Best headlights for 9004 bulbs? '95 Mazda Mx6

[passive101]

I have a 1995 Mazda Mx6. The bulbs are little 9004 bulbs which have very poor output. I have cleaned the plastic housings for them so they are pretty clear now, but the light output is still very poor.

What are the best bulbs or solution I can do next to installing european HIDs that were available? (I won't be able to afford those for a year or so). I'm not looking for blue lights. I wanted a pair of Phillips extreme lights, but they do not make them in 9004s.

I am interested in a pair of HIR bulbs if anyone makes them to fit my vehicle. I can upgrade the wiring harness I'm sure if they require a bit more power, but they'd have to be able to fit into the headlight housings.

 

[-Virgil-]

The 9004 system was engineered by Ford with heavy priority on low system cost and long bulb life. It's very difficult to wring good performance out of it unless you have a very large headlamp (such as the giant ones on the '86-'93 Volvo 240) and even then, you can't get anything resembling precise beam focus.

The best 9004 bulbs are the GE Night Hawk 9004NH and the Philips VisionPlus 9004VP. Amazon.com has the Night Hawks on sale.

Avoid Sylvania Silverstar, Wagner TruView, PIAA, Hoen, and all other brands of blue-glass "extra white" or "xenon matching" bulbs. They are a scam; despite the advertising hype, they reduce headlamp performance rather than increasing it (because the colored bulb glass blocks light that would otherwise reach the road).

You would also want to put in heavier-gauge headlight wiring with relays, to feed full power to the bulbs. There's another thread with info on that topic over here.

There are no HIR bulbs for 9004.

 

[passive101]

Where would I purchase an upgraded wiring harness for my car with fuses?

 

[-Virgil-]

Stern might have one.

 

[passive101]

How much brighter would my headlights be by upgrading the wiring that goes to them for little 9004 bulbs?

 

[-Virgil-]

Bulb output changes exponentially to the power 3.5 or so with voltage change, so a small increase in voltage gives a large increase in beam power (and a small decrease in voltage gives a large decrease in beam power).

For a 9004 low beam rated 700 lumens at 12.8 volts, operating it at 12 volts (94% of rated voltage) will give you ~560 lumens (~80% of rated output). Plus the filament luminance will drop like a rock, so beam performance suffers even more than just the output number would suggest.

Go the other direction (run the bulb at 13.6v instead of 12.8v, that is 106% of rated voltage) and get 865 lumens (124% of rated output) plus higher filament luminance.

 

[passive101]

I sent him an email as it seems he doesn't have a way to order from his website.

 

[Diesel_Bomber]

Stern's a good guy, and knows his stuff. He believes in a personal touch and selling you what you need and what will work best, instead of what will put the most money in his pocket for the least amount of his effort expended. That's why there's no way to purchase directly from his website. He assumes, quite correctly, that he knows more about the subject than most of the people he's selling to. He'll want to speak with you and make suggestions.

 

[passive101]

There doesn't look like there is much I can do with mine. I can get an upgraded wire harness for 150 dollars and replace the bulbs, but I will get a minimum return from it. I may be better with just paying the 500 for the european edition headlight housings that came with HIDs and has projectors for a huge increase.

 

[bshanahan14rulz]

Make your own wiring harness. Basically, all you are doing is instead of running your lights off the stock wires, you make the stock wires turn on a switch (called a relay) that lets electricity flow directly from the battery to the bulb with lower resistance wire.

 

[Buck91]

Haven't made the plunge into relay upgrades yet (Thanks to Schwein for linking my thread ), but I did speak with DS- very knowledgeble and straight forward. Definately drop him a line.

 

[Kuryakin]

One possibility is to see if there's an export headlight for these cars. Specifically European markets. A lot of cars built for here have FAR better headlights installed for these markets, as DOT headlights are actually illegal there, they are so bad. So, if your car is sold over there, there may be help!

I have a 1995 Mazda Mx6. The bulbs are little 9004 bulbs which have very poor output. I have cleaned the plastic housings for them so they are pretty clear now, but the light output is still very poor.

