Help with LED brake lights

[Dannyx]

Re: Help with LED brake lights

Didn't look too deep into it when I got them: I was just looking on an auction site where I usually buy stuff from and though ".....looks good enough. I'll buy a set" thinking it will just be plug and play, so imagine my dismay when I wasted practically an entire day trying to get them to light up but failed....For this reason, I don't suppose they're some super-high quality lamps, though they sure look well built and cost a considerable amount, so they're not bottom of the bin either...yeah I just realised to this day I probably haven't posted an actual pic of them....can't remember. Have I ? :|

[Longbow]

Re: Help with LED brake lights

Reading the schematic (with difficulty), refer to page 3-16. #160 appears to be the brake switch, which then connects to a computer unit, #118. The stop light is #639, appearing on page 3-45. If I'm reading things correctly, page 3-45 is a simplified drawing that does not show the computer that controls that stop light. This car has a pretty complex electrical system. It would be helpful to get the English schematic.

[Dannyx]

Re: Help with LED brake lights

If there were an english version, I would've posted it the first time I'm afraid. It's technically a Renault, so perhaps, just maybe there is an ENG schematic somewhere, but for now I haven't been able to find it and didn't dig too deep for it either since the one in my native language sufficed. It doesn't matter....

Yes you read it correctly, although I explained this myself in the post where I posted the link to the schematic The switch itself is indeed only 3 wires. I can confirm this, so the schematic is correct and actually draws each wire...it's a shame I didn't keep some pictures of it from under the dash I took a VERY long time ago when I was running some wires and couldn't see what I was doing.

The transistor symbol used throughout that manual usually denotes an electronic circuit or computer of sorts. That one, 118, is for the ABS no doubt, but I'm not sure how or if it's involved in the actual switching of the lights themselves. I recall the wiring for that pedal switch being decently sized, so that's what further makes me think it's a direct command to the lamps and not a control wire to any computer, though we cannot dismiss that entirely until I maybe tear apart the interior and follow the wiring loom from the lamps to the fuse box....which I'm not ready to do just for that LOL

On a side-note, assuming it's just a regular plain-ass toggle switch, makes me wonder how the ABS computer is able to distinguish between a normal pedal press and a press that requires activation of the ABS module...I can only assume that it does some sort of speed sensing on the wheels to determine if ABS is necessary, since simply applying power from that switch would cause it to come on all the time, even when not needed.....yes, a lot more complex than it looks. People used to make fun of this car being the most common and plain car you can possibly get where I live and in some of the surrounding countries, but it actually IS packed with a bit of tech

[Dannyx]

Re: Help with LED brake lights

This is what the lamps look like BTW...about time I posted an actual picture Not that it'd matter too much but anyway...

[WildPuppy]

Re: Help with LED brake lights

...
On a side-note, assuming it's just a regular plain-ass toggle switch, makes me wonder how the ABS computer is able to distinguish between a normal pedal press and a press that requires activation of the ABS module...I can only assume that it does some sort of speed sensing on the wheels to determine if ABS is necessary, since simply applying power from that switch would cause it to come on all the time, even when not needed.....yes, a lot more complex than it looks. People used to make fun of this car being the most common and plain car you can possibly get where I live and in some of the surrounding countries, but it actually IS packed with a bit of tech

You are right, the transistor symbol denotes "smart/complex" circuitry and in our case it stands for the ABS sub-module.
It is connected to the brake switch to sense when the driver wants to lower the speed and then combines speed readings from all 4 wheels and one accelerometer (usually found under the front right seat) to see if it needs to react or not. I used to drive two generations of very similar cars (Dacia Duster) and had some problems with the said accelerometer sensor... therefor I learned quite a bit about it.
ABS is only "reading" the status on the line (HIGH or LOW), it is not outputting anything to the lights in our case as far as I know.
In theory, the main computer ("borrowed" from Renault Clio) is internally fitted to sense broken light bulbs but that feature is not (yet) used by Dacia in any of their models.

Back to the topic, I have a funny feeling this will turn out to be very simple to explain/solve but somehow we are chasing our tails on it but that simple explanation still eludes me...
Did you try the suggestion about the reversed +12V wires ?

