Computer logic for bulb out error...

[clemsongt]

Hey fly...or anyone for that matter, what is the computer logic for checking to see if a bulb is out or not?

I assume, since resisters are needed, that it checks for a certain range of resistance, and if the resistance in the circuit is more or less it throws the code. Since a blown bulb introduces and infinite resistance, why didn't they just program it to be when the resistance is greater than a certain amount. If the resistance was less than that amount (which it would be for OEM bulbs and LED's) then the bulb out warning would not appear.

[ECS Tuning-Audi]

You've got it understood correctly. But why would you want to not know if your bulb was out or not?

[clemsongt]

You've got it understood correctly. But why would you want to not know if your bulb was out or not?

My main thought is that it doesn't make sense for the computer to check if the resistance is below a certain amount. All that means is that the lights are more efficient (ala LED's). By changing the computer logic to "if resistance > XX, then throw error", then it would only throw the error when the bulb was out (OEM or LED).

[walky_talky20]

I think I can answer your question:

The bulb out warning does indeed look only for resistance greater the X. I believe any circuit resistance at or below factory spec would not trigger the warning, given that the fuse did not blow. If the resistance was lowered too much, circuit current (amps) would increase enough to blow the fuse. The blown fuse would then create infinite resistance (circuit open) and the bulb out warning would subsequently trigger.

I believe where you are getting confused is the process of "adding resistors" to fake the bulb-out warning when using LED's, etc. In fact, when you are adding these dummy resistors you are actually *lowering* the circuit resistance, not raising it. This is because you are adding the resistors in parallel. The equivalent circuit resistance (ie: total circuit resistance, aka the value that the computer is checking) is figured not by adding the resistor values, but by adding the *reciprocal* of the individual resistor values. Hence, the more parallel resistance you add, the *lower* the total circuit resistance - keeping the bulb-out warning off. This is especially useful when using very low-current (high resistance) lighting such as LED type lamps.

Another area of confusion is the resistance of the LED lamps. As stated above, they are very low current and so are quite efficient. This equates to *high* resistance, not low. The relationship between circuit current and circuit resistance is *inverse*.

I hope that helps answer your question.

[trevor51590]

I think I can answer your question:

The bulb out warning does indeed look only for resistance greater the X. I believe any circuit resistance at or below factory spec would not trigger the warning, given that the fuse did not blow. If the resistance was lowered too much, circuit current (amps) would increase enough to blow the fuse. The blown fuse would then create infinite resistance (circuit open) and the bulb out warning would subsequently trigger.

I believe where you are getting confused is the process of "adding resistors" to fake the bulb-out warning when using LED's, etc. In fact, when you are adding these dummy resistors you are actually *lowering* the circuit resistance, not raising it. This is because you are adding the resistors in parallel. The equivalent circuit resistance (ie: total circuit resistance, aka the value that the computer is checking) is figured not by adding the resistor values, but by adding the *reciprocal* of the individual resistor values. Hence, the more parallel resistance you add, the *lower* the total circuit resistance - keeping the bulb-out warning off. This is especially useful when using very low-current (high resistance) lighting such as LED type lamps.

Another area of confusion is the resistance of the LED lamps. As stated above, they are very low current and so are quite efficient. This equates to *high* resistance, not low. The relationship between circuit current and circuit resistance is *inverse*.

I hope that helps answer your question.

Thumbs up!

I should study for my Physics 2 Final... :(

[clemsongt]

I think I can answer your question:

The bulb out warning does indeed look only for resistance greater the X. I believe any circuit resistance at or below factory spec would not trigger the warning, given that the fuse did not blow. If the resistance was lowered too much, circuit current (amps) would increase enough to blow the fuse. The blown fuse would then create infinite resistance (circuit open) and the bulb out warning would subsequently trigger.

I believe where you are getting confused is the process of "adding resistors" to fake the bulb-out warning when using LED's, etc. In fact, when you are adding these dummy resistors you are actually *lowering* the circuit resistance, not raising it. This is because you are adding the resistors in parallel. The equivalent circuit resistance (ie: total circuit resistance, aka the value that the computer is checking) is figured not by adding the resistor values, but by adding the *reciprocal* of the individual resistor values. Hence, the more parallel resistance you add, the *lower* the total circuit resistance - keeping the bulb-out warning off. This is especially useful when using very low-current (high resistance) lighting such as LED type lamps.

