DIY Electric Car Forums banner
1 - 1 of 1 Posts

70 Posts
Discussion Starter · #1 ·
Re: [EVDL] LED Taillights - a money saver in gas cars? -rangeextenderin EV's? - an an

Sounds fine. If there's no silvered reflector, half the generated
lumens only show up as some backscatter. So let's say 60%- 240 lumens-
make it forward.

That's 240 lumens of r, g, and b. I don't have any actual info on how
many lumens are preserved out of the white, but let's say 1/3rd because
we block all the g & b. So that would be only 80 lumens of red coming
out from the original 400 lumens.

I have used a 270 lumen red Lamina BL2000 and I can tell you it's MUCH
brighter than my old Ford taillights. I know, I went out 200 ft away
and took a side-by-side picture. At that distance the perception is not
biased so much by a large glowing panel of a taillight vs the blinding
point of a tiny LED emitter.

Now something to keep in mind. These high power LED emitters lose a
significant amount of efficiency and thus output power when the
emitter's junction heats up, and highly elevated temps will permanently
degrade the device's efficiency. Now 3W may not sound like a lot of
heat, we dissipate that in a big resistor all the time. But there's 2
critical differences. One, a resistor can run at a very high temp
without degrading. Two, the resistor has a pretty efficient thermal
path from the element to the case so the inside's not so much hotter
than the outside. The emitter can only take an absolute max of 135C at
the junction- and the case has to be much cooler than that at 3W to keep
it below that. But- look at the graph on the spec sheet. The emitter's
down to only 20% of its rated light output (25C) with the emitter at
120C! Heck, at even 60C- 140F- inside the LED, it's down to 70% of its
rated output.

It takes a thick slug of aluminum and a lot of cooling fins to run 3W
inside a housing with no airflow and still keep the emitter down. I
suspect that this works well only because the highest power is only seen
when the brakes are on which is pretty intermittent. Even at reduced
power, it's probably running a bit hot with efficiency issues.

That's what's so great about these 7.6mm 4-pin SuperFlux packages. Not
a lot of intense heat in one point so there's no real thermal problems.
"Sunset Orange"/"Hyper Orange" is the approved taillight color. 2800mcd
@ 70mA (a bit over 3 lumens given the 70 deg viewing angle).


Paul wrote:

>First off I should point out that I misquoted the specs <http://
>>. I was looking at the ma
>column instead of the lumens column. At full brake light power it
>draws 240ma (current regulated.) That would be 2.88 watts of power
>for the brake light function at 90 lumens. That is a lot less than
>the 400 lumens of a 1157 bulb with both filaments lit.
>I've tested a few cheap LED auto bulbs and found them lacking (not to
>bright unless you are right in the center.) I have not yet tested
>this bulb (I've been mostly occupied with the actual conversion of my
>'66 Datsun right now, plus its not a cheap LED bulb.)
Danny Miller wrote:
>>Well lemme point out something else.
>>Unless you have a buck converter, the red 3W LED is like 3V @ 1A. But
>>we start with a 13.8V system, so we have to burn off 10.8W in a huge
>>Maybe you do have buck converter in there and if so great.
>>With multiple LEDs, the situation is simpler. We can put 3x in series
>>for a 9V forward voltage and the ballast resistor only needs to absorb
>Look at the specs on the LED replacement bulb I was proposing. Try
>finding lumen specs for cheap LED auto bulbs (I didn't have any
>luck.) It is current regulated and mounted in a heat sink. It would
>be internally dissipating 2.88 watts in the brake light condition if
>the LED didn't convert any to light (note the .24 amps at 12 volts
>spec.) I know that an incandescent bulb converts about 90% of the
>applied power to heat (good ones are in the 80's, but this isn't a
>"good" example.) So I can expect a 1157 bulb to make around 18 watts
>of heat (and 2 watts of light - to be filtered and reflected.)
>Using resistors as your voltage drop is not so simple. How many LEDs
>do you want in series? 3 is typically less than 6 volts (for red
>LEDs.) You would still be dropping 1/2 the voltage across each
>resistor (around 1/8 watt per 3 LEDs, times HOW many?). If you put
>more in series the current drops pretty quickly as the voltage drops;
>the result is an LED lamp that dims about as fast as the incandescent
>one (the incandescent bulb shifts toward red as the voltage drops.)
>>The 7.6mm P4 pks are easy to use IF the red lens is screwed onto the
>>back half. If the whole thing's one glued together piece and the bulb
>>is socketed through the back, we may have a problem.
>That package may be easy but how easy is it to find replacement tail
>light housings for a 1966 Datsun 411? I'm *not* going to hack my tail
>lights; I would rather run 1157 bulbs (it has kept the car intact for
>over 40 years now.)
>Paul Gooch
>P.S. - there is no significant power savings in LED illumination for
>on-road EVs at this time. The only advantage is faster brake lights.
>That is nothing if they don't notice them!
>For subscription options, see

For subscription options, see
1 - 1 of 1 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.