RedimprezaOB

Ok. After way too much fooling around with this.........i.e., hooking up monitoring lEDs on circuits and then eliminating each circuit yada, yada, yada. Dang. So, the switched and dimmable illumination circuit comes in on a violet wire. Works great no issues. Just because the LCD backlight is a 'light', the circuit was in question. Removed that circuit / wire from the plug at the radio and still had the problem. Over time removed all wires except for the constant 12v supply for the memory (think mine was a Blue w/ Red stripe) and the ground wire. Turns out the constant 12v supply comes in from a circuit off the Interior Illumination circuit. Thus, when the Interior illumination fuse is removed under the hood the LCD backlight turned off. Long story short, issue is internal to the radio itself. Good grief. All that to know the issue was the radio!

Yes. I waffle on the fence by the hour. It's the radio head. It's something else. Seems to now have gone intermittent so I'll work with it a bit more and report back if I find anything new and of course the outcome. Thanks for the ideas so far.

Well, had some time to think about it a little more and do some testing the meter as well. Meter shows the illumination circuit is of (0 V) when the key is, and wiggled a bit. It only comes on when the key is in run position and lights are on. All other dash lights operate normally. Interestingly, radio properly plugged in or disconnected, when key is in the run position with the lights on, the clock looks normal. Flip on the parking lights at this point and the clock gets "flashy" - perhaps like a fluorescent having a problem turning on might be a good description. So, disconnection of the radio just eliminates the drain through the LCD. There is still some issue. This one might be beyond me. Wouldn't think it would be the illumination module as the other dash lights operate properly. I don't have a great schematic but many of the dash lights appear to be on the same circuit. Better sit down with the shop cat discuss this over libations.....

See where our minds go?! I was just thinking of that and looking at where I would add (the connections) a transistor that would accomplish the task. Hand free and thoughtless operation.

Funny the question about the CD! When I was working on it in the garage (nice and quite) and connected the power again I heard the CD cycle. I thought to myself, this better not be the case. But, it wasn't. Whew!! Don't get me wrong, it would have been nice to have resolved at that point. So, yes. Pulled the radio and then removed the illumination pin from the circuit and reassembled. No, current draw. I'm concluding that the issue is in the radio head itself. Now, to find something nobody will want to break into it for when I leave it and am out on bike.... or as suggested maybe I don't need a backlight tuner screen most of the time Thanks for the ideas. This car has been an absolute electrical pill pretty much since 30k miles. Guess we need something to challenge us!

Thanks for that insight. The glove box was the only thing that really was causing me (and the cat) to think something outside the radio might be amiss. However, with key on / lights on it works. The actual radio illumination lights work as they should with the light and dimmer circuit. Just the LDC backlight that never turns off. Thanks again.

I'm trying to run down an odd electrical issue on my 2004, impreza ob. This popped up several weeks back just before vacation so removing a fuse to keep the battery from draining has been the solution until I had time to look at it. (Searched past discussions but didn't locate this issue) Symptom: The LCD 'backlight/illumination' never turns off on the OEM (never touched) radio. Pulling the radio fuse (fuse 9 under the dash I think) does not turn it off. Removing the Illumination fuse (12 under the dash as I recall) does not shut it off. In fact, I removal of any single fuse under the dash had no effect. I have to remove the "interior illumination' fuse under the hood to shut it off. Also the glove box light does not come on - bulb tested and good. I'll test the glove box switch short.....ly. Schematics don't show but recalling how the LCD backlight performed prior to this issue I would assume it is powered by the interior illumination circuit and the issue may be internal to the radio so a replacement may be the fix. But that doesn't explain the glove box issue which occurred at the same time. I'd hate to throw down on a replacement stereo and still have the problem..... Any thoughts, similar experience, better schematics? The shop cat has been no help on this one choosing to nap on the manual instead of providing any real insight.

I probably shouldn't say amything because I own a 2005. But, I was at my sending units last year at this time. In this newer car the closer to 0 ohms the closer to full the reading and 90 to 95 Ohms was empty. ...rrrr the other way round. Point being a Megaohm seems really high. 50 ohms put me at a perfect half tank.

Man am I glad I'm not the only one this has happened too! You can bet the booger was #3. I did what GD discibed, a pair or 40'ish degree needle nose did the job. Hope your back feels better.