Another area of confusion is the resistance of the LED lamps. As stated above, they are very low current and so are quite efficient. This equates to *high* resistance, not low. The relationship between circuit current and circuit resistance is *inverse*.

I hope that helps answer your question.

Interesting...I just assumed that the fix was to put the resistor in series (as I've never actually done it)...

Now that I've thought about it...I guess you're right...since V=IR if current drops and the voltage is the same 12V, then the resistance must increase.

[RVT]

I'll bow to someone who says "It does indeed" assuming you have studied the circuit in detail where I have never looked at it, but I submit that the alarm trigger may instead be "current sensing" has that has been around the automotive and motorcycle industry since the early eighties. Perhaps wrongly, I had assumed this to be the case with my Audi[confused].
Also a current sensing circuit is relatively easy to design (voltage drop simply holding the latch on an IC) and incorporate into a alarm circuit.
Your reference to a parallel resistance would still hold true as it is a current divider, allowing higher current to flow in the circuit and maintain the latch back at the control module.
Again, I'm old school, and have not studied the Audi electronics at all, so it was just my assumption to how it was done.

2 cents
Vince

[clemsongt]

I'll bow to someone who says "It does indeed" assuming you have studied the circuit in detail where I have never looked at it, but I submit that the alarm trigger may instead be "current sensing" has that has been around the automotive and motorcycle industry since the early eighties. Perhaps wrongly, I had assumed this to be the case with my Audi[confused].
Also a current sensing circuit is relatively easy to design (voltage drop simply holding the latch on an IC) and incorporate into a alarm circuit.
Your reference to a parallel resistance would still hold true as it is a current divider, allowing higher current to flow in the circuit and maintain the latch back at the control module.
Again, I'm old school, and have not studied the Audi electronics at all, so it was just my assumption to how it was done.

2 cents
Vince

Well current in a circuit is 'conserved'...yes, when there is a parallel circuit the current is divided up, but when it comes back together it has the same total current as before the split.

[djpaw]

V = IR

I/R = 1/R1 + 1/R2

aaah......1995

[fly300kts]

Interesting...I just assumed that the fix was to put the resistor in series

This will NOT work either.
Believe me, I spent so many hours trying to figure out the ECU logic with the LED rear Lights for example, and I could not find any, that I came up with an adjustable bench:

1/ Connect the bench to the circuit
2/ Start the car
3/ Wait for the warning code to appear on dash board
4/ Slowly increase the resistance value on the bench until the fault code disappear
5/ With fluke, check the resistance on bench
6/ Order resistor on line in accordance with value found
7/ Done! and fault code free

Phil

[clemsongt]

This will NOT work either.
Yeah...I get that now. So that means that the regulators that people buy are just putting a resistor in parallel with the LED.



Believe me, I spent so many hours trying to figure out the ECU logic with the LED rear Lights for example, and I could not find any, that I came up with an adjustable bench:

1/ Connect the bench to the circuit
2/ Start the car
3/ Wait for the warning code to appear on dash board
4/ Slowly increase the resistance value on the bench until the fault code disappear
5/ With fluke, check the resistance on bench
6/ Order resistor on line in accordance with value found
7/ Done! and fault code free

Phil

The LED's for your tail light come on with the head lights, right? Do they replace the existing tail lights? And then your brake lights are the same bulbs as stock, right?...Was the original setup not just a two stage bulb with one filament for tail light and another for brake lights?

[fly300kts]

The LED's for your tail light come on with the head lights, right? Do they replace the existing tail lights? And then your brake lights are the same bulbs as stock, right?...Was the original setup not just a two stage bulb with one filament for tail light and another for brake lights?

No,

LED's comes ON when:
1/ Brakes are pressed (bright)
2/ City/headlight selected (but dimmed)

Turn signals and Reverse bulbs remain the same

Phil

[walky_talky20]

I'll bow to someone who says "It does indeed" assuming you have studied the circuit in detail where I have never looked at it, but I submit that the alarm trigger may instead be "current sensing" has that has been around the automotive and motorcycle industry since the early eighties. Perhaps wrongly, I had assumed this to be the case with my Audi[confused].
Also a current sensing circuit is relatively easy to design (voltage drop simply holding the latch on an IC) and incorporate into a alarm circuit.
Your reference to a parallel resistance would still hold true as it is a current divider, allowing higher current to flow in the circuit and maintain the latch back at the control module.
Again, I'm old school, and have not studied the Audi electronics at all, so it was just my assumption to how it was done.

