LED brake lights

[Tervuren]

Originally Posted by wild man

UPDATE: That's looking like it was a bad call. From what I see (on velocity network), they are already available on VERY high-end cars right now, albeit with 3 per side for a total of 6 per vehicle!

I have already seen a 3.3 watt single LED flashlight that put out 130 lumens. I think that's about 1/3 of what is needed per headlight for a single LED. It comes down to more a matter of getting the reflector/lens setup right to form the emission pattern of the LED into to the desired beam pattern in an efficient manor.

We have 400 watt single LED's we've created at my job. Bright sun of a guns. I think the real problem with having them as headlights, is getting rid of the heat, LED's need a much lower operating temperature than regular bulbs. When you make one of large wattage, it can be difficult to drawl all that heat away, which means a very short life if its not done properly.

[wild man]

That sounds like a real eyeball burner if you look directly at it without wearing a #10 welding helmet!

So it all comes down to support issues, such as heat dissipation and collimating the photon energy into suitable headlight beams.

If the supercars (yea, I'm talking higher level than p-cars) are now sporting 6 LED headlights to light their way, I would guess that it'll be down to 2 by sometime next year.

But to have LED replacements for like H3s & H4s (9003, 9004, etc.), where the original reflector/lens needs to be utilized, it will definitely take longer, if it is even possible.

[wild man]

Hey Tevuren, since you obviously work at the forefront of this technology, do you believe that LED bulb replacements are possible for existing headlight reflector/lens assemblies, such as the H3?

If so, are you able to predict a timeline for when it will happen?

[egil]

Originally Posted by wild man

The (hot) resistance of a brake light bulb should be around 6-7 ohms (25W). But it is very possible that the circuit could be tripped by a (much?) higher resistance. The question still remains:

What is the highest value of resistance that will trip it every time?

Unless you ride the brakes, which is surely a bad practice, power consumption should be a non-issue in this case.

That said, you are surely right in your assumption that a fairly high resistance will work. No need to emulate the lamp resistance (and if it was, it is the cold resistance which is relevant here, which is much lower than what you specify). It is very easy to verify, just by measuring the voltage of the brake lamp circuit when the brakes are not engaged. I am not sure what the switch voltage is, but if you see a voltage below, say, 0.2V, I would be very comfortable it is good enough.

[wild man]

Cold resistance?

I always thought that it was the brake lights going hot (on) which cancelled the automatic speed control (CC).

Anyways, I'd now like to see a schematic diagram of the CC control module circuit.

Anybody have it?

[vdubr928]

Originally Posted by egil

If you care to read the thread, I have explained how you fix this by adding one single resistor from the brake light circuit to ground. This is all that is needed.

If you do not believe me, then by all means, you are of course free to do it differently. You will require extra wiring, and adding of an extra switch and fabricating a bracket, or finding a compatible dual pole switch. Plus you will need a resistor to ground. The choice is yours. (Of course, if you do not believe in my first solution, I am not sure why you should beileve in anything else I wrote)

I just didn't see it. Surely that would work as well as that's how we defeat the check control modules on bmws for leds. I was a little confused by your explination if the circuit itself so I sudgested whick could be done very easily with boneyard parts and about 2' of wire and appropriate resistance of two bulbs. It sounded like you were saying that it was the light off resistance but now that I think about it, a resistor would effect light off as well as long as its uphill of the diodes. Either way would work I suppose. If your way does then its easier. Wasn't a matter of not beleiving. You just touched on it so fast I didn't get the whole message. Since these guys are in this perdiciment why not one of you test this?

Point is. I did read the thread thank you. But it sounded like you guys were arguing that the resistor fix would not work. Just offering another possibility that would be a gaurenteed fix.

[vdubr928]

Why don't the led companies get wise and do a dual circiut where the led is fired by a transister acting as a relay but have a dummy circuit that shows the correct value fir on board computers. This I think would he very simple.

[wild man]

May seem simple on the surface, but it would not be cost-effective, if that type of setup is only required in a small percentage of cases.

By "uphill" side, do you mean the "hot" side? If so (or even if opposite), that implies placing the resistors in series, which would be incorrect. They need to be in parallel with the diodes (shunt them), in order to achieve the desired effect.

EDIT: (this is just an FYI) I have found that the HOT resistance of tungsten filament light bulbs is about 10X the COLD resistance.

[vdubr928]

Yes, now that I think about it it wouldn't matter. Not sure why I was thinking about current flow when you are actually just shorting.

I think alot of people would use them and it would be inexpensive to manufacture as you are already using surface mount for the leds. You could even just parrallel the leds with a surface mount resistor internally. Couldn't you? I would think you would have alot less returns and there is plenty of space both in the barrelled of the light and where the leds are mounted. Think about all the people that buy these lights and take them back or theater refuse to buy because of threads like this. Seems like an easy bird to kill and alot of potential gains for the manufacturer. The lights themselves are getting hard to find on shelves simply because there are lots if returns.

[egil]

It seems many of you are making this way more complicated than it is. The only thing that is needed is one resistor from the brake lamp circuit to ground. Period. No bad side effects. The only question is the optimum value of that resistor. I have had a full LED brake light conversion on my to-do list for a while, esp since the cabrio central brake light is a pita. I will try to prioritize this, and come back with a verified solution soon.

