My own ramblings on the S2000 headlights.
If you blow a fuse on one of the low beams (or just remove it), the light will not go out like you would expect, but instead flickers like a wimpy strobe light. You'd think if the fuse is blown, the light would get no power at all. Not.
On top of that, both high beams will come on, even when they're switched off, but only at partial brightness. It will make you say "Hmmmmm" and maybe "holy crap" if you don't know what's going on. So what is going on?
Honda had a better idea, so they have the lights wired so the low beam relays provide positive voltage to both the low and high beams. Then they wired the high beam switch to supply the ground to the high beams. Why? To confuse us of course. And it makes for a cool light show if a fuse blows.
The result is that when an open circuit occurs (like a blown fuse) in the +12v side of the low beam circuit, current from the other low beam relay feeds back through both high beam elements to the low beam with the blown fuse. So all 3 lights are in series with the 12 volt power supply, and only get partial voltage. This makes the high beams light up dim, and the low beam flicker as it tries to start up on inadequate voltage. Cool.
Just for grins, I pulled out my handy-dandy light meter the other night and tested my Honda's headlights against my wife's ES300 with quartz-halogen headlamps. The light meter I used is a cheapie ($100 digital) I bought 10 years ago, so there's no telling how accurate it is, but since I used the same meter and methods on both cars, at least the relative readings should be fairly reliable.
I tested the lights at night (duh) in a poorly lit neighborhood, so there was no outside light sources to give false readings. These measurements are a little subjective, especially the "Average Coverage" reading. Just moving the light meter an inch could change the reading by 10%. If you look at the 25' measurements, the Halogens had a 500% variation between minimum and maximum brightness over a 3' vertical area, whereas the HID lights only varied by 50%.
It was very obvious that the HID's had a much more even light pattern, as well as being considerably brighter. Incidentally, both cars seemed to have peak brightness at the 25' foot distance at about knee height. (how's that for scientific method? was that the low knee or the high knee?)
I don't know if my light meter is more or less sensitive to the yellow or blue spectrum, so there could be errors there too. Anyway, here's the results. fc = foot candles by the way
1 Foot from headlight | 25 Feet from Headlight | |
HID Peak Brightness | 5,000 fc | 45 fc |
Quartz Halogen Peak Brightness | 1,300 fc | 17 fc |
HID Average Coverage over a 3' x 3' area | 30 - 40 fc | |
Quartz Halogen Average Coverage over a 3' x 3' area | 3 - 10 fc |
Just as a reference, here are some more measurements with the same light meter:
Reasonably well-lit office | 20 fc |
Daylight very overcast late afternoon | 700 fc |
Daylight on a sunny clear day | 10,000 fc |
Based on my interpretation of the shop manual, you can use the following tests to get a feel for how your headlights are aimed. These tests are easy to do at night, if you can find a reasonably level road.
With our lights being so much brighter than other cars, headlight alignment is much more critical. Plus, unlike regular lights that have a light pattern that gradually gets brighter as you approach the center of the beam, ours have a more defined pattern that abruptly changes from almost no light outside the beam to "damn, turn those friggin' high beams off you idiot" as soon as the beam hits someone's rear view mirror.
After I noticed my headlights seemed to hit lower on the truck in front of me as I got closer to him, I decided I should check my headlight alignment. I found the passenger side light was actually pointing up! No wonder people seemed annoyed when I got behind them. Then again, maybe they just recognized me.
Here's an easy way to align your headlights. Any body who has used concrete blocks and string to do a four-wheel alignment will recognize this get-up. The basic principle is that you don't need a flat or level surface to align your headlights. You only need a way to transfer the car's position in vertical space from the front of the car to your 'test wall' 25 feet away. As long as you know the height of your wall relative to the height of the car, the alignment will work.
String works great for this, but on this occasion, I decided to use my new high tech portable battery-operated "laser alignment" system, (i.e. a $6.95 laser pointer and a squeeze clamp).
Here's how it works: (there are pictures at the end)
Note: Each headlight creates a light pattern that is lower on the driver's side and higher on the passenger's side. See photo B. This is to make it easier to annoy people on your side of the road, as opposed to oncoming traffic. You should adjust so the top of the light pattern (passenger side) is 2.1" below the headlight height.
One more note I should make here. These alignment instructions are sort of based on the shop manual instructions, which as far as I'm concerned are almost useless. The shop manual shows a light pattern on the wall that looks nothing like the actual headlight pattern, so I made the assumption that when the manual says "the lights should reflect 52 mm (2.1 in.) below (B) headlight height" they meant that the top of the pattern should hit the wall 2.1" below the headlight height. Because of the unique light pattern of these lights, any other interpretation would cause the passenger side of the pattern to point up. I don't see the advantage of illuminating overpasses instead of the road in front of me.
(A) Laser at rear wheel | (B) Headlight pattern on wall |
|
|
(C) Location of adjustment hole | (D) Close-up of toothed adjuster wheel, inside adjustment hole |