2 cents
Vince

I have to say, that I have not studied the circuit in detail, as there really is no circuit to study. It is all just a black box on the diagram. It could be looking for resistance or current, but it is all the same in the end anyway. I do believe that (at least on the B6), the bulb-out warning for the tail lights will show before you even turn on the tail lights. This would mean it is testing resistance.

When I said "it does indeed look for resistance greater than X" I was more so referring to the fact that adding higher load to the circuit (lowering the total circuit resistance) would not trigger the warning. Only raising the resistance/lower the circuit load would trigger warning. This is the case regardless of how the computer measures the circuit.

[BoSNiaN]

Interesting thread that deserved a bump [:D] So is there any way to disable this function using Vag-Com? Worst case scenario and its not possible, where is the speaker located that beeps when driving, in the cluster? Anyway to disable it with Vag-Com?



No,

LED's comes ON when:
1/ Brakes are pressed (bright)
2/ City/headlight selected (but dimmed)

Turn signals and Reverse bulbs remain the same

Phil

Are you talking about the LED taillights you were testing Phil? Did you end up getting them to run without throwing an error? I don't see how you could throw in a resister in that tight space between the taillight OEM harness and your lights[:/]

[fly300kts]

I run my tails LED for more than 2 yrs now and no fault codes

Phil

[BoSNiaN]

I run my tails LED for more than 2 yrs now and no fault codes

Phil

So how did you work out the errors? Did you use special LEDs or did you have to throw in resistors somewhere in the circuit?

[fly300kts]

So how did you work out the errors? Did you use special LEDs or did you have to throw in resistors somewhere in the circuit?

Both

phil

[BoSNiaN]

Both

phil

Mind explain the resistor part Phil? [:)] Are they located in the trunk or did you squeeze them between your wiring harness and the OEM taillight harness?

@disabling the module completely (that checks for bulbs out), any word on that? I found this video (http://www.youtube.com/watch?v=-J91O3ClPCY) on how to do it on a Golf using Vag-Com, think it would work for our cars?

[fly300kts]

they get hot so not in contact with any plastic or wire
Our car coding is different and VAGCOM will not help

phil

[schnell]

I have resistor bulbs for my license plate. When the lights are off, there is no warning. The moment I turn them on, boom! There goes the bulb out warning. Is there a way I can vag it, where the light will never come on?

[fly300kts]

No
If you get a code, it means that your resistor value is wrong, either to high or too low

phil

[BoSNiaN]

they get hot so not in contact with any plastic or wire
Our car coding is different and VAGCOM will not help

phil

Is it a trade secret or something? [>_<] My aftermarket LED tails are throwing errors so I am trying to figure out where I could put in some resistors...I am also in contact with Ross-Tech to see if there is a way to disable the bulb out function completely as I think it causes more problems than it solves,,

[jmarch1122]

My LED plate lights from Emil were throwing codes, so I sent them to phil for analysis. Once he and the CSI team are done, I should have a couple resistors I can wire in-line. Ill be so happy to be done with this.

[BoSNiaN]

My LED plate lights from Emil were throwing codes, so I sent them to phil for analysis. Once he and the CSI team are done, I should have a couple resistors I can wire in-line. Ill be so happy to be done with this.

Was running those same LEDs! Please let me know what happens [:D]

[rickescazu]

based on the above info, it should be just a wire wound resistor that is capable of restoring the total resistance show by a factory taillight bulb, why not just measure your stock incandescent bulb for resistance, then the LED assemble then calculate by the formula above 1/R total= 1/R1 +1/R2 where R1 is what you need in a wire wound to achieve stock resistance R total and R2 is the new LED. unless they have to be in a live circuit as Phil suggests in the Bench test?

updated 1 hr later

i also tested a license plate bulb and it was a .35 amp draw at 13V so that indicates a resistance of 37ohms. since most LED without resistors are very high resistance you can ignore them (almost no current flow at this size LED) and just go for the wire wound size directly to match the bulb resistance in the circuit. i don't think you an find a 37-40 so a 50 is close enough. that would indicate a current flow of about .26 amps which according to phil in another post must be close enough to the factory ,35 amps i got. but if you are thinking like me there are 2 bulbs so the combined current flow should be about .7 amps, and if you replace one bulb with a camera and a cheap little LED like i have (read china back up camera) you need to figure this together. so one resister to cover both bulbs would be about 20-25 ohms.

Rick