CAR LED headlight - load resistor (aka "CANBUS")

Thread Starter

Dulus

Joined Jul 15, 2014
14
Hello,
short story:
I have a car light harness where i have added an 8Ohm wirewound 200W resistor parallel to the LEDs. When testing and i input 12V (from table power supply) on input of the wire harness i measure the voltage on the output of the harness (where the LEDs would be connected) is ~10V(voltage drop across the resistor ?, and as the LED is in parallel, it receives the same voltage, correct ?). When installed in the car, when i turn on the lights without starting the motor (volatage ~12V), the LEDs are ok, no problem, but when i start the car (~14V) the LEDs are flickering, sometimes only one, sometimes both , sometimes very short, sometimes for few minutes, but after a while they stop flickering. What is causing this and what should i change to end the flickering, i suspect the resistor values have to be changed, but i would like some opinions for what value.


Long story:
i am currently in the process of changing the lights in my car from halogen to LED, because the better halogens i was buying (Philips RacingVision 150% H7) were only lasting very short time. I now bought Philips X-treme Ultinon 200% H7. When i fitted them in my car they were performing very well, no flashes from oncoming traffic, good cutout etc, but were giving an error in the car, that the bulbs are burned down (obviously because the LEDs are drawing much less power than halogens). My car has sensitive "bulb failure detection".
So obviously next step is to add some resistors to "fool" the car the halogens are still in place.
I have seen some already made harnesses selling on some shops, but the cabling was thin, without any protection, and the resistors were very small, and surely would get very hot.
So i created a wire harness myself.
I bought H7 sockets, cables, cable sleeves etc., big 200W 8R wirewound resistors, glue heatshrinks etc.
I created a harness, where the resistor is parallel to the LED bulb. As showed on the diagram.

When testing and i input 12V (from table power supply) on input of the wire harness i measure the voltage on the output of the harness (where the LEDs would be connected) is ~10V (voltage drop across the resistor ?, and as the LED is in parallel, it receives the same voltage, correct ?). When installed in the car, when i turn on the lights without starting the motor (voltage ~12V), the LEDs are ok, no problem, shining bright, no flickering but when i start the car (~14V) the LEDs are flickering, sometimes only one, sometimes both , sometimes very short, sometimes for few minutes, but after a while they stop flickering.
I was thinking it might be a voltage drop on resistor, so the LEDs are receiving toow low voltage to function properly, but why they are not flashing when light are on without motor/alternator running, but when motor is running they flicker.
They were functioning ok when connected directly without the resistor harness.
What is causing this and what should i change to end the flickering, i suspect the resistor values have to be changed, but i would like some opinions for what value.
 

Attachments

Reloadron

Joined Jan 15, 2015
7,501
Your 8 ohm load at 12 volts will draw about 1.5 amps. Will your table top bench supply deliver the current? Typically when a load exceeds a power supply rated max current the voltage will start to sag and it sounds like that is what is going on using your bench supply.

Normal automotive system voltage is about 12.6 volts and when the engine is running about 13.6 volts or slightly higher. Automotive LED design headlights are designed around this. They should work fine anywhere between 12 volts, maybe a little less right up to 14 volts or slightly higher.

My truck uses the headlights also as DRL (Daytime Running Lights). The DRL mode is pulsed which with an incandescent lamp or halogen lamp is not noticed but might be visible to the human eye with a LED type lamp.

You may want to get with the LED Headlight manufacturer and ask them about what is happening. I also do not understand the need for the 8 Ohm additional loads? When I went from incandescent lamps on my motorcycle to LED lamps I only needed to use load resistors in my turn signal lamps to "trick" my turn signal module into believing that incandescent lamps were out there. Going from an incandescent lamp headlight and my spots to LED did not require any changes.

I would start with the guys who made the lamps.

Ron
 

Tonyr1084

Joined Sep 24, 2015
7,853
Not the expert here on this, but the flickering may be due to the CANBUS trying to control a load it is not set up to. As for adding a parallel resistor - not sure that's the right solution either. OR it could be the BCM (Body Control Module) causing the flickering. It's not seeing the expected load and should be throwing a code. You might look at the codes and see if there's anything there. Not sure OBDII (On Board Diagnostics 2) will recognize or capture a code for something coming from the BCM. But it might. I just don't know for sure. However, the solution to the problem might just be an add-on device that you can pick up probably fairly cheap and not have to waste all that electrical energy as heat.