(Back to the nit picking dept: It is the cold resistance that applies. When the lamp is lit, the circuit will have 12V anyway since the brake pedal switch is closed. The issue is when the switch is open, and the lamp is no longer lit. The voltage sensed should be fairly close to zero for the cruise control to sense a "brake not engaged" situation. Within the cruise control module, there is a pull-up resistor to 12V. A regular filament will pull the circuit to ground, but the LEDs will not.)

[djnolan]

Guys I think you have the correct technical solution to keep the cruise control working with LED's. The resistor approach seems like sound logic.

Now what if you live in an area with major bad traffic all of the time and you never get to use the CC? Are the LED's any brighter than regular bulbs? If you go to the trouble of changing to LED's is it gonna keep someone texting in a mini van from plowing in to your 944?

Ha Ha!

[vdubr928]

Originally Posted by djnolan

Guys I think you have the correct technical solution to keep the cruise control working with LED's. The resistor approach seems like sound logic.

Now what if you live in an area with major bad traffic all of the time and you never get to use the CC? Are the LED's any brighter than regular bulbs? If you go to the trouble of changing to LED's is it gonna keep someone texting in a mini van from plowing in to your 944?

Ha Ha!

Possibly. Conditions would dictate the answer. There is at least one person on this thread that could probably explain some of the variables to the answer. But what you should get for sure is not having to replace the bulbs near as much since leds last way longer.

[wild man]

Originally Posted by vdubr928

leds last way longer.

Right on. The only other advantage to them is more efficient, and that is actually a hinderence in this particular case. Like said, it is the optimum resistor value that needs to be determined.

Tell you what, I just screwed in a 12 watt 93 LED flat array (about 2 1/2" diamater) that is supposed to replace a 150 watt, 120 volt incandescent floodlight. Tell you what, it takes getting used to, or something. That's the most "piercing" white light I think I have ever seen!

[egil]

Originally Posted by djnolan

Now what if you live in an area with major bad traffic all of the time and you never get to use the CC? Are the LED's any brighter than regular bulbs? If you go to the trouble of changing to LED's is it gonna keep someone texting in a mini van from plowing in to your 944?

I've tested the red "14 Ultra Power LEDs" that look like this:


They have a light pattern that is compatible with the standard item, and when installed, seem to emit approx as much light as the standard item (LED in the left lamp, bulb in the right - I can take some photos later if anyone are interested). In fact they are quite indistinguishable from the standard item when installed. Except that they light up instantly, giving the car behind you maybe 1/10 of a second earlier warning. So in theory at least, you should diminish the chances of being rear ended by that mini van.

These bulbs cost around $20 a piece. Although I haven't tried them, I would assume LEDs with less light output would be less suitable. It also seems logical that red LEDs would be more suited than white for the brake light.

[Cochezz]

Why would you want to use LED's?

What is the advantage to you or the car?

[egil]

Originally Posted by Cochezz

Why would you want to use LED's?

What is the advantage to you or the car?

Very good question. I have three answers for you, which you may want to sort in your own order of importance:

1. Give car geeks something to do, and keep their minds occupied.

2. Reduces the reaction time of the vehicle behind you by probably over 1/10 second. This translates to about 3 meters at a highway speed of 100 km/h (10 feet at 60 mph for the metrically challenged), which seems not very much, but that you will find to be a rather significant figure if your own car happens to be stopped anywhere within that range.
See http://deepblue.lib.umich.edu/bitstream/2027.42/64045/1/84696.pdf

3. Pretty much avoid the risk of driving with burned out lamps. Thus avoiding a safety risk, a slight maintenance chore, as well as, depending on your jurisdiction, the chance of a fine, and being pulled over by the the boys in blue.

[egil]

Comparing LED brakelights (left lamp) to regular:

-020 ms

+020 ms

+060 ms

+100 ms

+140 ms

+180 ms

+220 ms

It is obvious that the center, halogen, lamp reacts quicker than the traditional lamp on the right. The assumption that the regular lamp is approx 1/10 slower than the LED seems to be about right.

From the photos it does seem like the LED is slightly weaker than the regular lamp. For the eye the I judge the apparent brightness to be approx the same. Also, for the eye, both lamps appear to be red when lit, not yellowish as registered by the camera. The LED is the type with 14 lamps as seen earlier. The stripes are from falling tiny leaves -- it was somewhat windy.

The recording was made at a frame rate of 40 ms, and I have just assumed the switch was closed between the 1st and 2nd frame. A high frequency capture camera is required for more accurate readings.

[vdubr928]

That's really cool. Thanks for the play by play.

[vdubr928]

Originally Posted by wild man

Right on. The only other advantage to them is more efficient, and that is actually a hinderence in this particular case. Like said, it is the optimum resistor value that needs to be determined.

Tell you what, I just screwed in a 12 watt 93 LED flat array (about 2 1/2" diamater) that is supposed to replace a 150 watt, 120 volt incandescent floodlight. Tell you what, it takes getting used to, or something. That's the most "piercing" white light I think I have ever seen!

Lol. HD light

[egil]

Here is the parking light at night. The LED is to the extreme left. (The image is exposed such that the lamps are not overexposed, so that it is possible to judge their relative brightness.)



The next image is with the brakes applied as well. I would say the difference between the inner parking lamps and the outer braking lamps appear more pronounced to the eye than in the photo.

In addition to the standard central brake light (three bulbs currently working, one not), I have added a strip of LEDs for comparison. You can also see their reflection in the trunk lid.



The central LEDs appear significantly brighter than the halogen bulbs when seen straight on, but the LEDs are more directed.