[Dannyx]

Re: Help with LED brake lights

Did you try the suggestion about the reversed +12V wires ?

Nope, I haven't. I shall this weekend.

Assuming we simply can't get to the bottom of this, I'm thinking of adding a transistor on the brake wire. We can't toggle the ground, since that's permanently connected, so we can't use an N-channel FET like I originally though, since that one switched the GND. A P channel seemed to work at first, but I ran into two issues: the body diode will allow the brake terminal to get connected to whatever mysterious force causes it to switch off when the brake is not pressed, so that's not good and second, to turn on a P-channel you need to apply GND to the gate....which is not possible since the pedal switched VCC.

Come to think of it, there's another thing I just thought about: I'll try replacing BOTH lamps at once to see what happens. The theory is that having just one LED lamp like I had so far for testing allows the filaments in the other two to act as a low value pull-down resistor for the brake terminal, which, as we've established, kills the whole lamp for some reason. Replacing the other one will hopefully increase the resistance to GND enough to keep it from shutting off. There will still be one filament bulb left for now - the one in the window stop light - but just like on the bench, I can only hope that ONE filament in parallel will not be enough to pull down the brake terminal...will see what happens. If it works, I'll replace that one too...not too difficult. I'll first unplug it to see if it indeed makes a difference. I already blew a couple of fuses when my meter probes or other silly wiring I had lying around slipped and made contact with the wrong stuff in all the wrong places

[Longbow]

Re: Help with LED brake lights

Key point: unknown manufacturer, item purchased on auction site.
From personal experience, there is a wide variation in the quality of discreet led's as well as with automotive LED modules. Some work OK, but burn out quickly. Some don't work at all, like yours.

The point: save a few bucks on the part, but be prepared to waste countless hours of your time trying to reverse engineer your vehicle's electrical system, or figure out why the part you have won't work in your particular vehicle.

Someone has tried this same thing, and has the answer. Grab a friend who speaks French and can post a question to a French auto forum. Or alternatively continue troubleshooting until you become an honorary Renault engineer.

Just to satisfy my curiosity, confirm that the LED unit you have in your hand will light up if you parallel the correct power resistor across the brake light socket. Success means that the car is smarter than you, and knows you are trying to mess with the designated lamp type. My own vehicle had a lamp-out relay module which monitored all the lights. Fortunately all the sensing connections could be shorted together, disabling the unit. Now, I have to personally look to see if any exterior bulbs are out, but I can use LED modules without any issues. Yes, that is like asking someone to get up off the couch to change channels on the TV.

[Dannyx]

Re: Help with LED brake lights

I tried it again with both LED lamps in AND removed the middle bulb...unfortunately no, it works exactly the same: no light at all. The ABS module is interfering somehow I reckon...there's no other reason I can think of at least :|

[WildPuppy]

Re: Help with LED brake lights

Bummer!
No light on brakes, I get. Bu no light on the other "filament" ? That should be straight 12V, no protections, no crap.
Further studying of this matter taught me that there are different variations of control relays, ones that are simple ON OFF, ones that ground the live wire on OFF positions, etc.
It seems you have the later option installed in the car, the kind that grounds the OFF position. Or am I wrong?

[Dannyx]

Re: Help with LED brake lights

No, it's the other way around: brake is fine, parker doesn't work. The LED lamps don't light up at all if the brake terminal is grounded - I confirmed this on the bench - although it doesn't make sense on the car, since nothing is actually going to GND...unless the ABS control wire is interfering in some way. I don't think it goes through a relay at all, according to the schematic.

[WildPuppy]

Re: Help with LED brake lights

But in my opinion this behavior should be corrected by installing a diode on the brake light line, before it splits to the three light sockets. That way, when the computer or brake switch (whichever is grounding that signal wire) will not be able to force the wire to ground.
If this causes a problem with the smarter part of the circuitry, a resistor in parallel with the diode should help sense "some" ground, but the resistor needs to be of higher resistance than the internal resistor used in the LED bulb so it does not interfere with the diode light-up.