Well it sounds like you know what is not causing the problem now! From what you have written, this is sounding like an electric issue. Grab those schematics and find the coolant temp circuit in.....engine electrical I think. At least that is where it is for the 04. I would trace the circuit in the car from the meter assembly to the sensor. Probably a short to ground. The temp sensor is a variable resistor so you could get a reading off your old one or disconnect and measure the new one then work through the circuit to see if the meter is getting the same reading you got at the sensor. Also I think the ECU monitors this circuit, again it does on mine but you'll have to check your diagrams. If so, you should unplug that connector from the ECU so don't get funny readings. Good luck. I finally ran down my electrical gremlin in the meter assembly PCB. I suspect though that is rare and an abraided / rodent chewed wire is usually the culprit.

I can surely attest to DEET being able to soften and make plastics / rubber very sticky!

Doh, must have read the pricing incorrectly. For 40 more I get the fuel temp gauge assembly and no messing around. Part will get here late this week. Thanks

Your right, it was pretty straight forward. It is just so brittle..... I see I can purchase just the gauge component or the fuel/coolant temp gauge sub-assembly. Tried removing the indicator (needle) but didn't want to go too far. It didn't budge. Ever remove the needle and is it indexed. I ask because if I go for the component instead of the sub-assembly I would need to remove the (functional) temp indicator as well as the fuel indicator to be able to remove the mylar guage face in order to get at the screws holding said component.

Hi Guys, Finally had time to run down a electrical intermittent that has been causing an 0462 fuel circuit voltage low issue. So I have it pinned down to the gauge cluster, who would have ever thought. ?Has anyone taken the cluster apart (2004 impreza OB (sport))? Looks like a pretty easy task but that thought has back fired before! Once I'm in is the fuel gauge a replacable unit? Thanks.

Hope your having luck finding the noise! As for what that thing actually is, it is a horn. On models with A/C one horn sits in front of the a/c accumulator. Strange place for one I thought. Check the Haynes manual 12-21. I also checked my car, sure enough....

2004 Impreza OB, just under 50K, stock as they come. 30 mpg around town, 35 on the highway unless the bike has to go on top, then 33 highway @65'ish.

So.... I have had good luck removing pins on connectors throughout the car. However, on this connector (green) I released the strain clip (bail on the wiring side of the connector, part of the connector) but the terminals are so small and access rather tight, I have been unable to locate the lock tab on the terminal itself.

Anyone have experience removing the terminals from the harness connectors at the guage cluster or the ECU?

There was quite a thread on this subject on another site in the fall. Can't seem to lacate it right now. By the end of the thread someone had been in touch with the makers of Peak. Peak Global Lifetime and Peak Long Life were matches to the subaru coolant minus the stop leak additive. I have been running it since 2004 when some crazy porupine gnawed some hoses......

No real development on this issue as of yet, chores, bicycles, whatnot. However, I ran an interim harness as described in the first post from the connector under the seat to the R58 connector on top of the fuel tank. The level sensors were not connected opting instead for my ohm gaunbe.... Been getting good use of this word...thanks Nipper! For some reason I opened connector R59, the sensor on the sub tank and went about my errands. 3 drive cycles CEL cleared and there had been no low fuel warning flashes. That was odd because it was not even in the active circuit. Drove to emissions and got my new tags! Not a 1/2 mile from the emissions center on the way home CEL and low fuel lamp!! Doh! Intermittents are no fun. So, you guys think there could be some type of grounding issue? I'm reasonably sure I have not seen a ground strap from the tank itself. I'll try running a ground but from what I know the FP has its own 12V circuit. That obviously works. The Thermistor circuit I cannot find an exact circuit schematic. Some show One line in and connectivity to a fuel level sensor. I have 2 wires for FP and 2 wires for level sensor and 2 wires I assume are for the thermistor on R58. All schematics show the thermistor run to a "micro computer". So, if the thermistor is really a temperature controlled resistor, is there really any reason I couldn't place a resistor gaunbe (again thanks) into the circuit to eliminate the real one as a culprit??

So, this is a topic somewhat beat to death in some forums, but different.... My 04 impreza OB threw a CEL. I also noticed the low fuel indicator would come on and go off many times in a 20 minute drive....never mind a full or half full tank level. Code 0462 - Fuel Level circuit, low voltage input. Did the ohm meter on each sensor and the sensor circuit putting them in series, checks out ok. Dealer says rodents likely gnawed the harness and want 10 hours to pull the tank. I made a jumper harness to go from connector under rear seat to R58 on fuel tank. I also jumpered the level sensors completely by removing them from the circuit and adding 50ohms across the leads at the connector under the rear seat. No go, still getting the intermittent low fuel indicator. ??Anyone know what the voltage should be in the low fuel thermistor circuit?? But oddly, the fault code is the level sensor. ??thoughts, observations?? Thanks