Fix for infamous LBF problem

[MitchSF]

It seems to me that we should be able to change the value of the resistor in the CCM that's used to determine whether to generate an LBF. Since we can find the current resitor value, and we know the standard lights use 55w, and HID's use about 110w at startup (which I think is the problem), we can use a simple ratio to calculate a new resistor value. I guess it's also possible that the LBF generates an error because the running (not startup) current is lower for HID lighting than standard bulbs. In that case, we just go the other way with the resistor.

I'd rather do this than cut wires to disable the CCM for low beams. I'd like to give it a try. Clearly this is the best solution, and it seems very possible that it will work.

Mitch

[clumpymold]

It seems to me that we should be able to change the value of the resistor in the CCM that's used to determine whether to generate an LBF. Since we can find the current resitor value, and we know the standard lights use 55w, and HID's use about 110w at startup (which I think is the problem), we can use a simple ratio to calculate a new resistor value. I guess it's also possible that the LBF generates an error because the running (not startup) current is lower for HID lighting than standard bulbs. In that case, we just go the other way with the resistor.

I'd rather do this than cut wires to disable the CCM for low beams. I'd like to give it a try. Clearly this is the best solution, and it seems very possible that it will work.

Mitch

I thought the LBF error message only goes on if the value is LOWER than 80 total or something. I didn't have the error message when I only had one HID with one halogen. But the moment I plugged in both HID bulbs, I had the message. Not sure if it matters what the wattage is on startup.

[00poop6x]

It seems to me that we should be able to change the value of the resistor in the CCM that's used to determine whether to generate an LBF. Since we can find the current resitor value, and we know the standard lights use 55w, and HID's use about 110w at startup (which I think is the problem), we can use a simple ratio to calculate a new resistor value. I guess it's also possible that the LBF generates an error because the running (not startup) current is lower for HID lighting than standard bulbs. In that case, we just go the other way with the resistor.

I'd rather do this than cut wires to disable the CCM for low beams. I'd like to give it a try. Clearly this is the best solution, and it seems very possible that it will work.

Mitch

Definitely, because this "fix" everyone is talking about is very GHETTO and NOT a "fix" at all, instead it is a loop hole. Your method is more of updating your CCM to monitor HID lighting's current draw and wattage output. I will have to wait and see what this is all doing when I get to upgrading to HIDs.

[mmaigret]

Mitch,

I also looked into this, and went so far as to measure the resitance in the LBF circuit. Unfortunately, the resistors are very high precision thin metal "fins". I speculated that I would not be able to duplicate the precision of the resitance required with over-the-counter 2% resistors, so I didn't bother trying. However, it may be possible if you have a lot of patience and access to high-precision components.

Mike

[MitchSF]

Mike,
I think we would first have to establish why we get an LBF. Is it because the running power requirement is less than the 55w that standard bulbs use, or because the startup power of about 110w appears as a short to the CCM? In other words, do we only get errors if the circuit is open, or do shorts generate errors too? My guess is that the CCM only checks for low or no current draw.

If there was a short, the fuse would blow creating an open circuit, generating an LBF. Maybe if we document resistor values used for other loads such as brake lights and turn signals, we can get an idea of how the value changes based on power used by that circuit.

I don't think a precision resistor would be necessary because any value that doesn't generate an LBF for a drain of about 30w, would work. If the ballast blew out, an LBF would be generated, and if a bulb blew, I guess the current drain from that ballast would decrease and an LBF would also be generated.

I switched my fuses from 10 amps to 15 and it seems to have cured my LBF problem. Maybe at a current drain of close to 10 amps where the fuse is about to blow, it acts a resistance, limiting current flow, thereby generating an LBF. Or maybe I don't know what I'm talking about. But for now, the problem is gone. If it comes back, I'll start experimenting with this. I don't think it will take very long to figure it out, but for now, I have other pressing projects. Hopefully there's an electrical engineer around here who's also getting LBFs!

Mitch

[clumpymold]

Definitely, because this "fix" everyone is talking about is very GHETTO and NOT a "fix" at all, instead it is a loop hole. Your method is more of updating your CCM to monitor HID lighting's current draw and wattage output. I will have to wait and see what this is all doing when I get to upgrading to HIDs.

Well, it is somewhat of a fix in that it "fixes" your OBC from saying that LBF error all the time.

