Repairing interior light circuit in Ford instrument cluster

[Skauber]

Hi!

I'm new here so go easy on me.. :p

So... I blew up something in the soft on/soft off interior light circuit in my '05 Ford Expedition. The circuit which controls this is inside the instrument cluster. What happened was that I was attempting to brighten up things a bit by swapping my interior lights and puddle lights (same circuit) with LED bulbs, I bought the bulbs off Deal Extreme, chinese stuff. For some reason, these LED bulbs, while apparently working just fine, blew something in the dimmer control circuit in the instrument cluster. The output is fine, the LED lights up as they should, and now when I put incandescent bulbs back they also light up fine as they should, but there is now a constant power draw on the circuit of about 460 mA even with the lights off, ignition off and so on.. This drains my battery, and I had to pull the fuse feeding this circuit which renders me without interior lights. I took the instrument cluster apart to try to find out what is going on, and see if it's repairable without having to replace the whole instrument cluster, and I narrowed it down to a couple of FETs and a few caps which might have something to do with this, using the wiring diagram for the connector and tracing the PCB. I measured the caps in circuit, the smaller ones were identical but the bigger ones were different. I know I can't get accurate values with them in circuit, but the fact that the two are different tells me that one of them is out of spec. There was another one on the same PCB on a different circuit which had the same value as one of the bigger.

Well, after that long winded text, here's a couple of pics to show what I'm up against (click for bigger size):







What I would need some assistance with is:

1. I need to identify the two FETs if possible from the pictures, and the 4 caps. I've tried google on the caps, but the datasheet I found didn't seem to match 100% with the marking.. I.e. for the larger caps, is this a 330 uF? Or a 4.7 uF in the 330 VFK series? On the FET I found a datasheet, could it be this? http://pdf.datasheetcatalog.com/datasheets/90/318399_DS.pdf

2. Why would connecting LEDs to this circuit cause any issues? I don't understand why a LED would be such a big difference to power compared to a incandescent bulb to the point where it would damage the circuit... Any thoughts on that?

I highly appreciate any help on this.. Thanks!

[amyk]

If the LEDs have active electronics in them they could be a capacitive load, while incandescent bulbs are pure resistive loads.

It looks like that heatsink is pressed on, you'll have to pull it off to read the marking correctly.

[Skauber]

Thanks for your reply!

I've looked a bit into it and to me it seems like it's the Infineon TLE4274 V50 low drop voltage regulator. I need to take it out to verify the number, but so far it seems to be. You can kinda half of see the Infineon logo on one of the pictures. http://www.infineon.com/cms/en/product/power/linear-voltage-regulator/linear-voltage-regulator-automotive/TLE4274+V50/productType.html?productType=ff80808112ab681d0112ab6dc83713e2

For the caps, they seem to be 330 uF and 220 uF, FK series Panasonic SMD aluminum caps. I need to know, does it matter if it's specifically the FK series of caps, or could it be a FT series or FP series without causing any issues? I found a 330 VFT 4F7 cap on RS Online, could that be used? On the 220 cap, I found multiple with the 220 CFK, but can't seem to find one that matches the 4C8..

What I understand is

220 = capacitance, uF
CFK = FK series, not sure what C stands for, or what's the difference between a VFK and CFK..
4C8 = ?? What does this signify? Package size?

And lastly, it makes sense what you say about capacitive load. The LED bulbs I got has a PCB with a heat sink on the back, and can sense polarity so it doesn't matter which way you insert it. The puddle lights also has some SMD components on them, looks like small caps, so that is probably the cause.. Does that mean I cannot use LEDs on this circuit at all with the type of voltage regulator that is feeding the circuit? Or do I simply have to find the correct type? And, what is most likely the damaged component? One of the caps, or the voltage regulator itself? I'm planning to swap out all those six components, but perhaps that's not necessary? The caps are connected between the ground pin and one of the other pins, so if one cap is damaged I suppose that would allow a current flow to ground, thus causing the draw I'm seeing from the battery. That could be said about the voltage regulator as well, allowing current flow to ground, but since it seemingly is still turning on and off the lights as normal with soft on/soft off, the voltage regulator must be functioning, right?

[Skauber]

Took it apart again today and checked it, and it is indeed the Infineon TLE4274 V50 low drop voltage regulator. The weird thing is though that this is a 5V voltage regulator, which doesn't exactly make sense as I would expect the output to give 12V to the interior lights when on, so need to investigate a bit more. Very hard to follow the traces without any wiring diagram, and so far I've only been able to follow the 12V supply pins from the harness connector for the instrument cluster, the output pins doesn't beep on anything around the connector so it's difficult to follow them anywhere.

