, and I've read reports on Jeep forums about how it actually does drive people crazy. Other people just pretend they're looking through their gun sights, as if at the range .

            If Truck-Lite had done what upashi discussed, and "smoothed" it out then it would be a total non-issue to everyone.

            To me, however, I'm more interested starting with the basics: does the resultant beam pattern meet requirements at the test points? Actually, does the LED bulb in a halogen reflector or projector even give me more light at the relevant points/zones? If I can just get more light in the hotspot region, or about 0 to 0.9 degrees below the horizontal, and 0 to 1.5 degrees right of the vertical, and without blowing up the glare zone, then I'd be very impressed. If I find any bulbs that do this, then I'll start examining the distribution of light within the entire beam pattern, perhaps by creating my poor man's isocandela chart.

            There's just so little useful testing out there, and while the truth is 95% likely to be what upashi said: LED bulbs are fundamentally different from halogen bulbs, so don't expect current-gen products on the market to work well...I'm still curious. We have one guy at TacomaWorld doing some testing, but it's not, in my opinion, up to snuff really. I wish he would test more than just one point, and I know he probably could, and he has much more advanced equipment than me, like a spectrophotometer! And he has 42' of testing space. I'm really curious how the Diode SL1 bulbs do, as they've been out for years and been making big claims, as well as the "new players" from Headlight Revolution and the such and the Amazon specials like the "eye of Megatron" Hikari bulbs.
            Last edited by projector-head; February 17th, 2020, 02:35 AM.

            Comment


            • #7
              I was able to take some more detailed measurements today after I bought some poster board. My wife would probably shank me if I dared to scribble on the living room wall. I was also able to move my entire setup back to >25 feet away from a wall, thanks to a tripod. I no longer had to put everything on the kitchen counter.



              I'm going to keep things simple for now and just measure the 3 hotspot points, as well as the 2 oncoming traffic glare points. If I find something that can put up good hotspot numbers while keeping the glare under control, then I'd dig deeper into that bulb.

              Being that this Hella projector is a visual optical left type lamp, I sought to aim the left side of the cutoff so that the cutoff would rest on a imaginary horizontal line located 0.4 degrees below the the true horizontal line.



              I ran the Sealight H9 LED at 13 volts, and it chose to draw just over 1 amp.

              Sealight H9, 13 volts, 1.082 amps (14.06 watts):

              Lower glare point (0.5U, 1.5L): 8.0 lux, 521 candela, legal limit 700 cd

              ~50 meters downroad (0.86D, V): 71.6 lux, 4671 cd, legal minimum: 4500 cd

              ~50 meters ahead, right edge of road (1.5D, 2R): 79.3 lux, 5173 cd, legal minimum: 15,000 cd

              ~75 meters ahead, right side of road (0.6D, 1.3R): 61.9 lux, 4038 cd, legal minumum: 10,000 cd

              Max search: I waved the lux meter around a la Headlight Revolution/Bulb Facts and found a peak of 83.5 lux, or 5447 cd at just a bit below 1.5D, 2R, and well to the left of it. In other words, a lot of light in the area ~130-150 feet ahead of the car. Which is kinda useless, since what we really want is light 300 feet ahead of us. 130-150 feet is fog lamp range.






              It makes me laugh that people are saying "the wall is dead" and "#wallshotsaredead" and then going out to test LEDs on a road with a max distance of 180 feet.





              At 30" off the ground, and setting the cutoff to "terminte" at 180 feet, you're dialing in a -0.8 degree aim for the headlamp. That's ridiculously low. But it's really no surprise that they're dialing in such a low aim figure. The Sealight LED I tested performed best at ~0.7 to 0.8 degrees below the horizontal. Examining LEDs at realistic distances of 200-300 feet would instantly reveal their shortcoming: lots of light up close, and little distance illumination compared to halogen.

              Speaking of halogen, here are my results for the H9 halogen bulb in the same Hella projector:

              Again, the H9 halogen is running at a pretty severe handicap. My power supply maxed out feeding the H9 bulb at 12 volts and 4.937 amps. We're feeding a 65 watt bulb with only 59 watts. We can look at it through lumens too: at 13.2 volts, a H9 bulb is supposed to emit 2100 lumens. Lumens and voltage are related by an exponential factor of 3.4. This means that at 12.0 volts, the H9 bulb here is only emitting ~1500 lumens. In fewer words, the H9 bulb is at a severe handicap.

