Interior LED conversion.

[Harley Dave]

Originally Posted by Laird Scooby

Define "ballast resistor" please?

My understanding is that it is used alongside a LED replacement lamp to "fool" the control circuits into seeing a similar resistance as an incandescent lamp - which is fairly low resistance, as against a high resistance LED lamp.

Without the ballast resistor, the control circuit would see the high resistance LED as open circuit which would trigger the "bulb blown" warning.

Ballast resistors are usually quite meaty as they have to dissipate as much energy (heat) as the original incandescent lamp would have

Cheers

Dave

[Laird Scooby]

Originally Posted by Harley Dave

My understanding is that it is used alongside a LED replacement lamp to "fool" the control circuits into seeing a similar resistance as an incandescent lamp - which is fairly low resistance, as against a high resistance LED lamp.

Without the ballast resistor, the control circuit would see the high resistance LED as open circuit which would trigger the "bulb blown" warning.

Ballast resistors are usually quite meaty as they have to dissipate as much energy (heat) as the original incandescent lamp would have

Cheers

Dave

Thanks Dave - that's exactly my understanding of them - i was actually asking "SwissXC90" as he said it wasn't a ballast resistor but a pull-down resistor.

I've always know a pull-down (or pull-up) resistor to be used to prevent the switched side of the load from floating too much and is in parallel with the switched end of the load and the switched rail - normally for status sensing. That's status as in on or off, not blown or open circuit/short circuit load although i suppose it's easy enough to modify to detect that as well.

The big thing nobody has questioned is how much current the new LED pulls.

If that was known then it would be possible to work out a definitive value for the ballast (or "shunt" resistor) so the CANbus system saw it as the original filament bulb.

Keeping the maths easy, replacing a 12V, 12W bulb with an LED bulb drawing 200mA.
Original current would be 1A and resistance of the filament bulb, 8.5 Ohms.

As the LED replacement takes 200mA, it makes the resistance of the LED 60 Ohms.
Using Kirchoffs Law, 12V divided by 800mA = 15 Ohms so a 15 Ohm resistor would be the ideal shunt resistor/ballast resistor in this example.

As the ballast resistor would draw 800mA and the LED replacement draws 200mA, that's a total of 1A, as per the original filament bulb in this.

Wouldn't get away with a 1/4 resistor for it though!

[GenericVolvoDriver]

Originally Posted by SwissXC90

Yes it will be fine. Anything from 4k to 10k will be fine.

Totally. For reference, I just used 4.75k because I had a bulk roll of 1% film resistors. First resistance I tried... enough of a pull down to turn off the LEDs and cancel bulb out messages in VIDA was good enough for me. I was going to try 10k next but only had 'em in 5% carbon resistors... only the best for the XC90 and the blue of the film resistors matched my exterior...

[SwissXC90]

Originally Posted by Laird Scooby

Using Kirchoffs Law, 12V divided by 800mA = 15 Ohms so a 15 Ohm resistor would be the ideal shunt resistor/ballast resistor in this example.

Except that it would dissipate so much heat that it would melt the surrounding plastic....

Weigh the heat issue against the (hidden) bulb failure error message

The decide what do you want most:
Visible, melted plastic?
Invisible, irrelevant bulb failure messages?

[Laird Scooby]

Originally Posted by SwissXC90

Except that it would dissipate so much heat that it would melt the surrounding plastic....

Weigh the heat issue against the (hidden) bulb failure error message

The decide what do you want most:
Visible, melted plastic?
Invisible, irrelevant bulb failure messages?

Of course, the original filament bulb also produces the same amount of heat as the resistor as my method is only using the same amount of total current.

This has all reinforced my decision to never have a CANBus equipped car so it is in fact irrelevant for me, i really can't see the point in over-complicating what is essentially the simplest circuit on the planet - a battery, a switch and a bulb - with a computer controlled version of the same.

[SwissXC90]

Originally Posted by Laird Scooby

Of course, the original filament bulb also produces the same amount of heat as the resistor as my method is only using the same amount of total current.

So the light comes by magic?

The energy consumed by a filament bulb is used up as 10% light, 90% heat
The heat is inside a vacuum chamber and the outside temperature of the glass is a lot less than the glowing white hot filament

The resistor on the other hand converts 100% of the energy to heat, and is designed to radiate that heat

If you could position the resistor in the central point of the bulb space, you may get away with it. Bur I challenge you to find space inside a plastic lamp fitting for a resistor that needs to dissipate 4W of heat....

[Laird Scooby]

Originally Posted by SwissXC90

So the light comes by magic?

The energy consumed by a filament bulb is used up as 10% light, 90% heat
The heat is inside a vacuum chamber and the outside temperature of the glass is a lot less than the glowing white hot filament

The resistor on the other hand converts 100% of the energy to heat, and is designed to radiate that heat

If you could position the resistor in the central point of the bulb space, you may get away with it. Bur I challenge you to find space inside a plastic lamp fitting for a resistor that needs to dissipate 4W of heat....

Yes, but the resistor would only be converting 3.6W out of that 4W, possibly less if we knew how much current the replacement LED consumed. For example, if the new LED consumed 0.8W of energy, the resistor would only have to dissipate 3.2W - goes back to Kirchoffs Law again and the need to know who much current the replacement LED actually draws.

Also let's not forget that to get to those heat levels the resistor would have to be on for quite some time and a filament bulb would also melt the housing if left on for that kind of time. The life of a courtesy light bulb tends to be only on for a few minutes at a time (if on a delay) at the very most.

What does surprise me though is that with all the technology available to produce the CANBus system, nobody has looked at it to see how to reset it so LED replacement bulbs won't trigger a bulb failure warning.

[SwissXC90]

Originally Posted by Laird Scooby

What does surprise me though is that with all the technology available to produce the CANBus system, nobody has looked at it to see how to reset it so LED replacement bulbs won't trigger a bulb failure warning.

Easy.
You incorporate it into the software and hardware design IF you want to
The car maker does not want to.
So he doesn't do it.

[Laird Scooby]

Originally Posted by SwissXC90

Easy.
You incorporate it into the software and hardware design IF you want to
The car maker does not want to.
So he doesn't do it.

With all the software hacks and hackers out there, someone must know how to do it then?

There will come a time when it won't be possible to buy filament bulbs for cars - it's already happened on the domestic market - so a lot of owners will come unstuck.
Maybe that's what the manufacturers want so people suddenly find their nice, shiny new box of tricks has got bulbs that can't be replaced without a £3k software mod to the CANBus system.

Worse still, the manufacturers might say the software mod can't be done and they need a new car................

[SwissXC90]

Originally Posted by Laird Scooby

With all the software hacks and hackers out there, someone must know how to do it then?

These are not hardware devices running software that is known and can be modified

They are hardware devices with specific chips, often programmable gatearrays, that have very special code in them, with no operating system as you would know it from a PC

The embedded controllers are programmed by the manufacturer to have a specific function. And that function is supported by the hardware around the controller.

You would need hardware and firmware development experience at the manufacturer and all the right tools to be able to change anything

Not easy at all. Not at all.

Have you every seen hardware level code for a micro controller?
it's a different world altogether.