What are the best bulbs or solution I can do next to installing european HIDs that were available? (I won't be able to afford those for a year or so). I'm not looking for blue lights. I wanted a pair of Phillips extreme lights, but they do not make them in 9004s.

I am interested in a pair of HIR bulbs if anyone makes them to fit my vehicle. I can upgrade the wiring harness I'm sure if they require a bit more power, but they'd have to be able to fit into the headlight housings.

 

[-Virgil-]

One possibility is to see if there's an export headlight for these cars. Specifically European markets.

Careful, though, it has to be European headlamps for right-hand traffic. The left-hand traffic Euro-code lights they use in Great Britain, Australia, and New Zealand are just as dangerous and useless in right-hand traffic as are the left-traffic lights from the Japanese market.

(Also, since it's a Mazda we're talking about, the US-code lights are technically "export" lights, too!)

A lot of cars built for here have FAR better headlights installed for these markets, as DOT headlights are actually illegal there, they are so bad.

That's not quite accurate. It's a myth that the European headlight regulation is uniformly better than the US regulation. Both regulations allow a huge range of performance, and there are useless Euro-code lights just as there are useless US-code lights. The laws of physics apply to Euro-code lights just as they apply to US-code lights; if the vehicle stylists opted for a small reflector-type headlight, especially if it's not very tall, and the Euro-code light is a single-bulb high/low-beam type, it's not going to be very effective. The same goes for US-code lights.

US-code lights are illegal in many European countries because they don't meet the European ECE regulation, not because they're "bad". Same as European-code headlights are banned in the US because they don't meet the US regulation, not because they're "bad".

 

[Kuryakin]

Agreed, must be headlights for LHD nations.

As far as better, E codes are WAY better than 9004 headlights!!

And yes, DOT spec headlights don't pass muster because of poor cut off definition, not meeting minimum CP requirements, etc. Sure, there are SOME DOT headlights that meet E code standards, but they sure aren't 9004 headlights! Some DOT headlights use H4 bulbs (called 9003s which simply means they are more precisely built than H4s to eliminate the reaiming requirement when relamping), some have H7s with E code cut offs, etc. But E code requirements ARE more stringent than ours. It used to be E codes were illegal here because they pumped out too many CP (in some models), the E code cut off wasn't approved here (it is now), etc.

Remember, DOT stuck to sealed beams long after better stuff was available.

Careful, though, it has to be European headlamps for right-hand traffic. The left-hand traffic Euro-code lights they use in Great Britain, Australia, and New Zealand are just as dangerous and useless in right-hand traffic as are the left-traffic lights from the Japanese market.

(Also, since it's a Mazda we're talking about, the US-code lights are technically "export" lights, too!)



That's not quite accurate. It's a myth that the European headlight regulation is uniformly better than the US regulation. Both regulations allow a huge range of performance, and there are useless Euro-code lights just as there are useless US-code lights. The laws of physics apply to Euro-code lights just as they apply to US-code lights; if the vehicle stylists opted for a small reflector-type headlight, especially if it's not very tall, and the Euro-code light is a single-bulb high/low-beam type, it's not going to be very effective. The same goes for US-code lights.

US-code lights are illegal in many European countries because they don't meet the European ECE regulation, not because they're "bad". Same as European-code headlights are banned in the US because they don't meet the US regulation, not because they're "bad".

 

[-Virgil-]

Agreed, must be headlights for LHD nations.

That's right, but boy does the terminology trip people up! "LHD" left-hand drive refers to the position of the steering wheel in the car. "RHT" right-hand traffic refers to the position of the car on the road. While most cars in RHT countries have LHD, this is not necessarily the case. The steering wheel position within the car doesn't matter to the headlights...the vehicle position on the road does. So it's better to keep things simple and understandable by referring to headlights by the side of the road they're for, as the lamp makers do.

As far as better, E codes are WAY better than 9004 headlights!!