The circuit is supposed to feed 7 bulbs, 5 interior and 2 exterior puddle lights. Total wattage would be about 30-50W or around there, which makes it draw up to about 4 amps, which I'd suspect would have fair sized components to handle, and those two voltage regulators with the associated caps are the only components with heat sinks and with a decent size on them. There are two quite large diodes on the board as well, but these seems to handle the main power to the board as they are connected to the main 12V trace and the ground plane. Before diving more into the rabbit hole, I'm going to put it back into the car and troubleshoot the wiring a bit more to make sure I'm looking at the right place for the power draw issue...

[Clear as mud]

You say you measured the capacitors.  How did you check them?  Only with an ohm meter, or some sort of capacitor test meter?

I am doubting that this is really the circuit that controls the interior lights.  A 5V linear regulator like that one is probably there to supply current to a microcontroller or other logic that runs on 5V DC.  It's possible the microcontroller creates a PWM signal that goes to a power transistor to dim the lights.

Also, without knowing anything about the Ford Expedition, I was doubting very much that the interior lights are controlled by anything inside the instrument cluster.  That would not be something any normal car manufacturer would do.  So, I spent a few minutes looking it up.  It turns out that late-model Fords have something called a "Smart Junction Box."  See post number 15 in this topic on a Ford F150 forum.  And here is a description of where the part is located.  Also see the pictures in the ebay listing 03-06-NAVIGATOR-EXPEDITION-UNDER-DASH-FUSE-RELAY-BOX-OEM-4L1T-14A067-AC-5022.  See the green around the little relays in the middle?  I think that is a printed circuit board in there, and I think the FETs for the interior light dimming are on that board.  I would be very surprised if they are really inside the instrument cluster.

[Clear as mud]

It is possible the 460 mA current draw is a problem unrelated to your replacement of the interior lamps, and it is just a coincidence that it happened at the same time.  Or even that it is normal for this vehicle.  See if you can find another Ford Expedition and check to see if it also draws current when turned off.

[Skauber]

Hi! Thanks for taking the time to help me out.

What I'm using when diagnosing this is the factory shop manual from Ford, which has wiring diagrams for the entire vehicle. It shows the circuit going from battery "hot at all times" feeding through fuse 21 in that relay box you linked to, then from there straight to 2 pins supplying voltage to the instrument cluster. Then it has a rudimentary drawing of a relay, and shows the circuit to go out through 2 pins from the instrument cluster and then to the bulbs. Yes, it's the very first vehicle I've had or heard about that uses an internal circuit in the instrument cluster to power the interior light, but according to the wiring diagram in the shop manual, they actually do this.. Here's the wiring diagram showing this (click for bigger):

It starts from battery, goes through F1.21, fuse 21:



After instrument cluster, follow B, the other circuit coming from fuse 24 has to do with the manual on/off supply, the circuit on fuse 21 supplies the circuit for auto on when door opens or unlocks and so on.



Goes to puddle lamps, then follow C.



And this last shows the interior lamps. Nothing else in the circuit, no relays or anything...

And no, 480-ish mA drain from battery is way above spec. Factory spec for battery drain when everything is off, and modules are in sleep mode should be less than 50 mA, when pulling fuse 21 it reduces current drain to around 15 mA.

The way I measure the caps is with a DMM with capacitor capability, I have two of them. I measured them in circuit, which I know will not give me an accurate value, but it's a quick step to check if something seems to be out of place. Since I have two fairly identical circuits with the voltage regulators, then it could be assumed that measuring them would give roughly the same value, which they don't. The two smaller give the same value, the two bigger has different value. In order to confirm a failed cap I would of course have to desolder it and measure it out of circuit. Right now I'm attempting to do as much as I can without desoldering anything, until I am sure I'm on the right track and not chasing the wrong rabbit...

[Skauber]

Well, this is a bit interesting. I may have been chasing the wrong rabbit.. I installed the instrument cluster in the vehicle again, measured battery draw of around 480 mA. I then proceeded to disconnect the connector marked as C238 which effectively disconnects the interior and puddle lights from the instrument cluster, and the battery draw was reduced to around 45-50 mA. Still on the upper limit, but acceptable. So, I am beginning to think that this is not due to the LEDs, but rather an issue with wiring to the lights. It can't be a short to ground, that would blow the fuse, but something is definitely amiss with that circuit. I've removed all the sockets for interior lights, and the only thing left is the socket for puddle lights. I will have to check it further there, might be a short to 12V+ or something which backfeeds the circuit and causes a power draw.