Besides, why does it matter so much? Most people who do this mod have HIDs. It's usually pretty obvious if one of them is out.

[MitchSF]

Good point, but I think that since the CCM functionality is there, I and others would like to make it work correctly with HIDs. We certainly don't need to do most of the stuff that we do to our cars. And once we figure it out, it's just a simple resistor change.

Mitch

[mmaigret]

Hopefully there's an electrical engineer around here who's also getting LBFs!

Heh, I am an electrical engineer :P Anyway, here's the deal with the CCM. Each low beam's power wiring is routed through two series resistors. The node between the two resistors is monitored by the CCM via a comparator which compares the node voltage to a fixed reference value (this is called a resistor divider). Known quantities are the battery voltage (~12.5V), the resistor values, the reference voltage for comparison, and the appropriate current draw of the _stock_ halogen bulbs.

Thus, with the known quantities and Ohm's law (V=IR), the comparator has to decide if the voltage at the resistor divider's node is correct. Now, the plot thickens a bit here. I believe that the comparator operates in a binary fashion, basically, it decides yes we have a LBF, or no, we don't. If the voltage at the resistor divider node is sufficient to trip the comparator, the appropriate current (or more than appropriate) is being drawn and the bulb is functioning properly, thus no LBF. However, if not enough current is being drawn (as would be the case with a bad bulb), the voltage is too low at the node, and the comparator says no dice, LBF for you.

Have you ever noticed that it usually takes a moment or two after turning on the HID's before the CCM indicates an LBF? The current draw of HID's at start-up is sufficient to satisfy the comparator's test; however, the steady-state current is not.

I think modifying the CCM is possible and probably the "cleanest" way to get rid of the LBF messages. However, it is not as easy as it sounds. First, you need to figure out the resistor values that will work. Then you need to source some fairly precise resistors that can handle ~35W of steady-state power or more that will not get so hot as to set your dashboard smoldering (hence the exotic plate-type resistors in the CCM). Finally, you will need to hack up the CCM and solder you resistors in and hope it still works.

Seems to me snipping a couple of wires that can easily be tied back together again is a lot less painless. Maybe I'm just lazy.

Mike

[clumpymold]

^^^ Ahh, yes, that makes sense. I had a feeling it was LESS than a certain amount and I'm estimating that amount to be around 90 watts. One halogen and one HID capsule will total 90 watts and I get no LBF. With TWO HID capsules on, I get the LBF.

I'd just rather "fix" it this way, then to go and hack up my CCM. WAY too complicated and not worth it.

[///AlpinePower]

Good point, but I think that since the CCM functionality is there, I and others would like to make it work correctly with HIDs. We certainly don't need to do most of the stuff that we do to our cars. And once we figure it out, it's just a simple resistor change.

Mitch

true. somebody on this forum has a Compaq DL380 which has all of its fans (power supplies x2, chassis fans, etc) tied into the "BIOS" (a piece of Compaq config software, not an actual BIOS). it was too loud, so he had to do the same thing with resistors in order to install quieter fans and have the machine boot without throwing a fan error.

this is an accurate way to accomplish our objective:
http://www.8886.co.uk/ref/resistance_formula.htm

[infamous]

YESSS!! i found my answer to my problem...THANK YOU!

[Eric Giles]

Just wanted to say thank you to the original starter of this thread. I have been living with the LBF failure for months since the install of my HID's. Tonight I had to pull down the driver's side underpanel to install a new center console piece, so while I had it out I did this fix. It works great, and if I want to use the CCM again I will just splice the wires back-not a big deal at all for me.

Thanks again!

[Mrmontecristo]

Mike,
All things being equal, your fix - whether ghetto, loophole, or any other descriptive adjectives that have been thrown out here - simply removes the nuisance LBF with HID installed. Until the eletrical engineer gurus come up with a better schematic, I'll sleep well knowing that I followed you directions to a tee and now I no long get the LBF upon lighting up!
I also have some seeeerious lingo to throw at my nine year old son who recently asked the burning question, " why does that light always come on when you turn your lights on?" I'll wow him with the technical reasons and then the real world blue collar solution.
THanks for taking the time to share this with the rest of the bimmer world.
MrMonte

[cossie]

Excellent Guide, I recently fitted some HID's to my 93 UK M3, and it was fine for the first night.