Sorry, no, it's just not possible to make a generalization like this that's accurate. The beam focus of many 9004 headlights is quite sloppy, with more glare than is allowed by ECE regulations, but the peak intensity and therefore the seeing distance is often quite a bit higher/longer with 9004 versus H4 of the same size and shape. This is what I mean by "not necessarily better/worse". Pick a size/shape/level of technology/level of build and material quality, and the 9004 headlight will let you see a longer distance on low beam (but with high glare, narrow beam width, and uneven road surface illumination); the H4 will do a better job of controlling glare and will have wider, more even road surface illumination, but shorter seeing distance.

E code requirements ARE more stringent than ours.

For low beam, ECE headlight requirements are more stringent in terms of glare control (ECE allows 350 candela at the glare test point at 13.2v, US allows 1000 candela at 12.8v), less stringent in terms of minimum allowable intensity for seeing distance (ECE requires at least 5100cd at 0.6D/1.2R at 13.2v, US requires at least 10,000 candela at 0.5D/1.5R or 0.6D/1.3R at 12.8v), and less stringent on control of backscatter (self glare in bad weather - ECE allows up to 438 cd at 12v, US allows up to 125 cd at 12.8v). For high beam, ECE minimum intensity requirements at the center of the beam are less stringent (ECE requires at least 27,000 cd at 13.2v, US requires at least 40,000 cd at 12.8v) but the ECE spec allows about double the light at the center of the beam. On the other hand, US specs call for a wider high beam distribution, while ECE spec permits a narrower distribution. The US regulation is much stricter on reflector durability (corrosion resistance, solvent resistance, etc.). The ECE aim procedure during headlight approval testing is more stringent. Neither spec is strict enough on plastic-lens durability, but the ECE test is faster (days instead of years) and more readily repeatable.

So no, it cannot be said that either spec is categorically more stringent than the other.

It used to be E codes were illegal here because they pumped out too many CP

They were (and are) illegal in the US because the US does not recognize ECE regulations. Back in the '70s, the US code allowed only 37,500 candela per side of the vehicle in the center of the high beam pattern. Now the US figure is 75,000. The ECE figure has ranged from 225,000 to 300,000, though it's only recently with Xenon light sources that any actual main high beam headlamps have begun to approach the lower end of the ECE limits.

the E code cut off wasn't approved here (it is now), etc.

There's never been a prohibition on a European-style low beam cutoff in the US. All that was necessary was to make sure the various US beam test points were complied with. Since 1998, there has been a technical definition for two kinds of low-beam cutoff in the US regulation.

Remember, DOT stuck to sealed beams long after better stuff was available.

Sealed-beam construction makes a lot of sense for car headlights. The photometric performance of the American sealed beams wasn't great, but that's more an implementation issue than a concept problem; various firms in Europe and one American outfit produced halogen sealed beams in the early 1970s producing much more precisely focused beams than the typical US unit. They saw fairly wide use in England, but the H4 sealed beams (w/well-focused DOT beam pattern) GE produced at that time didn't get carried on in production very long.

It has always been spurious to claim either headlight regulation is overall better than the other. Have you seen the beam performance of a Hella H4 7" round ECE headlight (on an objective goniometer-generated plot, not a subjective "wow, lookit that sharp cutoff on the garage door!")? It's pathetic! Maximum intensity of about 11,000 candela; just plain not enough light. A GE H6024NH sealed beam of the same size and shape has a max intensity of about 25,000 candela. Less glare from the Hella. About the same amount of backscatter from both. You're not going to sit there and keep insisting the Hella light is better because it's an ECE light.

Both regs have way too much room for poor headlamps. If the Europeans and the Americans were both to tighten up their regs (lower maximum glare in America, lower backscatter in Europe, higher minimum low beam seeing light in Europe, higher maximum high beam seeing light in America, stricter plastic-lens requirements in both regs) the situation would be a lot happier.

 

[Kuryakin]

This guy is what I would consider the ultimate authority.

http://www.danielsternlighting.com/nhtsa/NHTSA.html

Check ALL the annotated links, and become enlightened!