              But the H9 bulb still manages to cream the Sealight bulb despite "having one hand tied up behind its back." Data below:

              Osram H9, 12.0 volts, 4.937 amps, 59 watts:

              Lower glare point (0.5U, 1.5L): 8.3 lux, 541 candela, legal limit 700 cd

              ~50 meters downroad (0.86D, V): 114.7 lux, 7483 cd, legal minimum: 4500 cd

              ~50 meters ahead, right edge of road (1.5D, 2R): 212.3 lux, 13,850 cd, legal minimum: 15,000 cd

              ~75 meters ahead, right side of road (0.6D, 1.3R): 151.2 lux, 9864 cd, legal minumum: 10,000 cd

              Some interesting notes:

              1) Despite a much fuzzier cutoff with the halogen bulb, the projector didn't create excess glare. The glare readings differ by 0.3 lux, or by a minuscule 20 candela. The glare limit at the tested point is 700 cd for comparison.

              You can see that the halogen projector has a pretty big color "band" or color "flicker."



              While the LED bulb has a much cleaner cutoff:



              The correct conclusion here is that a cutoff has almost nothing to do with glare. The sharpness of a cutoff has almost nothing to do with glare.

              2) I wanted to quantify the effects of heat soak on the performance of the passively cooled Sealight H9 LED. It doesn't draw as much power as some of the "big boy" LEDs. I've seen tests by members on here showing the actively cooled LEDs drawing as much as 25 watts and more. This passively cooled Sealight H9 LED draws ~13 to 14 watts. Could its heatsink be adequate?

              I took two sets of measurements for the Sealight H9 LED to examine the effects of heat soak on lumen (or I guess in this case, lux/candela) maintenance. The measurements were taken approximately 20 minutes apart.

              Sealight H9, 13 volts, 1.082 amps (14.06 watts) after 20 minutes of continuous operation:

              Lower glare point (0.5U, 1.5L): 8.0 lux --> 6.9 lux,

              ~50 meters downroad (0.86D, V): 71.6 lux --> 67.4 lux, 5.8% loss.

              ~50 meters ahead, right edge of road (1.5D, 2R): 79.3 lux --> 77.6 lux, 2.1% loss.

              ~75 meters ahead, right side of road (0.6D, 1.3R): 61.9 lux --> 60.0 lux, 3% loss.

              The ambient temperature was about 75 degrees, so nothing like a car's engine bay, which would easily see triple digit temperatures. So even after ~20 minutes at ambient temperature, the Sealight exhibited intensity losses in the order of 2 to 5% when it really couldn't afford to lose any more intensity. It's already being creamed by the halogen bulb it replaced, and one of its numbers falls below the legal minimum. I suspect that we'd see double-digit losses in intensity had I tested these Sealight LED bulbs in a more realistic environment (hotter environment) to better simulate operation in an engine bay.

              The next steps for testing are to use an actual projector from a car instead of a standalone Hella module. The Hella H9 projector I'm using is actually meant for retrofitting (into classic cars). I want to make my results more applicable to end users by using a projector actually found in their car, but I suspect results will largely remain the same: the Sealight simply doesn't have enough "oomph" to compete with a handicapped H9 bulb running at 75% capacity.

              Last edited by projector-head; February 21st, 2020, 04:35 AM.

              Comment


              • #8
                Edit - Sorry, started to post a question earlier, got distracted and didn't finish. Off-topic question, anyone else not able to see their text when typing a response due to the space-themed background showing through the text edit box? Same thing happening in my mobile browser, as well.

                Back to what I intended to post:

                1-Glad to see some detailed testing being conducted, thank you for being clear and thorough!
                2-Glad to see some testing being done with Hellas 90mm projectors. I'm considering my options for a halogen projector swap project (or the 50mm, or the 60mm units), Hella doesnt even give their "birds-eye" beam shots for some of their different projectors - makes trying to compare then... difficult without shelling out and getting subjective impressions, and returning whats not used.
                3- Finally the question: This being a projector meant for an H9 halogen bulb, is there any reason to think that a high focus +130/+150 H11 might outperform the much higher output H9?
                4- Do you plan to test this projector with any HID burners?
                5- Do you plan to test any other Hella halogen projectors in a similar manner, actually comparing intensities at the different test points? Would you be willing to do so if a projector was donated?
                6- Do you have any other projectors from the aftermarket (e.g. commonly retrofitted like the morimoto mini h1, etc) to compare? I have not liked the idea of retrofitting primarily due to not being able to easily make any semblance of an information driven choice.
                Last edited by cibiefan; February 23rd, 2020, 01:15 AM.

                Comment


                • #9
                  Originally posted by cibiefan View Post
                  llll
                  Hmm?

