Frustration with HID Headlights

[furinax]

Just spent 20 minutes trying to debug an electrical issue. This is getting quite frustrating. I'm on HID Headlights with Phillips ballasts.

The driver's side headlight will sometimes not turn on and I have to hit it with a highbeam to get it to turn on, but recently it just stopped working all together. By swapping the ballast + bulb I can see that neither the ballast nor bulb is bad.

Last time it took me an hour to diagnose and it turned out to be a corroded fuse. I guess it's due to all the exposure to water now, even though my car is under a cover. However, this time I cleaned the fuse, swapped my relays, wiggled some contacts... nothing is working. Next I may have to get a multimeter to check the circuit coming from the fuse box.

Does anyone have any tips or tricks for debugging faulty HID headlights? My 2007 MS3 had this same issue and I just replaced the whole system. I don't think that would work in this case as I think there is now a break somewhere from the fuse box to the ballast.

I wonder if LED kits are this complicated.

 

[jwmelvin]

If you can swap the L/R assemblies without a change in behavior it seems it has nothing to do with the headlights themselves. Do you use a relay harness? Perhaps the increased current of HIDs has stressed your cars OE harness.

 

[furinax]

Well after some debugging I am still quite frustrated. Swapped everything I could to find out the source. Ballasts, bulbs, fuse box fuses, HID relays, HID fuses... all seems to work on only the passenger's side. Sprayed my fuses and connectors with contact cleaner too...

I measured the voltage at the input to the ballast and it reads 11.8v at the passenger side and 11.5v on the driver's side. I figure something was corroded so the driver's side has more resistance (more wiring) and hence a lower voltage. Maybe the battery is too weak? My battery was 10 years old so I got a new one and it still didn't fix it.

At this point I can only think of replacing the wires... or doing some resistance measurements along the wires to find the bad portion. I have never done this before and just have a simple multimeter with me. Any other clues I can look for?

 

[kennyvb]

Cutting out all the aftermarket stuff, does a halogen bulb work on the driver side?

 

[Old Guy]

Your avatar suggest that this is a 2002 or later car. So, do you have a pre 2002 with a 2002 face lift applied or do you have a 2002 and later on which you have ditched the OEM HID system for the Phillips system? It makes a difference because I believe the headlight wiring is different between the pre and post face lift cars.

If you have a 2002 or later vehicle I expect the replace with halogen test is not a viable option.

 

[furinax]

Yeah I have the 02 conversion with the Phillips ballasts. I just wish I had a garage so I could do it in privacy.... Everyone thinks I'm some sorta scared guru if they only knew....

 

[Old Guy]

First, a voltage of 11.5 - 11.8 volts is probably reasonable assuming the engine was not running at the time. I am going to assume that the Phillips system is like some other HID systems that I have seen which get power through the existing vehicle headlight socket. If you have a system with separate non stock relays and fuses for the ballast, then this may not be of much use. I suggest that you try kennyvb's suggestion with an original style halogen bulb in the low beam socket. If the halogen bulb lights up, then, if you confirmed that the HID bulb and ballast are operational, you may have a mating problem between the plug on the Phillips system that fits into the socket on the wiring harness. Checking for damaged contacts in the Phillips plug or the socket on the wiring harness would be appropriate.

If the bulb does not light up, remove the bulb from the socket and with a voltmeter, check to see if you still have 11.5 volts at the socket terminals (the black wire to the socket should be ground and the red/yel wire should be +12v).
- If you don't get any voltage, check from the red/yel wire to a good body ground. If that gives you 12 v then the ground connection to the socket is faulty. If measuring between the red/yel wire and a good body ground gives you no voltage, then you have an open circuit between the headlight socket and the headlight fuse.
- If you do get voltage at the socket when you remove the bulb, then you have a high resistance connection in the circuit someplace. When there is no current flowing in the circuit (bulb removed) there is no voltage drop across the resistance and when the bulb is in it draws enough current to cause a voltage drop across the resistance which prevents the bulb from lighting up. If that is the case, again check the socket for damaged contacts. My next favorite pick would be to check the condition of the body connection point for the black wire coming out of the headlight socket. If that yields nothing, then you have to trace the black wire from the socket to the ground connection and the red / yel wire back to the fuse box looking for damage which is a royal pain because they are in a wiring loom.