 

[-Virgil-]

Yep, I talk with Stern on a regular basis; he definitely knows his stuff and has worked in the car lighting industry as have I -- we go have coffee every now and then since he's local to me. We don't tend to disagree with each other on technical points, and my previous post in this discussion (the one that prompted you to post the link to Stern) is more or less word-for-word from a conversation he and I had the last time we chatted a couple weeks ago -- the candela values I quote are directly from the regulations themselves, so it's kind of silly to sit there saying "Is too! Is too! Is too!". You got a beef with the requirements of the regulations, go argue with the people who wrote 'em!

Back to the question of which reg is better: go ask Stern yourself; he does not espouse a "USA bad, ECE good" philosophy (though he used to, in the '80s and '90s when he had more enthusiasm than education or experience), he advocates a "Good lights good, bad lights bad" philosophy. In my library I have a paper copy of the report you link to. It's a sound analysis, but it doesn't demonstrate that the European headlight code is better than the US code; that's not what Stern set out to demonstrate. What it demonstrates, in accordance with Stern's purpose in writing it, is that there's probably no practical safety advantage to the US code versus the Euro code. That's not a new conclusion or a revelation; lots of research in lots of countries over lots of years has come to more or less the same conclusion: US-code headlights distribute light differently than European headlights, giving the driver a differently-shaped field of illuminated view, and each system has benefits and drawbacks, but overall in terms of crash rates (i.e., proof) there's no substantial safety benefit to either system. That being the case, there's no sound safety-based reason for the US to maintain its own unique headlight standards.

Stern suggests if there were one standard instead of two, car makers' headlamp program dollars would go farther toward high-performance headlights because they wouldn't have to spend money on engineering, tooling, and building separate headlight designs for the USA. That might be true in theory, but I wouldn't count on it; I think automakers would likely just pocket the "extra" cash and keep reserving high-tech/high-performance headlights for high-dollar models. Stern suggests a coordinated two-step solution: US adopts ECE lighting reg, and ECE lighting reg gets tightened up. That would relieve automakers of the multiple-lights cost, but would force them not to cheap out too badly. Good idea, but I don't see it happening anytime soon, unfortunately.

 

[Kuryakin]

I've spoken to Dan as well, many times.

So, back to the topic at hand, Mazda headlights with 9004 bulbs, they SUCK, big time. My old Subaru had 9004s and I've yet to meet a 9004 headlight that's good in any way, besides CHEAP.

US export headlights are nearly ALWAYS head and shoulders above what comes on them here. Caprices, C5 Corvettes, and many others have these available, and they are WAY better than what we get here.

E codes with VERY few exceptions, are superior in performance to DOT spec headlights. Over the years, I've equipped every car I've ever owned with E codes, where possible, and have enjoyed the improvement. Sure, there are probably exceptions, but not many. And E codes definitely allow me to see more of what I need to be seeing while being fine with opposing traffic.

DOT means fuzzy cut offs, meaning lower aiming points and less throw. High beams are usually the same pattern, higher and centered. No thanks.

Sure, a common standard would be great, however, our DOT STILL believes in NIH, which means no one else can POSSIBLY have a better system than us. Alas...

Yep, I talk with Stern on a regular basis; he definitely knows his stuff and has worked in the car lighting industry as have I -- we go have coffee every now and then since he's local to me. We don't tend to disagree with each other on technical points, and my previous post in this discussion (the one that prompted you to post the link to Stern) is more or less word-for-word from a conversation he and I had the last time we chatted a couple weeks ago -- the candela values I quote are directly from the regulations themselves, so it's kind of silly to sit there saying "Is too! Is too! Is too!". You got a beef with the requirements of the regulations, go argue with the people who wrote 'em!