[Skauber]

More progress, in a process of elimination the fault has now been isolated to the instrument cluster. Disconnecting the previously mentioned connector must have removed power to parts of the instrument cluster and in that way reduce the power draw.. The pinout diagram for that particular connector just references circuit numbers, and I haven't cross referenced it yet (big connector). But, I took the connectors to the instrument cluster, removed the output pins to the interior lights from the instrument cluster, power drain was still there. I then removed the power supply pins for interior lights coming from the fuse box and supplying 12V constant power to the instrument cluster, and power draw disappeared. In other words, the only thing that draws an excessive amount of power on that circuit is the instrument cluster, and that power draw is there regardless if something is connected to the output or not. And the fault must then be internally in the instrument cluster. For now, I've temporary fixed the issue by pulling the fuse on that circuit, and I'll take time this weekend to once again check the PCB, and I'll thoroughly check the components that are able to handle a big load. My primary suspect would be the caps, so I'll start there and see if one of them traces to the outputs on the connector.

Why Ford designed the interior light circuit like this is really beyond me, not a smart move, specially when it seems to be very sensitive to what is connected to it. I'd think an external relay, perhaps a solid state one with a dimmer feature, would make much more sense as it would probably be more robust. Doing it like this, and then have people replace a whole otherwise fully functional 500-dollar instrument cluster is pure evil... lol..

[Clear as mud]

The bottom line of letters and numbers on those capacitors is just a lot number.  The datasheet explains the markings.  If you want to replace the capacitors, the exact type probably doesn't matter that much, just match the capacitance rating and match or exceed the voltage rating, and get one with the same footprint.  You said it is one of the 330 uF ones that you think is bad.  From the picture, it looks like this is a capacitor on the input side of the voltage regulator, so even if that is going bad, it probably doesn't affect the circuit too much.  At some point the regulator may cease to function properly, but the capacitor can probably degrade quite a bit before that happens.

Thank you for posting the relevant pages of the shop manual.  The information I was looking at for the other Ford vehicle showed three different FETs controlling the interior lights.  Does yours have that?  It looks like all the lights are fed from the same wire, from the shop manual pages you posted.  I didn't have time to look at it for very long, though.

[Skauber]

Ok, thanks. No, the shop manual wiring diagram only shows the one I posted above in the switching of interior lights. I will take another look at the PCB this coming weekend and try to trace the circuit from input to output. If connecting an incompatible load on a dimmer circuit, is there anything that is a common failure point? Caps? Regulators? FETs? Or is it different from circuit to circuit?

[Skauber]

Well, things are FINALLY making sense! (LEDs killing an instrument cluster does NOT make sense... :p )

After battling with this elusive rabbit hole for a week, I have now solved this issue. It became apparent a couple of days ago, when I came to start my trusty (well, maybe not so trusty the last week :p ) truck on wednesday morning, and it barely wanted to start again. I went to refuel, thinking it would charge on the way there, but after refueling it wouldn't start. Dead battery.. I bought a battery charger (finally got an excuse to buy one), and charged it for 7+ hours at home, at 6 A constant charge the whole time. That kind of gave it off, since it's supposed to go down as the battery charges. I unhooked it, let it sit for 15 minutes and measured 10.5 V, obviously a dead cell.

Replaced the battery yesterday, and today I redid the battery drain test to check if the failing battery was the culprit, as they often do make gremlins appear out of nowhere, and after sitting for about 45 minutes I had a current reading of 15-20 mA drain, jumping to 20 mA every time the anti theft light blinks, and it marks the end of a week of chasing non-existent gremlins..

At least, this shows that it's important not to jump to any conclusions, just because you replaced your interior bulbs to LEDs the night before, and the next morning the battery barely has enough cranking power to start the engine, and a subsequent battery drain test shows the interior light circuit draining power through the instrument cluster, does not necessarily mean there is anything wrong with it, even if it seems apparent that there are.. At this point, running on ebay and buying a new one would not have fixed the problem, it would be a waste of money.

Bottom line, diagnose it properly, in every way and once reaching a diagnose, test that diagnose to make sure it's correct, BEFORE ordering parts.. In my case, a dead battery almost made me buy a new instrument cluster... Gotta tell ya, the timing was friggin perfect for throwing me on a wild goose chase though! haha At least now I have a fresh looking cleaned instrument cluster, as it was a bit dusty inside so at least something good came out of it...


I'd like to thank you for replying and offering assistance, specially on the electronic stuff. I have some knowledge on through hole stuff, but not so much in SMD circuits when it comes to identifying markings and such..