Riddle me this: Where is the current coming from?
 

A while back I added LED upper brake lights to my car by tapping into the brake wire right where the harness leaves the engine compartment in the back driver's corner.
This has worked as intended for the last couple years.
http://forums.pelicanparts.com/uploa...1508367160.jpg

Recently one of the 35W halogen brake bulbs burned out so I decided to replace them with LEDs. (Running a high wattage bulb through the brake light pressure switches isn't the best thing.)
Since the Phillips LED 1157 replacement has been in place the upper brake lights are dimly on when the brake lights are off.
http://forums.pelicanparts.com/uploa...1508368165.jpg

So where is the current coming from to dimly light the upper brake lights? My guess is it is back feeding from somewhere but where?
I assume I could "turn out the light" by switching back to incandescent bulbs but I'd really like to keep the reduced current draw.

Lights are creating a ground from another source. Not from brake light switch. Sometimes happens when an incorrect bulb is jammed into a socket. Such as a running light bulb sending power through the ground circuit.

Quicksilver- I’m no help in your plight but I am a fan of your third brake light set up. Where did you source the lights and are the mounted on exterior or in cabin?

I'm not familiar with later cars so bear with me.

Are there any lights on when your car is running? Meaning, does your car have running lights or side marker lights that are on during normal driving, even during the day?

The side marker lights and running lights both share a common ground with the brake lights. Additionally, the power feed for the left brake light is split from the power feed from the right side brake light. If you have a poor ground connection for any bulb in the fixture and you have active lights during driving, you could be getting the feedback through the feed to the brake lights where the ground is made at your rear window light bar. Since the feed to the left brake light shares a common wire from the right brake light, the poor ground might be from the right side light fixture.

Replacing the incandescent bulbs in the brake lights with leds greatly reduced the resistance and may now allow the leaking current to back feed into the circuit to your light bar.

Just a thought that you might check out, but it will be discarded if you don't have any lights on in the rear fixtures during normal day time driving.

http://www.pelicanparts.com/911/911_...SC_Part3-1.jpg

I have an issue where my LED parking/turn signal lights are barely light when i have no lights on and the engine is running.

I have checked all the bulbs I can think off but haven't found my issue yet. So for what its worth, you're not alone in suffering with LEDs on these old cars. Looking forward to what you discover.

Quicksilver, isn't the brake light bulb 1156 not 1157?

I would trace V24 from the brake sw.
or, pull that wire off the brake sw and see if it goes out

Look on Spokes led lighting thread. I think there was a diode or resistor issue. They had to wire some stuff into the system to make sure the lights only came on when the switch was on. My LED tails do this now and I haven't corrected them because they get people's attention on the road.


-Steve

LEDs are polarity sensitive where incandescents are not, also you should not mix led and incands on same circuit, if you only replaced one bulb on the upper brake light you may need to replace them all (assuming it has more than one bulb) check that

I'm with T77911S on this.
You should have 2 switches wired in parallel - maybe one developed some carbon on the contacts from trying to drive 35W halogens.
Disconnect 1 and then the other.
You should be able to run with 1 due to the decreased I draw.
Two are used b/c of redundancy.

The lights should turn on at the same voltage level, unless there is an internal resistor in one of them.

Wait...why?

I experienced very (similar) strange behavior from the taillights when installing LEDs in place of the 1156/1157 there. Connections would switch from ground to 12v in a very confusing way. I never quite figured it out and went back to incandescents.

Perhaps I should have said in the same fixture which is more accurate. For some reason the LED (which is polarity sensitive to begin with) does what Tremelune (and many others) has experienced when placed in line with incandescent bulbs.

I modded a center reflector and placed leds in it, when car was running they had a slight glow as if they were being energized by static electricity.

I used cheap ones from fleabay/china and they were all but useless ..I believe its the way they are wired, I tried to put them on my front yard pole light and no matter what could never get both to light properly, whether in series or parallel, if both lit they were not bright enough .When one was at proper light out put the other was dead so I took it upon myself to make it just one bulb. Maybe someone more versed in electronics can explain it , I just am a hack and can only relate what I have experienced in using the things for a few years

I fitted my tail lights and euro center reflector with Sylvania Zevo LEDs and have not had any issues with that "phantom glow"

Sometimes in brake circuits there are other sensing circuits that detect when the brakes are active, cruise control, etc. When the brake circuit is not active, normally the incandescent bulb presents a low resistance path to ground for these other sensing circuits. However, when the incandescent bulb is replaced with a led, this low resistance path to ground is lost. These sensing circuits often apply a low current to the circuit that may be powering up your very nice center top brake leds. Sometimes a resistor is placed parallel to the led incandescent replacement bulb in order to shunt this sensing circuit to ground.

You can determine the shunt current required by removing the led bulb, and with the brake lights OFF, measure the current that would normally flow through the incandescent bulb. A resistor of about 1-2V divided by the current you measure should be sufficient to shunt the sensing current. The other little detail is the power rating of the resistor, this will have to be much greater than 144 divided by the resistance value you determined.

I’m not sure what year you have, but the cruise control completes a ground through the brake lights and disengages when it gets +12v from that wire. With all incandescent bulbs, the resistance to ground is 6 ohms. With led/smd bulbs, it's not. ;)

The brake light also is one of the running lights so it has 2 filaments. The "correct" bulb is an Osram 7528 which has oval contacts instead of round ones. But everyone uses 1157s because everything else is the same (plus you can find an 1157).

_________________________________________
I'm suspecting that the current is backfeeding through the LED 1157 because an LED has no resistance in the reverse direction. It would probably require that there be a weak ground somewhere.

I hadn't thought about a possible connection to the cruise control unit. I'll try disconnecting the cruise brain. If it is upstream the upper lights should stay on if I remove the LED 1157s.

I guess another thing to try is to run a direct ground wire to the light socket and see if that knocks it down. Obviously wiring a diode in would "fix" it but that would just put a bandaid on the real issue.
I'm not sure if I'll have time to chase it down before the car goes on the trailer tomorrow. (Targa Baja California here I come!!!)

If you have cruise control, then the question is answered ;)

I’d say it’s definately pushing electrons through the negative wire . ;)

But I know nothing about led lights .
other than they can cause “bulb out” issues on later cars ..
yeah there is a type that guarantees no problems but I can’t remember the details .
I think I went with a Phillips model to stop the issues

where does V24 on the drawing go???

on older drawings I have I see it goes to some electronic device but the print is not clear enough for me to read any writing.
most likely your are getting power feeding back from that device.

remove the blk wire from the brake sw and check it

I have the same problem, only different...

My third brake light glows when the car is off and glows brighter when the car is on (both without pushing the brake pedal). Does cruise require key-on to send current, or am I looking for something else? The third light is tapped into the left brake light, and I have Spoke LEDs for marker/brake.

This is fascinating. Why wouldn't Porsche use the brake light sensor/activator directly and a proper ground instead of this roundabout do-my-lights-work trigger?

I guess my followup questions are, can I re-wire it to do this or maybe just disconnect cruise control completely...?