Back to the question of which reg is better: go ask Stern yourself; he does not espouse a "USA bad, ECE good" philosophy (though he used to, in the '80s and '90s when he had more enthusiasm than education or experience), he advocates a "Good lights good, bad lights bad" philosophy. In my library I have a paper copy of the report you link to. It's a sound analysis, but it doesn't demonstrate that the European headlight code is better than the US code; that's not what Stern set out to demonstrate. What it demonstrates, in accordance with Stern's purpose in writing it, is that there's probably no practical safety advantage to the US code versus the Euro code. That's not a new conclusion or a revelation; lots of research in lots of countries over lots of years has come to more or less the same conclusion: US-code headlights distribute light differently than European headlights, giving the driver a differently-shaped field of illuminated view, and each system has benefits and drawbacks, but overall in terms of crash rates (i.e., proof) there's no substantial safety benefit to either system. That being the case, there's no sound safety-based reason for the US to maintain its own unique headlight standards.

Stern suggests if there were one standard instead of two, car makers' headlamp program dollars would go farther toward high-performance headlights because they wouldn't have to spend money on engineering, tooling, and building separate headlight designs for the USA. That might be true in theory, but I wouldn't count on it; I think automakers would likely just pocket the "extra" cash and keep reserving high-tech/high-performance headlights for high-dollar models. Stern suggests a coordinated two-step solution: US adopts ECE lighting reg, and ECE lighting reg gets tightened up. That would relieve automakers of the multiple-lights cost, but would force them not to cheap out too badly. Good idea, but I don't see it happening anytime soon, unfortunately.

 

[-Virgil-]

I've spoken to Dan as well, many times.

Oh, good. I'll mention your name next time he and I get together. UPDATE I just gave him a jingle. He says he doesn't remember conversing with an Ilya Kuryakin. Wonder what's up with that; usually his memory is pretty sharp and that seems like a memorable name.

So, back to the topic at hand, Mazda headlights with 9004 bulbs, they SUCK, big time.

True.

My old Subaru had 9004s and I've yet to meet a 9004 headlight that's good in any way, besides CHEAP.

True - it's difficult to think of many 9004 headlamps that are much of any good.

US export headlights are nearly ALWAYS head and shoulders above what comes on them here.

Sometimes. Not always. Not even nearly-always.

C5 Corvettes and many others have these available, and they are WAY better than what we get here.

The C5 Corvette domestic headlight was a 1992 Cadillac Seville headlight with a plastic rather than glass lens (to save money by specifying a weaker headlight lift motor). 9005 high beam, 9006 low beam. Cheap, crude, and nasty. The C5 Corvette export headlight was a Bosch 4" x 6.5" rectangular H4 reflector cemented to a plastic frame with a GM-Guide front lens. Cheap, crude, and nasty. They're differently nasty, but neither of them is much of any good.

E codes with VERY few exceptions, are superior in performance to DOT spec headlights.

Keep saying it as often as you want, and it still won't be true. Not objectively. Not scientifically. Please learn to distinguish between "I prefer this over that" and "that is WAY better than this". The former is an opinion. The latter is an assertion of fact, and in this case, it's easily disproven.

Over the years, I've equipped every car I've ever owned with E codes, where possible, and have enjoyed the improvement.

OK, so your subjective preference is for European headlamps. That's fine, but it doesn't mean they're objectively better. Headlight performance cannot accurately be judged by subjective preferences.

E codes definitely allow me to see more of what I need to be seeing while being fine with opposing traffic.

The science and data don't support this opinion.

DOT means fuzzy cut offs

Not so much any more -- remember, since 1999, most US headlights have been the visual/optical aim variety, rather than the old mechanical-aim type. V/O aim headlamps all have low-beam cutoffs. Some of them have a European-concept cutoff (low left, high right) and some have just a right-side cutoff at the horizon or a flat-across cutoff at the horizon, but they all have cutoffs. Your info appears to be out of date.

meaning lower aiming points and less throw.

Nope...US headlights are aimed higher than ECE headlights, as a matter of regulation and specification. And US low beams tend to give substantially greater seeing distance down the road. Not all of them, but most of them. That doesn't make them categorically better, but your claim that ECE headlights give longer beam reach is just plain incorrect. Sounds like you need to go update your knowledge; maybe it's time for another chat with Stern. Or maybe you or someone else will have some luck inviting him onto this discussion. Last time I raised the subject was quite awhile ago, but at the time he seemed pretty clear about his preference for staying off the forums.