 

[furinax]

Spent some time trying to debug it again this morning. It really is mind boggling.

Voltage measured at the contacts to the ballast are okay. With the new battery, it's 12.4v. 12.8v and .4v for each lead to ground (the radiator). The other working side was the same.
I connected a new ballast to the wire and nothing happened. The question is, why does neither the new nor old ballast not get the input voltage? I figure it can only be the connection corrosion but it looks good and the voltage reads okay. The new ballast came with some connectors, if I was feeling risky I could solder them on and have a new connector.

Here are some pics.

The working side, showing how things are supposed to look. I can circle the extra fuses and relays that are next to the fuse box but I don't think they are a problem. The relays both click on when I turn the lights on and the voltage reads okay on the other end.

The broken side. The connector on the bottom right goes to the ballast input. It reads okay voltage-wise but doesn't turn on the ballast. I think that is the culprit, but no way to fix it.

My old ballast for reference.

Due to the symptoms I can only guess it is corrosion somewhere. Maybe it is in the wire. It has always been hard to turn on: have to wait for it to warm up, or blink the highbeams to get them to work. Since I park my car outside with a cover I figure all the rain seeps into the electronics and causes corrosion over time.

Anyway, thanks for your help. I'll find some more time to look at the contacts like you suggested, Old Guy.

 

[nikk0bacardi]

Have you tried reversing the polarity?

Sent from my SM-G955U using Tapatalk

 

[furinax]

Yes, reversing the polarity did not help. I think the resistance of the wires or the switched circuit has increased to a point where it is too weak for the ballasts. It may be high time to replace the wires.

I wired the new HID kit directly to my old battery directly and it works fine. The new HID kit has a bunch of adapters and even has wires endings I can solder on. In relation to the old low beam plugs: does anyone know what the below plug is for?

Currently my Phillips ballast is connect to what I would expect to be the proper lowbeam connector, except it has fuses and relays thrown in for safety reasons I guess:

 

[furinax]

OK revisiting this thread with my solution which I found a few months ago. Essentially everything was working but one of the grounds went bad. My driver's side HID ballast was hooked to the chassis but if you measured the chassis point to another part of the chassis there was no continuity. Likely the rain had caused corrosion between the connection and the bolt which was in the chassis. I just moved it to a different location and it worked fine.

 

[jwmelvin]

OK revisiting this thread with my solution which I found a few months ago.

Thank you for the update.

 

[Old Guy]

OK revisiting this thread with my solution which I found a few months ago. Essentially everything was working but one of the grounds went bad. My driver's side HID ballast was hooked to the chassis but if you measured the chassis point to another part of the chassis there was no continuity. Likely the rain had caused corrosion between the connection and the bolt which was in the chassis. I just moved it to a different location and it worked fine.

I am guessing that you were using a non-sanctioned ground location (i.e not one of Honda's grounding points)? However, if you were using a sanctioned ground, this might be a wake up call that some inspection type preventative maintenance on other ground locations might be appropriate, particularly if you are suffering from exposure to a lot of wet conditions. Although Honda has taken preventative measures, the NSX with its aluminum body is more susceptible to galvanic corrosion at these connections. Also, certain items such as the igniter module do seem to have an anecdotal history of suffering from flakey operation caused by poor grounding. Inspecting the ground connectors and perhaps cleaning and treating them with a little Oxgard grease might be useful preventative maintenance.

 

[furinax]

If you look at the image below, it was the yellow circle on the right side which was the issue. If I had did some logical deduction with a multimeter, I could have figured it out. After testing all the gear to be okay it ended up being just a simple faulty ground.