8Ω @ 12V comes to 1.5 amps and 18 watts. Long story short - you don't need a 200 watt resistor there. Won't hurt having it, but it's just way way way over-kill. A 35 watt resistor would handle the wattage for what you're trying to do. And as @Reloadron said, automotive voltages when running can range from 13.6 to 14.4 volts. Even a little higher, maybe 14.8, but I've never personally seen anything over 14.4 volts. So assuming we're talking about 14.4 volts, 8Ω @ 14,4V is 1.8 amps and 25.9 watts (call it 26 watts). 35 watt resistor would continue to be a safe margin. Using standard commercial engineering tactics (1 1/3 times normal loading) at 14.4 volts, 25.9 watts times 1.333 comes to 34.5 watts. I'd say the 35 watt resistor would do just fine.
 

narkeleptk

Joined Mar 11, 2019
558
@Dulus Year and Model might help a little.
I'm not real familiar with automotive lighting and even less with LED's but I don't think its anything to do with the canbus other then the canbus sending data between the controlling module and your instrument clusters bulb out warning. More likely it has some type of current sense in the controlling module for the supply. Perhaps when the voltage gets higher its sensing the current is dropping or raising out of the threshold and then it is shutting the output down. Some cars like BMW & MB will disable the output permanently if the faults happen too much. You then need to have the modules reset by specialists or replaced according to dealers. I think you need a little higher resistor. Do you know what value the pre-made kits use?

Not sure OBDII (On Board Diagnostics 2) will recognize or capture a code for something coming from the BCM. But it might.
Yes, you can connect to the bcm (and pretty much every other control module) with a good tool and pull codes but checking live data in the module that controls the lighting may be more useful as it can show you real time information such as the voltage that is being detected at the bulbs. The only code it is storing is probably going to be the one for bulb out.
 
Last edited:

BobaMosfet

Joined Jul 1, 2009
2,110
Contrary to what you may think, your 12V is not smooth under the hood with the engine running- it's closer to 14V, that's your alternator providing power, not the battery. It's providing 14V in order to keep the battery charged, and it's going to be rough and spikey, because it's being used to fire your plugs/coil-packs.

You need to filter your input voltage using a filter, comprised of one or more resistors, capacitors, and/or inductors, to give your headlights smoother voltage.
 

narkeleptk

Joined Mar 11, 2019
558
Contrary to what you may think, your 12V is not smooth under the hood with the engine running- it's closer to 14V, that's your alternator providing power, not the battery. It's providing 14V in order to keep the battery charged, and it's going to be rough and spikey, because it's being used to fire your plugs/coil-packs.

You need to filter your input voltage using a filter, comprised of one or more resistors, capacitors, and/or inductors, to give your headlights smoother voltage.
I assume he is using the OEM supply for the lights.
 
Let's ask:
1) original part # of automotive lamp, power, voltage, current
2) new part number and the same numbers.

What I would do is get a resistor when added in parallel to the new lamp draws the same amount of current as the old.

That's what I did to get my interior/trunk lights to behave in 2000 Chevrolet. I had to parallel a resistor for the trunk lamp only otherwise very wierd things would happen:

When a door and trunk were open and the system timed out, leds went off. When a door was disturbed opened and closed again, the LED would come on at about 1/2 brightness. If all doors were closed and one re-opened trunk would come on as full bright.

Car had auto dimming of the interior lamps, both on and off, retentive power and it would turn off the interior lamps after a preset time to avoid battery drain.
 

Thread Starter

Dulus

Joined Jul 15, 2014
14
Some additional info for all questions asked.

The car i am talking about is Opel Corsa E, MY2018. I have another 2 Opel cars, and every light i wanted to replace with led bulb (every light that was checked by the onboard diagnostics[for eg high beam, cornering, interior are not checked]) i had to add additional resistors. Opel has some of the most sensitive bulb checking system, so no LEDs with that integrated resistors (marked "canbus", altough it does not really have any communication over canbus) does not help. It would still show up a "bulb failure" error.

The headlights have a separate factory LED strip which is used (and PWMed) which is used for DRL, so i assume the headlight (which are only turned off at night) should not be PWMed and have a direct DC input.

My bench power supply is rated at 10A so it should be fine providing ~1.5A for the resistor. I have connected a different led (very "stupid" one) onto the harness, and when i connected the leads to the resistor (connected parallel) the LED became noticeably dimmer. So there is some voltage drop across the resistor correct ?

The harness you can usually buy of the shelf include 8R 50W wirewound resistor.

I was going with the 200W because on my tests the 50W resistor was going very very hot.
Even now with the 200W resistors (mounted on heatpads connected to the metal frame of the car), after a longer drive the resistors are so hot you can barely touch them (estimate ~60C). Same was when i replaced my lightning on the other car for license plate. The oem bulb is only 5W but 5W resistors with smallest current which did not trip bulb fault code were burning hot. I had to replace them with 50W and those are about 40C steadily when working.

OEM bulb is: 55W H7 12V (OEM is philips) (https://www.philips.co.uk/c-p/12972PRB1/vision-car-headlight-bulb)
replacement LED : https://www.philips.co.uk/c-p/12985BWX2/x-tremeultinon-led-car-headlight-bulb

The fact about ripple would make some sense, When turned when car not running, they are supplied from battery so steady ~12.5V while when engine is started, they is probably some ripple as stated. My car starts at ~14V+ (at least that what my OBD HUD are showing) and slowly settles about ~13.5V. Halogens does not flicker as they "take more time to dim" so the flickering will not be visible to naked eye.
When they were connected directly (without resistors), they were not flickering, because they received the full voltage?
Now when connected via resistor harness the voltage is dropped and thus they receive lower voltage which might explain the filcker.
I think i firstly try to experiment with additional values for resistors (i have 6R ready and 10R and 15R will have soon).

