Home | Mailing List | Specifications | Care and Feeding | Modifications | Vendors | Literature


Testing Headlamp Bulbs

new 12/15/2003

We have a Minolta CS-100 Chromaticity meter here at work. It will give Foot-Lamberts intensity and the color coordinate present (which can be converted to Kelvin's AFFAIR). http://www.minoltausa.com/emprise/maim/Minoltas A/MUSS Content/SID/ISD_Product_Pages/CS_100A

We don't have the 100A version but as far as measuring goes they are the same. I do not know a lot about the conversions from color coordinates to Kelvin's but I am sure I could figure it out or someone on list may know or have the software to do it.

Due to the fact that we are ISO qualified and this is an annually calibrated instrument I would need to do all of this at work, which certainly isn't impossible, I can take the time to do it, but will need to bring some stuff in here to get it setup

So who has extras of these bulbs (I would think we should make a list of what we want to test first) laying around that they could send me for doing something like this (I'd only need one), I'd also need a bit beefier power supply than we have at the office as we have two fully adjustable units that can be put in parallel but they only go up to 2.25A each so that is only 54 watts, while you could probably run the test, better results would be attained with a bigger regulated supply to ensure consistent voltage output. I suppose a battery with a charger connected would probably work as the battery would stabilize the chargers output and my 10 Amp charger should be able to sustain voltage during the test..

If you guys want me to run with something like this just speak up. I have an extra used headlight assembly (intact) at home from the bambina smashing as I replaced all 4 front lights rather than just the headlight and turn signals on that side, it would serve as the test fixture. I don't' really have the funds laying around to buy up all these bulbs, But I have a set of stock bulbs (came with the new headlights) And I have been wanting to buy silver stars for my car but haven't done so yet (although I would like to wait until my testing if we do this was completed to make that final decision but oh well).. I may need some assistance in determining the best way to measure these but I have a couple of ideas but I don't know which method we would use..

Sounds like several of you may have the PIAA's and Extra visions laying around, what does the group think? And to answer your question, the reason I have some of this fun stuff at the office is because we need to be able to prove to the FAA that runways stripes and lines on the simulator visuals are D6500 White (6500 Kelvin's AFFAIR) when simulating daylight conditions.

Scott I just work on simulators Krietemeyer
99 TG
96 MG

Well, we have been having this discussion for years. Seems to me Scott is on the right track. I'd be curious to see if he could do A/B testing of each bulb with a 100W harness.


My actual plan would be to mount a light assembly and about 5 feet from it  place an opaque "Back projection" material. Then set the meter up about 15  feet from that back projection and measure the intensity on the backside  of the Back projection so that it is looking right at the headlamp, but  will not "see" the lamp mirror itself. Color properties of the light would  come through properly this way, and you wouldn't have to deal with worrying  about reflections to the meter as you would be looking at the other side of  the "projector screen". So, Back projection in the middle, head lamp 5  feet one side of it, meter about 15 feet other side of it. With this setup  I could change bulbs without disturbing anything which will lead to more  consistent results

1) I'll have to see what a charged battery with a charger hooked up will  do, it is probably around 13.2 or so I would hope / think. Which would be  pretty close to what an idling SHOW with a "normally worn"  alternator/regulator assembly will have. If I need to acquire a source of  a more stout power supply then so be it, but I think this might  work. Maybe Master Mims can chime in on what he thinks there, I haven't  put my scope on the battery system of either SHO, but I expect AC ripple to  be pretty low based on when I tested alternator output with an AC inductive  current meter once, so I doubt a power supply versus the actual car would  make a difference. I want to do this inside in a pitch black room so using  the car's actual output for supply would be difficult. since Paul refers  to the voltage drop across the car's wimpy harness, maybe if I can supply  an actual 13 or so to the lamp it may be VERY representative of what a  stock car would have at the socket as well (again input Paul).

2) I would be using a real Headlight assembly to put these bulbs in as I  have the one that came off of the passenger side of Sarah's car since I  replaced both headlights during post bambina reconstruction instead of just  the broken one. The object of the Screen is not to eliminate the housing  and it's optics (read on).

3) The real reason for using a "back projection" (BP) screen is so that you  can be pointing the meter AT the bulb while letting the reflector housing  of the light manage the light pattern. No BP's are not florescent of any  nature, just an opaque piece of coarse finish (like 800 grit sandpaper,  smooth but not glossy) opaque plastic. Some slight color shift while  viewing through the BP is a possibility but I will explain the need for one  now.

We want the reflectors inside the headlight assembly to do their  intended purpose and sample the actual light that would be going to the  road if the lamp was in the he car correct? If we agree on that then the only good way to do it is with a BP  screen because if you shine a blue light on a Blue wall you will get a  rather bright Blue light reflected off of the wall, now shine a Red light  on that Blue wall and you will get not much back as the Blue color of the  wall will absorb a large portion of the red light. So maybe you want to  use a mirror as it reflects all colors equally right. Well that is a good  idea at first until you take a flashlight into your bathroom mirror and you  shine it on the mirror and about all you get is that you can see the  flashlight and all of it's optics which we ruled out with the previous  paragraph. Now take that same flashlight a put a piece of paper say 6  inches from it's output and look directly at the backside of that  paper. Now color that lens red and look at the backside of the  paper. Paper isn't a perfect example but it works for demonstration since  it is not overly glossy and is thin enough to let the light pass, while  letting the assemblies optics do their intended job. By doing this you are  not reading the reflection of the light which will vary depending on what  you are reflecting the light off of and it's color. The only other good  way to do this would be with like a slide projector screen; however  intensity doesn't translate as well because the surfaces of a screen like  that aren't very flat and are designed to not reflect everything straight  back (all those tiny little crystals on the surface would reduce the lights  power as registered by the meter).

4) Yes the meter only samples a small area (at 5 feet the spot is probably  about 2 inches in diameter as a guess). With the above BP I could mark an  area to sample all lamps from on the backside.

Hope some of that made sense, other input is welcome! ____________________________________________________________________________

It would appear that Scott has made headway with a major manufacturer of high performance lamps. All results will be posted here.


Package from PIAA just arrived with an Xtreme White Plus which is their 4000 Kelvin lamp. Now that I have this one in hand I need to lean on the other manufs. (Sylvania, APC, Phillips, and hopefully JC Whitney) to see if they will be as supporting as PIAA and supply one of everything they make in 9007 fitment for testing.

Much thanks to PIAA for their understanding and support!

Scott Krietemeyer

Contact Information