LED taill light / brake light frustration

[PhotonWrangler]

I'm starting to wonder why it's so difficult to come up with an LED taillight bulb that's a drop-in equivalent to the incandescent versions.

I found a 3157 style LED that's actually a little brighter than the incandescent equivalent, and it even has full beam coverage like a regular bulb, fully illuminating the reflector housing like it should. It actually outperformed the incandescent version in terms of visibility and visual crispness. It was really impressive.

However I discovered that the second "filament" connection doesn't work right, and it messes up the logic in the flasher controller. This is not the load resistor issue - it doesn't even work properly when I connect it directly to a 12v battery.

I don't get this. We now have lighting-class LEDs for home use that are far more complex than tail light bulbs and they work well. Why can't we get a decent LED tail light bulb?

[Scheinwerfermann]

I'm starting to wonder why it's so difficult to come up with an LED taillight bulb that's a drop-in equivalent to the incandescent versions.

Because a lamp (taillamp, stop lamp, turn signal, etc.) designed for use with a filament bulb has optics engineered and focused to look at a filament of a particular size, shape, orientation, and position, collect light from it, and distribute that light at the required levels through the required range of angles. If the shape, size, orientation, position, or luminance distribution of the light source changes, all that optical engineering is right out the window. The analogy that comes up again and again (because it's a good one) is putting on somebody else's glasses: no matter how nice they look on your face, you won't see properly because the optics aren't for your eyes.

I found a 3157 style LED that's actually a little brighter than the incandescent equivalent

No, it isn't. Not unless it produces 402 lumens, which it does not -- not immediately, and certainly not with usage that extends beyond a couple of seconds, because none of the LED "bulb retrofits" have anything like adequate thermal management for the emitters.

and it even has full beam coverage like a regular bulb

That may be your subjective impression, but objectively no, it doesn't.

It actually outperformed the incandescent version

That may be your subjective impression, but objectively no, it didn't.

in terms of visibility

Measured how? I'm guessing the answer is "subjectively", which is not a measurement.

and visual crispness.

This does not mean anything; "crispness" is not a quality of light.

I don't get this.

It's fairly easy to understand. The difficulty is that there are a lot of marketers working very hard (for very good pay) to make it seem like an easy, simple, carefree swap. It is not. The problem is with the concept, not with any particular implementation.

[PhotonWrangler]

While I understand and appreciate your well made points, a few thoughts -

1) When I said crispness, I was referring to the much shorter period for an LED to ramp up to full brightness and then turn off again. This makes for a more noticeable transition from off to on, reducing reaction time and improving braking distance.

2) No, the beam coverage isn't exactly the same as an incandescent, however if it appears to be objectively the same, it serves the same purpose. A driver following from behind doesn't care about the exact spacial distribution of the light - he cares about it being noticeable. There needs to be a slight relaxation of the beam distribution requirements to accommodate this.

Ultimately it is more important that a driver isn't traveling with a burned-out bulb than a bulb with slightly different optical characteristics.

[Alaric Darconville]

2) No, the beam coverage isn't exactly the same as an incandescent, however if it appears to be objectively the same, it serves the same purpose. A driver following from behind doesn't care about the exact spacial distribution of the light - he cares about it being noticeable. There needs to be a slight relaxation of the beam distribution requirements to accommodate this.

It may *subjectively* appear to be the same; to *objectively* appear the same it must pass the tests in, for example, SAE J585, J586, and J588 (if the vehicle is less than 2032mm in overall width).

The driver following from behind doesn't consciously care if it's not the exact distribution of light; the driver following from behind my misinterpret the signal or even fail to see the signal at all.

Ultimately it is more important that a driver isn't traveling with a burned-out bulb than a bulb with slightly different optical characteristics.

The traveler should be using bulbs that satisfy the requirements. Driving with a "permanent" bulb that is also permanently out of spec means the driver is permanently in danger.

[Scheinwerfermann]

When I said crispness, I was referring to the much shorter period for an LED to ramp up to full brightness and then turn off again.

That is called rise time.

This makes for a more noticeable transition from off to on, reducing reaction time and improving braking distance.

Yes, if (and only if) the photometric performance of the lamp is adequate, and with an LED "bulb retrofit" it is not. Faster rise time does not countervail improper photometric performance, and neither does lower current draw or long rated lifespan.

I can't claim to have tested each and every LED "bulb retrofit", but in the course of my work I have tested several hundred different ones, and have seen others' tests of a great many more, and in no case has an LED "bulb retrofit" performed adequately. So it is not unreasonable to state categorically that they simply do not work safely or effectively.

No, the beam coverage isn't exactly the same as an incandescent

And there you have it.

however if it appears to be objectively the same

When you say "appears" in this context, it looks as if you are referring to what you see (or, more accurately, what you think you see) when you look at the lamp. That is not objective, it is subjective. There is no such thing as a lamp that "appears to be objectively the same". If you had photometric test data for the lamp with the filament bulb and with the LED "bulb retrofit" and the two data sets were to show effectively identical (or at least equally compliant) performance, then the statement "it appears to be objectively the same" would mean something.