So, when i will connect the resistor with the smaller R value (6R), the voltage drop should be smaller, correct ? and the current higher obviously.
 
Last edited:

narkeleptk

Joined Mar 11, 2019
558
Now when connected via resistor harness the voltage is dropped and thus they receive lower voltage which might explain the filcker.
I think i firstly try to experiment with additional values for resistors (i have 6R ready and 10R and 15R will have soon).

So, when i will connect the resistor with the smaller R value (6R), the voltage drop should be smaller, correct ? and the current higher obviously.
There is no voltage drop if they are hooked up like your original attachment and they should not be that picky about voltage anyway.
Have you tried hooking the resister in series? I forgot if I missed that in your first post.
 

Thread Starter

Dulus

Joined Jul 15, 2014
14
There is no voltage drop if they are hooked up like your original attachment and they should not be that picky about voltage anyway.
THere is not a voltage drop ?
I was in impression there is a voltage drop across the resistor, and because the LED is parallel, there is also a voltage drop. I remember all components parallel shares same voltage, while all in series share same current. But then again, i am a noobie in these things.
I have also one off the shelf harness, and it also shows voltage drop. I will have to check it again with other power supply just to be sure and also when in th car.
Have you tried hooking the resister in series? I forgot if I missed that in your first post.
No. Same as before, i was expecting the components in series shares same current. and i didnt want to limit/overdrive the current to the LEDs.
 

Reloadron

Joined Jan 15, 2015
7,501
So the idea behind the parallel load of resistors is to provide the on board computer with current to look at and not "think" a lamp is burned out. There is no voltage drop, none to speak of anyway as the voltage across the load resistor and parallel LED will be whatever the system voltage is. An 8 Ohm resistor across just 12 VDC will dissipate about 18 watts so I would expect even a 200 watt resistor to feel hot to the touch as it is dissipating 18 plus watts. Higher system voltage it will dissipate higher power. At 14 volts it will be dissipating close to 25 watts.

The setup you are using, the LED lamps are designed for automotive use so any noise on the system should not be causing a problem. I would like to see the powered LED, engine running, on a scope just to see what system voltage actually looks like. I remember reading that flickering LED headlights can be a result of communication issues between the car/truck computer system and the headlight control module. The ability to actually look at the power at the LED headlights would be a big plus.

Ron
 

Thread Starter

Dulus

Joined Jul 15, 2014
14
I would like to see the powered LED, engine running, on a scope just to see what system voltage actually looks like.
Ron
I do have a very basic portable oscilloscope (some chinese DSO150), never actually used it before, so let me study how to work with it and maybe then i can provide some outoput :) .
Hopefully i will have some time to explore this problem soon.
 

bwilliams60

Joined Nov 18, 2012
1,442
I skimmed through this quickly but I think @BobaMosfet and @narkeleptk ate on the right track. I think you are doing the right thing to fool the supply but the supply is varying and probably needs to be stabilized and filtered. You cite a range of 10v on your supply to 14.4 v from the car where it seems to start messing up. On the bench it seems to be working okay. Voltage regulation and filter may help?
 

Thread Starter

Dulus

Joined Jul 15, 2014
14
I couldnt find the scope, must have put it somewhere , where i cannot find it.
In the end i actually found out that it looks like the ballasts themselves were the problem. because i switched them with the ballasts which are on high beam (i use same led lights on high beam and low beam, but only low beam is checked for working bulbs, so only low beam requires additional resistors), and the flickering stopped. i contacted the manufacturer, and he told me he will send additional balasts, as some of them could be affected by some quality problems. Its just that i am so unfortunate that for me both of them started flickering during short time one after another (basically during one ride). So i was expecting some "systematic" problem rather a individual component problem.
I will see if the flickering will come back after installing new ballasts.
 

Tonyr1084

Joined Sep 24, 2015
7,853
i was expecting some "systematic" problem rather a individual component problem.
That makes sense. But don't discount a systematic issue because a voltage spike could have knocked them both out at the same time. I've seen weirder things happen. For instance (and unrelated) my son-in-law had a bad fuel pressure regulator on his car. So he replaced it. That replacement was bad from the factory. They replaced it again. The car ran fine for two days. Then the car would hardly stay running and would stink of excess gasoline in the exhaust. "Couldn't be another bad regulator" we both thought. But it turned out the newest one developed a leak. So with that replaced all seems to be going well enough for now. Until something goes wrong again. I'm suspicious of the fuel pump (again unrelated to this thread).

Just when you think you have it all sorted out you discover a gremlin causing problems you didn't anticipate. Hopefully you've solved the problem(s). It's not unheard of for two ballasts to be bad at the same time, but again, for them to be bad at the same time COULD indicate a different problem.

Good luck with your solution. It may just settle everything down. Like I mentioned regarding my SIL's car, factories sometimes produce bad product.
 
Top