High beams are usually the same pattern, higher and centered. No thanks.

High beams are supposed to be higher than low beams, and they're supposed to be centered. Don't see what your objection is here.

Sure, a common standard would be great, however, our DOT STILL believes in NIH, which means no one else can POSSIBLY have a better system than us. Alas...

Yeah, that's a real problem, whether we're talking about different (not better, not worse, just different) headlight standards, red rear turn indicators, no mandatory side-visibility of turn indicators, high-glare daytime running lights, no mandatory self-levelling with HID headlights, or any of the other big deficiencies in the US vehicle lighting code. But it is just simply not the case that the ECE lighting code is anywhere close to perfect. It's got big deficiencies, too, whether we're looking at the front of the car, the side, or the rear. They're just different from the big deficiencies in the US code, is all.

 

[Kuryakin]

Oh, good. I'll mention your name next time he and I get together. UPDATE I just gave him a jingle. He says he doesn't remember conversing with an Ilya Kuryakin. Wonder what's up with that; usually his memory is pretty sharp and that seems like a memorable name.
>>Not my real name, obviously! LOL


True.



True - it's difficult to think of many 9004 headlamps that are much of any good.




Sometimes. Not always. Not even nearly-always.
>>Yeah, nearly always. 9004s are simply terrible bulbs to start with, and go into lousy headlights.


The C5 Corvette domestic headlight was a 1992 Cadillac Seville headlight with a plastic rather than glass lens (to save money by specifying a weaker headlight lift motor). 9005 high beam, 9006 low beam. Cheap, crude, and nasty. The C5 Corvette export headlight was a Bosch 4" x 6.5" rectangular H4 reflector cemented to a plastic frame with a GM-Guide front lens. Cheap, crude, and nasty. They're differently nasty, but neither of them is much of any good.
>>Having seen a few C5s with the E code conversion, it's hands down a better lamp than the DOT unit. Not the best, to be sure, but definitely an improvement.


Keep saying it as often as you want, and it still won't be true. Not objectively. Not scientifically. Please learn to distinguish between "I prefer this over that" and "that is WAY better than this". The former is an opinion. The latter is an assertion of fact, and in this case, it's easily disproven.



OK, so your subjective preference is for European headlamps. That's fine, but it doesn't mean they're objectively better. Headlight performance cannot accurately be judged by subjective preferences.
>>You keep talking of this 'scientific evidence', but I've yet to see any of it. Why is that?


The science and data don't support this opinion.
>>> Bring it on!


Not so much any more -- remember, since 1999, most US headlights have been the visual/optical aim variety, rather than the old mechanical-aim type. V/O aim headlamps all have low-beam cutoffs. Some of them have a European-concept cutoff (low left, high right) and some have just a right-side cutoff at the horizon or a flat-across cutoff at the horizon, but they all have cutoffs. Your info appears to be out of date.




Nope...US headlights are aimed higher than ECE headlights, as a matter of regulation and specification. And US low beams tend to give substantially greater seeing distance down the road. Not all of them, but most of them. That doesn't make them categorically better, but your claim that ECE headlights give longer beam reach is just plain incorrect. Sounds like you need to go update your knowledge; maybe it's time for another chat with Stern. Or maybe you or someone else will have some luck inviting him onto this discussion. Last time I raised the subject was quite awhile ago, but at the time he seemed pretty clear about his preference for staying off the forums.



High beams are supposed to be higher than low beams, and they're supposed to be centered. Don't see what your objection is here.
>>using the same pattern for both, just offset, is just not good engineering.


Yeah, that's a real problem, whether we're talking about different (not better, not worse, just different) headlight standards, red rear turn indicators, no mandatory side-visibility of turn indicators, high-glare daytime running lights, no mandatory self-levelling with HID headlights, or any of the other big deficiencies in the US vehicle lighting code. But it is just simply not the case that the ECE lighting code is anywhere close to perfect. It's got big deficiencies, too, whether we're looking at the front of the car, the side, or the rear. They're just different from the big deficiencies in the US code, is all.