                  Anyway, did some more testing with the Sealight LED in a Stanley projector. No surprises here.

                  Comment


                  • #10
                    DaY dOnyou think the incorrect/low placement of the LED chips on this bulb was an attempt to direct more light at the base of the reflector bowl and increase distance projection?

                    Comment


                    • #11
                      Highly likely not. In projector case, vertical cross section is generally close to ellipsoid, and horizontal cross section is generally close to paraboloid. Placing chip outside of original filament near by area will result ellipsoid surface to fail focusing ray at other focus point, which should be aligned to aspheric condenser lens back focus distance.
                      Then it only scatter ray at vertical axis. Long story short, it turn into glare and excessive foreground illumination.
                      Horizontal axis affect is fail to create parallel beam into lens surface, This can be either observed as unintentionally slightly wider beam or unintentional overlap at beam center to create excessive foreground illumination.

                      Having emitting area of LED closer to bulb base than filament position, it is studied to cause stronger glare.

                      Then, other important aspect to observe is thickness of center PCB ( emitting surface distance each other)

                      Optical design is very sensitive.
                      If LED bulb at least does not have emitter position match to filament reference distance, that is simply not good design.

                      Comment


                      • #12
                        You can fix the background issue by clicking on the top left-most button on top of the reply box. The icon looks like a piece of paper with "<>" on it. Once you click this button, the background should return to its normal, white color.

                        1-Glad to see some detailed testing being conducted, thank you for being clear and thorough!
                        2-Glad to see some testing being done with Hellas 90mm projectors. I'm considering my options for a halogen projector swap project (or the 50mm, or the 60mm units), Hella doesnt even give their "birds-eye" beam shots for some of their different projectors - makes trying to compare then... difficult without shelling out and getting subjective impressions, and returning whats not used.
                        Glad someone found it helpful!

                        I think you might be better off avoiding the smaller units. Headlamp performance tends to scale with the size of the unit. I think there is a reason Hella "hides" the isolux contours for its smaller projectors while it has no issue showing everyone the isolux contours of its, say, 90mm xenon projectors.

                        3- Finally the question: This being a projector meant for an H9 halogen bulb, is there any reason to think that a high focus +130/+150 H11 might outperform the much higher output H9?
                        That's a huge reason I purchased this projector first, before trying to obtain other projectors.

                        I figured that I could not only test H9 bulbs but also the various high-performance H11 bulbs in the same unit. Plus, being a projector designed around a H9 bulb, and therefore being a projector expecting to see ~2100 lumens, I hypothesized that it would be easier to suss out the pretenders from the contenders when I insert ~1350 lumen sources (H11 bulbs).

                        Now, as to whether a H11+130/+150 bulb would outperform a H9 bulb, here are my preliminary thoughts, without any data:

                        1) The H11 bulb has a blacktop; the H9 doesn't. I believe ~10% of the light exits the distorted tip of a halogen bulb. I'm not sure what a projector "does" with this light, if anything, because its leaving a distorted end of a bulb so the characteristics will obviously vary bulb-to-bulb, manufacturer-to-manufacturer (like Philips H9 vs Sylvania H9). Also, 10%, or ~200 lumens, doesn't account for the remaining ~600 lumen gap between a H11 and H9 bulb.

                        2) I read a test on Facebook that tested H9 and high-efficacy H11 bulbs in a H11 projector. The H11s came pretty close, like within 10% at many test points, of the H9 bulb, despite being down by ~800 source lumens.

                        https://www.facebook.com/AutomotiveL...38490493252366

                        3) With the above information in mind, I believe that a high-output H11 bulb in this Hella H9 projector will likely perform fairly closely to a H9 bulb, at least in terms of the hotspot. The H9 probably will have a leg up in other regions, like the edges of the beam. This is basically what the link above found.

                        4- Do you plan to test this projector with any HID burners?
                        Yes...that's actually another huge reason I purchased this projector first. The projector lens says "H9/D2S." From what I've read online, many people have taken similar statements on other headlamps as meaning that the headlamp can be used with HID and halogen bulbs interchangably.

                        This is obviously not true, and NHTSA actually explictly told a manufacturer NOT to do this (print "H9/D2S" or similar) since it would create confusion.

                        The only issue is whether my power supply can actually power a HID bulb's ballast and igniter. Hopefully it can. I plan to find out sooner or later. The limiting factor right now really is money--I'm already several hundreds of dollars deep into this project.