A driver following from behind doesn't care about the exact spacial distribution of the light

It's not a question of what a driver might "care" about. The performance of vehicle lights is specified in terms of minimum and maximum intensity, and intensity ratio, throughout a range of horizontal and vertical angles so as to convey the intended message quickly, accurately, and without ambiguity to other road users whose eyes are located in a wide variety of locations relative to the lamps, and in a wide variety of ambient lighting conditions. This is a great deal more complicated than "Is it bright?".

There needs to be a slight relaxation of the beam distribution requirements to accommodate this.

Not actually, no. The standards are already quite lax; there is room in them for vehicle lamps that are only marginally effective. Further relaxation of the standards would endanger roadway safety. Moreover, the (objective, real) safety performance degradation of vehicle lamps that have been equipped with LED "bulb retrofits" is a great deal larger than "slight".

Opinions are fine, but they must yield to facts.

[PhotonWrangler]

If you had photometric test data for the lamp with the filament bulb and with the LED "bulb retrofit" and the two data sets were to show effectively identical (or at least equally compliant) performance, then the statement "it appears to be objectively the same" would mean something.

When you consider some of the wacky shapes that OEM tail lights are taking on, we're already talking about photometric performance that varies quite a bit from a regular incandescent tail light. This is a plea for common sense - an acceptable LED taillight bulb is achievable. I can't understand why it hasn't been done yet. It can improve driver safety if implemented properly, and I can't see a logical argument against that.

[Scheinwerfermann]

When you consider some of the wacky shapes that OEM tail lights are taking on, we're already talking about photometric performance that varies quite a bit from a regular incandescent tail light.

No, we're not. The shape of a vehicle lamp is not regulated -- the safety performance is. All vehicle lights, no matter what their physical shape might be, comply with the extensive, detailed regulations for photometric performance (and durability thereof with protracted operation, in the case of LED lamps), colorimetry, effective projected luminous lens area, geometric visibility, and other criteria. They are not just thrown together at the whim of the vehicle stylist. There is a wide range of allowable performance for each lighting function, but bulb-type lamps equipped with LED "bulb retrofits" invariably fail to meet even the minimum requirements. That is why they are unsafe, and in turn that is why they are illegal.

This is a plea for common sense

No, actually, this gives every appearance of a plea for reality to bend so as to accommodate what you fervently wish to believe.

an acceptable LED taillight bulb is achievable

No, it is not. Your assertion that it is achievable is not based on any facts or science; it appears to be based on what you want to be true. Unfortunately for all of us, that is not how this universe works.

I can't understand why it hasn't been done yet.

Yes, I think you can if you choose to. Clear, detailed explanations have been offered. You appear to reject them simply because they make you unhappy.

It can improve driver safety if implemented properly

The LED "bulb retrofit" you have in mind cannot be implemented properly. Even the LED system specifically designed as a lambertian light source for use in reflector optics -- the Osram Joule unit used in some Ford and GM tail/stop lamps, recently discussed here -- is only conceptually similar to a bulb. The optics are completely different.

[Lynx_Arc]

I can't figure out why people would want to (for the most part) bother replacing brake and marker lights on vehicles with LEDs unless perhaps they have an electric car as the savings in fuel is probably minimal with the exception of headlights, marker lights don't take a lot of power to run and brake lights are not typically on continually when driving.

[PhotonWrangler]

Because they don't fail catastrophically like incandescents, and the quicker rise time results in quicker reaction time on the part of the driver behind you. In my mind it's more about safety than power savings.

[Scheinwerfermann]

Incandescent bulbs generally do not fail catastrophically. It's not entirely clear to me what PhotonWrangler has in mind when he uses the term. Here is a high-speed movie of a poorly-made incandescent bulb failing catastrophically. We don't see many failures of this nature in automotive bulbs.

[Lynx_Arc]

Because they don't fail catastrophically like incandescents, and the quicker rise time results in quicker reaction time on the part of the driver behind you. In my mind it's more about safety than power savings.

Most cars have 3 brake light bulbs (or sets of them) and the chances of even 2 of them both failing at the same time is rather small and all three failing would be attributed to not the bulbs but electrical problems that may have LEDs fail also. As far as the quicker rise time I don't really think it matters that much unless you are allowing someone to tailgate you at 80 mph.

[PhotonWrangler]

Incandescent bulbs generally do not fail catastrophically. It's not entirely clear to me what PhotonWrangler has in mind when he uses the term. Here is a high-speed movie of a poorly-made incandescent bulb failing catastrophically. We don't see many failures of this nature in automotive bulbs.

I'm referring to individual bulbs. When the filament goes, that's a catastrophic failure. When an LED chip fails, others in the assembly are still working. This is also why LED traffic lights are such a good idea - when one chip blows out, the lamp is still producing lots of light.

[Scheinwerfermann]

When the filament goes, that's a catastrophic failure.

No, it's just a failure, and that makes several times in this thread you have misused terms that have specific meanings you don't seem to be aware of. It shows a lack of basic, fundamental knowledge about the topic under discussion -- automotive lighting -- that probably explains why you believe effective, safe LED "bulb retrofits" for car lights are possible.

Fortunately, ignorance (i.e., lack of knowledge) is quite curable with education. Pay attention to those who know more about the topic than you do, seek out and absorb solid and real literature on the subject from reliable sources, and soon you'll be up to speed on the topic and able to participate productively in a discussion of this nature.

This thread is somewhere between a lopsided quarrel and a nascent flame war and for that reason is being closed.