Any battery can fail at any time. Not unusual to not be able to disengage the electronic parking brake if the battery is too low or dead. Measure your battery voltage at the battery terminals with the engine off and again with the engine idling and report those values here.

 

Thanks Jon. Battery after charging (with very small charger) at 12.6V. After one attempt at starting, down to 11.7V. My meter does say "Amprobe", and does have one setting for Amps, but not sure how I would use it here. Test to see what how many amps are happening with key off to drain battery? (Cannot give you engine idling spec, it isn't starting).

 

The "clamp connection spark" is normal, and is just the sum total of every module in the car trying to take a big ol' gulp of electrons when the power is first applied. It doesn't indicate excessive current consumption (though of course, you do get a spark when that happens, too).

FWIW, I broke down and spent a few bucks on a meter that has a clamp-on DC ammeter, so I can easily and quickly monitor parasitic drain. Even better, it works with a smartphone app to let you read the meter output from your phone from a good long distance away. This is nice because it lets you shut EVERYTHING (including the hood and/or trunk, depending on where your battery is located). The one I bought was pretty inexpensive - I'm not suggesting it's the best option out there, or the best deal, but I like it!

 

Around how much is your normal parasitic drain? I think I can measure that with my Amprobe. Thanks.

 

Just FYI:

Subaru Electronic Parking Brake Actuator Safety Recall (WTV-81) | Evergreen Subaru

Subaru electronic parking brake actuator safety recall description, recall code, and what parts are involved and how long it will take to fix.

www.evergreensubaru.com

 

Luckily that says only for manual, mine is automatic XM. Thanks.

 

12.6-11.7 after one start....sounds like a bad battery.

 

To measure current with a non-clamp ammeter (or multimeter set on current) you disconnect the battery negative lead, and put the meter leads in line (negative lead to the battery terminal, positive lead to the disconnected clamp). One issue is that most meters are fused at 10 amps or less, so if the "connection current" spike is over that, it'll blow the internal fuse (or possibly damage the meter if it doesn't have a fuse). That's why I'm a fan of the clamp-on meters, though not many of them will do DC current (99% do AC current only - you have to be careful choosing one).

 

I looked carefully at the fine print on my meter, and it was fused at 200 ma, so thought it not wise to even try.

 

UPDATE: Finally charged it enough to undo parking brake and get it started. Had somewhere to go, and didn't want to get stuck, so just let it run a while, then took off batt lead and took public transport. This morning it did start, but it took several tries, and what was even scarier to me than a dead battery is that it was not catching at all for the first few times, like no gas or spark. But then she started, and running voltage is 14.35. Let her run for 15 minutes, and am now letting her sit with the battery lead connected to see if it drains. If the brake was for sure the culprit, I could live without it. I just thought buying a car w 200K on it and no history, that maybe the pin in the transmission could use a bit of help (if that is actually the way it's done these days...). No idea of why she didn't even cough the first few times, maybe I didn't give it enough time for the electric fuel pump to fill up or something? She has sat longer periods and still started up fine before.
You all might hear back from me if she doesn't start after sitting with battery connected...
Maybe it is a faulty battery even tho it is marked as 9/24. Guess I could take it by auto store that checks them for free...
Thanks for all the help.

 

Yeah, at 200ma, there's no way you're not going to blow the fuse / meter. Good call!

Just a comment - while idling for a while IS a good thing, it's not really going to put the battery to a full charge, or even close to one. If the battery had been run down severely, that could have affected something that caused that hard start. A common problem is a leaky injector that bleeds down the stored fuel pressure in the system overnight, so nothing much happens until the fuel pump runs long enough to restore the fuel pressure, and that takes a while because the pressure is very high, so the flow from the pump is very tiny. Of course, this should have been happening prior to your current issues, with or without a dead battery.

Given that the battery has only a small fraction of a full charge, your current troubleshooting method is not really at all fool-proof. With only a very small percentage of a charge, the results of waiting to see if the battery droops over a few days will be suspect. I'd suggest getting a "Real Battery Charger tm" and doing a real full charge, and THEN see what happens. If you do indeed have a parasitic drain on your system, there are ways to trace it down, but they take a lot longer than a clamp-on DC ammeter. One way is to measure the voltage drop across every (!) fuse in the car, which can be compared to a table showing the "voltage vs. amperage" for a given size fuse. Fiddly, but free.

 

OY. I am certainly hoping it won't come down to that. I am really thinking it was the brake, but can't find any written precedence for it. Then of course there's also possible alternator, regulator, etc...

 

OK, so after giving it a good charge, I let it sit overnight with the battery cable connected, and it didn't even have enough juice to open a door lock. 3.6V if I remember correctly. So either major parasitic drain or bad battery? I guess the easiest is to charge it back up (hopefully) enough to get running, then take it to an auto supply store and have them check the battery, that would take at least one thing off the list... Since when did batteries cost $200? It's been a while for me...