On the second night the LBF warning came up and has done ever since.

I have tried most things, including removing the HID's and refitting the original bulbs, but the fault was still showing up even tho the HID's had been removed.

I also tried using 1 HID and 1 H1 but the only time the fault wasn't showing was when there was no bulb in the O/S headlight.

Anyway I have been looking at the different guides about sorting this, and this 1 seemed the best.

So I followed it, and found the CM (mine wasn't bolted in, just held with thin clips so came out easy) cut the wires and soldered them together, and it's working a treat.

Great work, thanks very much

[e-famous]

Did this today on a 95 M3. Worked great!

[Bav///Man]

Yup, Gary and I attempted it on my 95 M3 today. I cringed when I cut those four wires. Instead of the connector, we used crimpers. It worked flawlessly, and I even got the thing back in place securely.

One question, what is the huge metal thing in front of all of those wires? I didn't see it pictured in the original writeup.

[Quake]

Help Guys. Maybe i'm doing it wrong but i can't seem to get my lbf warning off. I connected the yel/green and yel/ grey then yel/blue and yel/red on the wiring side. Now i have lbf even when i don't have the lights on. Did i do something wrong? I am running the HIDs with a relay and a fuse. I'm not sure if that causes the problem. Hope you could help. Thanks in advance.

[Bav///Man]

when you say on the wiring side, it's the side coming from the car, right?

you did it right. Maybe check the connectors? I couldn't get his connector to work so I just connected them w/crimp connectors.

I'm not running a relay, how do you have the relay hooked up? Anyways, that shouldn't matter because this cuts the problem off at the source - the OBC check module.

[Quake]

when you say on the wiring side, it's the side coming from the car, right?.

Yes, By saying Wiring side i meant when the connector was cut the wire on the car side was tapped together not the connector wires tapped. I followed the pictures from the first page so i'm pretty sure i did that right.

you did it right. Maybe check the connectors? I couldn't get his connector to work so I just connected them w/crimp connectors.

i just put the wires together first for a test. I'm sure it is connected properly.

I'm not running a relay, how do you have the relay hooked up? Anyways, that shouldn't matter because this cuts the problem off at the source - the OBC check module.

The relay hook up came with the Hids that i bought. Basically the 2 wires from the original low beam terminal goes to the relay. Another wire goes to the positive (fused) and another goes to the ballast.

I think this is how it is set up.

Relay setup.
85, 86 - orig low beam wires
87 pos wire
30 goes to the ballast

If it is not the relay then i'm really stumped.

Thanks.

[Quake]

Hi I looked at the bently manual schematics and it seems that for my car 93 325 it may be grey/ blue and red/green connection. Tried it but to no avail.

Please help. Thanks.

[donoman]

I installed my HIDs tonight and the LBF failure w/ Check control came up . I assume the Check Control is related? Hopefully?

Mike, what makes you think there is a voltage comparator in the OBC? I'm also an EE and maybe we can brainstorm and find a real fix.

[MitchSF]

donoman, I switched from 10 amp to 15 amp fuses, and my lbf's went away, so I didn't pursue this further. Hopefull it won't be too difficult to come up with a modification, rather than a disconnect. The circuit is in the check control module, not the OBC. Good luck.

Mitch

[umnitza]

First and always look for the fuse.

But the fuse isn't the reason it will go out. Fuses are "passive" not active. They can't possibly affect it unless they have an decrease of resistance across their terminals right?

[mmaigret]

If you're not using separate relays for the ballasts, a larger fuse will definitely help the LBF situation (to a point). You don't want to use a fuse so large that it will allow dangerous currents delivered to the ballasts. I think 15A is certainly safe enough, and that's what I use. If you're using relays the size of the fuse is unrelated to the LBF and a 5A fuse is sufficient.

The reason a larger rating fuse helps the LBF when not using relays is because it has less resistance than the smaller fuse (it's a thicker wire). Remember the CCM monitors current draw, and also remember that we get the LBF because the ballasts draw less steady-state current than the stock halogens. Applying V=IR, you'll see that decreasing R while keeping V the same means an increase in I. Thus, if you're on the edge as far as the CCM is concerned, the larger fuse could help you not get LBF messages.

Mike

[umnitza]

Mike,

Thanks for the info. It makes logical sense. Thanks for explaining it