headlight failure (H7)

  • Thread starter Thread starter Mike Marlow
  • Start date Start date
M

Mike Marlow

Dan K said:
Well, I actually did something about it this weekend. This is what I
measured/did:

With 2002 XG350 running:
V battery = 14.2 v
V at headlight = 13.43 v
V headlight + to battery - = 13.55 v

This tells me there is a 0.12 volt drop in the ground wiring (seems
reasonable), and a 0.65 volt drop in the positive side which probably
includes wiring, a fuse, and a switch of some sort (relay, transistor, or
both).

So I placed a 10 amp diode in the passenger side headlight circuit. It
now reads 12.8v (headlight + to battery -). This should imply that the
voltage measured at the headlight is now around 12.68v (I did not measure
this). As I installed the diode and placed it in and out of circuit while
watching the headlight intensity, you could tell the difference, but only
because you had the 2 to compare. Its virtually impossible to tell by
looking now that the passenger side headlight is a little dimmer than the
driver side. My plans are to run this for a week or two and then install
a diode on the drivers side. I will then see how long it takes to burn
out a headlight or two. If they still burn out in under 3 months I might
try a second diode in each side - which should drop the voltage at the
headlight down to right around 12.0 volts.

Good stuff Dan. Thanks for posting this. Why bother waiting a week though?
That won't really show you anything, so why not just go ahead and install
the second diode now and start the clock ticking?

-Mike-
 
Well, I actually did something about it this weekend. This is what I
measured/did:

With 2002 XG350 running:
V battery = 14.2 v
V at headlight = 13.43 v
V headlight + to battery - = 13.55 v

This tells me there is a 0.12 volt drop in the ground wiring (seems
reasonable), and a 0.65 volt drop in the positive side which probably
includes wiring, a fuse, and a switch of some sort (relay, transistor, or
both).

So I placed a 10 amp diode in the passenger side headlight circuit. It now
reads 12.8v (headlight + to battery -). This should imply that the voltage
measured at the headlight is now around 12.68v (I did not measure this). As
I installed the diode and placed it in and out of circuit while watching the
headlight intensity, you could tell the difference, but only because you had
the 2 to compare. Its virtually impossible to tell by looking now that the
passenger side headlight is a little dimmer than the driver side. My plans
are to run this for a week or two and then install a diode on the drivers
side. I will then see how long it takes to burn out a headlight or two. If
they still burn out in under 3 months I might try a second diode in each
side - which should drop the voltage at the headlight down to right around
12.0 volts.

I'll post updates as they occur.

Dan
 
Mike Marlow said:
Good stuff Dan. Thanks for posting this. Why bother waiting a week
though? That won't really show you anything, so why not just go ahead and
install the second diode now and start the clock ticking?

-Mike-
The thought was that if something weird occurred, like the diode went out
for some strange reason, it would not take out both headlights. As soon as
I convince myself that this is a stable configuration, through multiple
startup - drive - turn off cycles and a number of hours of run time, then
I'll go ahead and do the other side. What could possibly go wrong with a
simple diode I hear you ask? I've been an engineer too long and seen too
many instances of the impossible happening to ever ask that question "what
could possibly go wrong?"

Dan
 
Well, I actually did something about it this weekend. This is what I
measured/did:

With 2002 XG350 running:
V battery = 14.2 v
V at headlight = 13.43 v
V headlight + to battery - = 13.55 v
the voltage drop through a diode
will be about 1 V or less and install
them in series until you get 12 volts.
measure across the filament for 12 Volts.
or use a Zeener 10W diode
for example a 1.5V zeener will give
you almost exactly what you need,
they come in all different available
voltages, such as 2 volt, 1.5 Volt,
1.8 volt....
some are stud mount for more secure
mounting. that would be recommended.
OK make sure it's an isolated stud.
Yes any electrical circuit has a voltage
drop, thats normal.
Well halogen lights are well known for
short life, over-voltage won't help that's
for sure. But a good alternator shop can
adjust the output voltage easily.
you might as well install diodes to get
12volts on both sides now.
you are right, over-voltage will fry the
filaments, no question.
That's funny, the headlights on my accent
have lasted 70,000 miles, 5 years. why
won't Hyundai do something to fix the
bug? Is there Xenon replacements?
 
The thought was that if something weird occurred, like the diode went out
for some strange reason, it would not take out both headlights. As soon as
I convince myself that this is a stable configuration, through multiple
startup - drive - turn off cycles and a number of hours of run time, then
I'll go ahead and do the other side. What could possibly go wrong with a
simple diode I hear you ask? I've been an engineer too long and seen too
many instances of the impossible happening to ever ask that question "what
could possibly go wrong?"

Dan

OK wait a second, how many watts is the bulb?
Low AND high beam...
let me know and I'll try to help here.
using load and voltage we could calculate the
watts load on the diode. You have apparently
overheated the diode. But don't forget there are
industrial diodes that can take the heat, heat sink
included, chassis mount or stud mount.
You have a good idea, just need better hardware.
Also industrial voltage regulators. cheap if
the bulbs go out every 3 months!
Interesting project, can't wait to see how it turns out.
 
Dan K said:
Well, I actually did something about it this weekend. This is what I
measured/did:

With 2002 XG350 running:
V battery = 14.2 v
V at headlight = 13.43 v
V headlight + to battery - = 13.55 v

This tells me there is a 0.12 volt drop in the ground wiring (seems
reasonable), and a 0.65 volt drop in the positive side which probably
includes wiring, a fuse, and a switch of some sort (relay, transistor, or
both).

So I placed a 10 amp diode in the passenger side headlight circuit. It
now reads 12.8v (headlight + to battery -).