[Dannyx]

Re: Help with LED brake lights

I already tried this with one lamp and it sort-of works, in that the parker does stay on, but now the brake doesn't make it come on brighter either, so it's only half-way there. Even though ironically it works perfectly on the bench, even with the diode in series and with another bulb in parallel....quite annoying It's mocking me

[WildPuppy]

Re: Help with LED brake lights

I can imagine. I must give some credit to the user(s) that said we will become Renault engineers if we fix this... BTW, if you find us annoying insisting with this topic, tell us and we move to private conversation but that kinda' throws away the "discussions forum" idea.
Back OnTopic:
1. Did you eliminate the other 2 light bulbs from the sockets when you tried on the car?
2. Did you observe if the light was dim or bright (compared to what you observed in the bench test) ? If yes, it may be that the bulb has reversed contacts and when the parker goes on it lights up bright as it should for the brake so when the brake comes on, you see no difference.
If the answer to 1 is no, it may be that the other two bulbs in parallel make for a good ground connection cancelling the signal coming in.

[redwire]

Re: Help with LED brake lights

Do you think this LED lamp has inside a buck-converter or is just cheapo resistors? It acts like cheapo resistors.

Because brake resistor R1 much lower than park resistor R2, grounded brake wire prevents it from working.

Try ohmmeter the LED lamp between park/brake.

[Dannyx]

Re: Help with LED brake lights

BTW, if you find us annoying insisting with this topic, tell us and we move to private conversation but that kinda' throws away the "discussions forum" idea.

No way...if anything, I'M the one who's bound to get annoying for pushing on, but that's half the fun for a tech junkie and a bit of a petrol-head like me

1. Did you eliminate the other 2 light bulbs from the sockets when you tried on the car?

Yes, although I only did the "plug-and-play" test, meaning I removed the other 2 bulbs, screwed in the LED one and tried it out to see if it indeed was a grounding issue. I didn't use the diode AND removed the other bulbs at the same time, that's the idea. I wrote this in one of my previous posts in detail

Did you observe if the light was dim or bright (compared to what you observed in the bench test) ? If yes, it may be that the bulb has reversed contacts and when the parker goes on it lights up bright as it should for the brake so when the brake comes on, you see no difference.

I thought of that at one point and I can't really answer for sure, though I BELIEVE it was dim, so the diode was doing its thing, allowing the parker to come on, but engaging the brake didn't cause it to glow brighter. I shall retest this to confirm, this time taking a marker pen and actually marking P and B on the terminals to know which is which once and for all.

So ATM there's two issues going on, to sum up again and clear out any possible confusion for those who might be joining the discussion later:
1. the parker "setting" doesn't come on (some mysterious force, most likely related to the ABS module, grounds the terminal and keeps the parker terminal from turning on the LEDs at all). This was confirmed on the bench and is not technically a defect: grounding the brake terminal kills the whole lamp, period.
2. if I put a diode in series with the brake wire, the first logical course of action, the parker DOES come on, since the brake no longer "grounds" (and I put that in quotes because we're not entirely sure it's going to chassis GND per-se), but engaging the brake doesn't cause it to glow any brighter either, so it's only half-way there...

[Dannyx]

Re: Help with LED brake lights

Definitely resistors. Either like in your drawing or a resistor divider on the parker wire, with the input terminal going to one end of the ladder "ladder", GND at the other end and the LEDs in the middle...classic divider, simple and easy to understand. I mocked-up a schematic like that somewhere along the way myself, since I immediately thought of the same thing. This perfectly explains issue no.1 (the parker not turning on when brake is grounded), but not issue no.2, why the brake causes no change in brightness once a diode IS added in series with the brake wire.

Actually come to think of it, issue 1 is not completely solved either, just "put aside for now", since we don't know for sure that there's a GND there...both the schematic and the facts (AKA putting my head under the bloody dash and looking at that switch in some weird power-stance/sex position sort of deal ) show there's no GND going to that switch, so the ABS module connected in parallel to the output of the switch is somehow interfering. Unfortunately I can't disconnect it, not even temporarily for testing, because the wire is crimped together with the other one going to the lamps (exactly the same as in the schematic - page 65 or 66 if memory serves).