                        5- Do you plan to test any other Hella halogen projectors in a similar manner, actually comparing intensities at the different test points? Would you be willing to do so if a projector was donated?
                        Absolutely, I would love to test all sorts of projectors (and reflectors) and I would be willing to accept donations, both temporary and permanent . I would obviously be willing to return a projector if one wants it back.

                        6- Do you have any other projectors from the aftermarket (e.g. commonly retrofitted like the morimoto mini h1, etc) to compare? I have not liked the idea of retrofitting primarily due to not being able to easily make any semblance of an information driven choice.
                        I don't have any personal experience or personally-obtained data on the Mini H1, but I did stumble across this a while ago, and it might interest you:

                        https://www.reddit.com/r/cars/commen...tor_retrofits/

                        dOnyou think the incorrect/low placement of the LED chips on this bulb was an attempt to direct more light at the base of the reflector bowl and increase distance projection?
                        Somewhat related: I read that another issue with LED chips is that unlike a glowing filament, the intensity of light emitted from a LED chip falls off dramatically as you go off-axis. I read that the emissions peak in the range of about -15 to +15 degrees, with 0 degrees being looking at the chip dead on. A filament, however, no matter whether you look at it on-axis or way off-axis to it, has roughly the same intensity. I read that the center portion of the reflector, responsible for distance illumination, is obviously at a pretty extreme angle relative to the chips on a LED bulb, and this hurts the intensity of light that strikes the center rearward portion of the reflector.
                        Last edited by projector-head; February 25th, 2020, 11:02 AM.

                        Comment


                        • #13
                          >you might be better off avoiding the smaller units.

                          Makes sense, but I had seen in passing some comments about the 60mm projectors being a newer design, and supposedly more efficient than the older 90mm's. And there's not much info on the 120mm bi-halogens either. It would make sense that a larger lamp format would allow for higher performance, but that doesnt necessarily hold true if it was designed for a price-point. And the usual source for Hella news, the gentlemen at rallylights.com just suggested looking over the myHella site. And that "all the hella projectors work great with HID". Not sure I want to go that route, thank you.

                          >This is obviously not true, and NHTSA actually explictly told a manufacturer NOT to do this (print "H9/D2S" or similar) since it would create confusion.

                          Interesting. I had seen those marking in the marketing photos, I had assumed it was just a stock image(s) that were being used for either lamp. Perhaps its just to help identify parts as they're being assembled?

                          >H11 testing

                          Cool, looking forward to see what the results may be.

                          >HID testing

                          Going to laugh if it somehow ends up with a compliant beam. Would be a great trick to see (and I winder if there would be a quiet increase in the price for the halogen 90 lamp?) And (Big) If it ends up with a compliant beam -and- is shown to be significantly higher performing than the h9, I may end up going that route for a swap. Time will tell.

                          >Mini H1 test

                          I had seen that linked on reddit. Disappointing but ultimately not surprising. It also reinforces the fact that a good headlamp is a good headlamp regardless of lightsource type (which somehow seems to be missed a lot of places).

                          >LED emission angle

                          That is probably a big part of it. Domed LEDs would have a wider/more even distribution of light across their viewing range, but the dome magnifies the apparent die size, which reduces the apparent intensity, which reduces throw. I had theorized here and elsewhere that manufacturers of these drop in may be aware of this, but attempting to make up for it by shifting more light closer to the base of the bulb, but as DaY ppinted out, that still wouldnt necessarily result in a correct reflecting angle from the base of the bowl to the front lens. So, either more light lost inside the housing, more glare, or more light flooding the foreground. All useless.

                          Comment


                          • #14
                            I just realized that you are using the Hella 90mm H9 projector. It doesn't have that great an output. I've used them with regular H9 Bulbs, 35 watt HID and 55 watt HID. In the end, swapping to different sealed projectors with 35 watt HID was yielded much better results.

                            https://ls1tech.com/forums/appearanc...rojectors.html

                            I have more pics in the thread, but here is one that is a direct comparison.


                            Last edited by VIP1; February 28th, 2020, 11:12 PM.
                            www.fadingarrow.com

                            2001 Black Firebird Formula 5.7L V8 --- 75th Anniversary Package
                            Headlights: Sinolyn 3" D2H Projectors with 35 watt HID

                            2013 Blue Mustang GT 5.0 V8 Premium
                            Headlights: Stock D3S Bi-Xenon

                            Comment


                            • #15
                              VIP1 thanks for the link to that thread. Did you attempt to shim the HID burner in the 90 and 60mm projectors to see if that altered the beam significantly?

                              Comment

                              Working...
                              X