Snip

pardon my question, if it has been answered in earier posts, but is the
voltage at the lamp known to be abnormal-too high? If every other of the
1000's of like cars and bulbs are at that figure, maybe heat or vibration or
some other thing more unique to your vehicle is the culprit ? Good luck !

(as an aside, I've changed one headlamp in 2 yrs/60k mi, 06 sonata)
 
Darby OGill said:
Snip

pardon my question, if it has been answered in earier posts, but is the
voltage at the lamp known to be abnormal-too high? If every other of the
1000's of like cars and bulbs are at that figure, maybe heat or vibration
or some other thing more unique to your vehicle is the culprit ? Good luck
!

(as an aside, I've changed one headlamp in 2 yrs/60k mi, 06 sonata)
That's just it, nobody knows for sure what the problem is. There are a lot
of Hyundai cars out there with H7 bulbs that go through them like popcorn.
I own two XG350's (his and hers) that both have had the alternator replaced
once, so that's 4 alternators and the H7 bulbs seem to last 6 months if we
are REAL lucky. So if its an over voltage problem, I suspect its a design
flaw and not due to a failed piece of hardware. I'm use to the old cars
that needed a new set of headlights once in their entire life (of coarse
that was only 100,000 miles)...You never thought to buy a spare
headlight...with the H7's I wouldn't dream of going anywhere without a
couple spare bulbs in the trunk!

So, anyway, this is a first attempt to see if a lower voltage will fix the
problem.

Dan
 
Dan said:
That's just it, nobody knows for sure what the problem is. There are a lot
of Hyundai cars out there with H7 bulbs that go through them like popcorn.
I own two XG350's (his and hers) that both have had the alternator replaced
once, so that's 4 alternators and the H7 bulbs seem to last 6 months if we
are REAL lucky. So if its an over voltage problem, I suspect its a design
flaw and not due to a failed piece of hardware. I'm use to the old cars
that needed a new set of headlights once in their entire life (of coarse
that was only 100,000 miles)...You never thought to buy a spare
headlight...with the H7's I wouldn't dream of going anywhere without a
couple spare bulbs in the trunk!

So, anyway, this is a first attempt to see if a lower voltage will fix the
problem.

Dan
One thing to remember, while we usually use the term 12V battery, it
really means 12.6V, no-load voltage battery. That's the normal voltage
for a fully charged lead-acid battery under NO-load condition.
With the car running, the alternator has to provide a higher voltage to
recharge the battery, so the voltage should be 14.4V plus or minus 0.4V
depending of the load.
The bulbs manufacturers take this into account.
For what you measure, your car is at the low-level of the range, so
over-voltage does NOT seem to be the problem.
Have you tried another brand of bulbs ???.
BTW, Elantra GTS 02, Santa Fe LX 03, both still original lamps .... so
no, I do not have any problem.
 
The voltage is okay in this case (14.2V). That's normal alternator output.
In my opinion, the culprit is the H7 bulb itself. For whatever reason, it
seems to burn out quickly.

Dan's theory, however, is that if we reduce the voltage to the bulb but
still keep it high enough for proper lighting, bulb life may be
significantly increased. Hopefully, he'll report back when he's been able
to conclude something.
 
I used to have these halogen headlight bulbs burn out very rapidly,
until I learned that you cannot touch them with your bare hands when
you are installing them, as the natural oils on your fingers will
cause them to fail prematurely. I started wearing cotton pallsbearer
gloves when I change my headlight bulbs and the early failure problems
went away completely.

My .02 cents, ymmv

me
 
Although I never checked the voltages across the headlights on my 2001
XG300, the 14.4V or so to the battery on charging sounds like mine.
That said I still have the original battery & I have never yet, touch wood,
changed a headlight on that car! The only bulbs to burn out have been the
licence plate marker & the glove box light. The other consideration here is
my car is Canadian & has always had daytime running lights on (all lights on
but reduced intensity during day).
BCinBC
 
For what its worth I work in medical electronics and I never buy any lamp or
battery from a certain country. No names mentioned but I,m sure you could
guess.(not Korea)
I,ve done comparative studies of different brands to
justify it. Even Panasonics who used to make excellent batteries shifted
manufacturing to this country and now they are just as bad as the others.
John
 
jusme said:
I used to have these halogen headlight bulbs burn out very rapidly,
until I learned that you cannot touch them with your bare hands when
you are installing them, as the natural oils on your fingers will
cause them to fail prematurely. I started wearing cotton pallsbearer
gloves when I change my headlight bulbs and the early failure problems
went away completely.

This is a very well known issue with any high temp bulb. Perhaps there are
people who still don't know about it and are handling their bulbs
incorrectly, but the real problem is that for the people who do know about
it, the problems with H7 bulbs persist. Your bulbs will fail also, unless
you've found a brand that is a bit longer-lived, in which case we'd all sure
like to know about that.
 
For what its worth I work in medical electronics and I never buy any lamp
or battery from a certain country. No names mentioned but I,m sure you
could guess.(not Korea)
I,ve done comparative studies of different brands to
justify it. Even Panasonics who used to make excellent batteries shifted
manufacturing to this country and now they are just as bad as the others.
John

Well..... that all by itself is useless, unless you tell us where you have
found good bulbs.
 
Mike said:
This is a very well known issue with any high temp bulb. Perhaps there are
people who still don't know about it and are handling their bulbs
incorrectly, but the real problem is that for the people who do know about
it, the problems with H7 bulbs persist. Your bulbs will fail also, unless
you've found a brand that is a bit longer-lived, in which case we'd all sure
like to know about that.

FWIW, I've had Hella Optilux XP bulbs in my Elantra for over 8 months
now, which is longer than any other bulb I've tried.
 
Back
Top