BTW, I bought a Duracell for my 2003 several years back, and it seemed like a decent battery, but if I left the country for 2 or 3 months, it would be dead when I returned. So I started disconnecting the negative terminal when gone, and it started right up after returning. So maybe Subies in general have big parasitic drain? I even took all the bulbs out of the bottom of the doors, and the vanity lights, just in case those were the culprits. NOT... Is there anything else that does not get turned off by key, besides enough to keep radio settings and such?

 

The days of decent $49 batteries are long gone. I like AGM batteries built by East Penn. I have been able to find them on sale for $160 to $180 if I check around often enough. But that was more than a year ago when I found one on sale at my local NAPA store for my wife's 2006 Jaguar XK8 convertible.

 

Well, I doubt auto stores have Black Friday sales, but I will be gone by then anyway... :-(

 

If the battery dropped to under 4 volts (!) I'd suggest that the battery is well and truly toast. I would get it checked, but would charge it up before doing so, as I believe that the auto store testers can be fooled by a "good but really discharged" battery. That said, it's really hard to drop - and keep - a good battery at that low a voltage. You really (!!!!!) need to actually measure the drain, or you're just shooting in the dark (and if that shooting involves the parts cannon, it can get pretty expensive).

 

So you think I need to go out and buy a clamp-on ammeter and do the one by one fuse thing? How do you find out what the amperage should be for each fuse? Or instead, could I figure out which side of each fuse is connected to positive, and see which ones have voltage on them with the key off? Then figure out what they are connected to?

 

I think it would make more sense to have the option of using an easy-to-use clamp-on (remote reading) meter for this kind of issue. I had some very extremely intermittent drain issues on my 5-series BMW and would just clamp the meter around the battery lead(s) and then check it occasionally (it produces a graph by time, so you can see the shutdown current drain in phases as different systems time out and go to sleep).

Once you know if you actually DO have high parasitic drain, then you can move on to narrowing down which circuit is the culprit. That's done by using a voltmeter (multimeter set to volts) at a very low range. Here's a link to a handy set of charts that will translate the voltage drop across a lot of different type and ratings of fuses into current drain. For example, if you measure 2 millivolts (that's 0.002 volts) across a 30 amp mini-fuse, it would correlate to that circuit drawing 1081 milliamps (or 1.08 amps). You can pretty much ignore any circuits / fuses that aren't powered with the key out, as they can't be contributing to your battery going dead overnight.

 

Yes, I was figuring that I would just concentrate on the circuits that had power with the key out. Though I am not sure the points on my meter probes are small enough to reach down into the fuse socket.
Another update. Charged it again, unplugged battery cable, few hours later hooked up cable and drove a short distance, was somewhere for 4 hours with battery cable unplugged, came back and hooked up cable, and it did the same thing again - turned over several times like it had no gas, waited a moment, tried again, and it started fine. Some kind of fuel pump thing when the battery is disconnected?

Also, it looks like the alternator was recently replaced, it's a whole lot cleaner that components next to it, and bright new bolts. But there is a bunch of serpentine belt dust on it already, is that normal? Do those normally shed some, or is it a sign of aging belt? Think you can see it on the second photo. So I imagine the battery is getting a decent charge from this new alternator. Of course there are also some other possible culprits, but I am pretty darn sure this is parasitic drain of some sort... Because the battery gives a good hefty start even after sitting for hours, but only with battery cable disconnected, otherwise it's dead in a few hours. Also, what would cause a parasitic drain to occur suddenly? I think I was wrong to suspect the parking brake, that might have been a coincidence.

 

If the battery is dropping from 12.x volts to 3.x volts overnight it is likely something is turned on in the vehicle (no not the parking brake)

Have you taken the time to check if the small light in the rear cargo area is turned on? The switch is easily bumped, turning on the light and flattening the battery.

This is the number one reason for flat batteries overnight in an Outback and you will not know the light is on until you physically check it.

Seagrass

 

OMG!! Seagrass nailed it! I have only had this vehicle a month, and never even noticed that light, because on these newer models it is hidden behind a big bump in the roof. But I did recently move a pretty big piece of furniture, and I bet it popped that switch into the ON position! So I looked, and turned the switch to middle position. When I put on the battery terminal, MUCH smaller spark, almost undetectable, and this time I finally remembered to put key in ON position for several seconds before starting, and she started right up. I guess the fuel pump may be something that stays connected to battery in OFF position to keep fuel in the system or something. Anyway, PROBLEM SOLVED! (Unless it occurs again, but I bet that light did it. That's why I took the bulbs out of the doors of my '03, they would stay on if the doors were open, and if I was working on the car with doors open it would drain my battery. Funny how a small 12V bulb can drain a big battery...).
THANK YOU! And everyone else! Subie Forum Rocks!

 

A newer, good quality battery can be brought back to life by using a quality charger equipped with a desulphination mode. This puts a controlled high current into the battery to blast away lead sulphide deposits that formed on the plates during a deep discharge cycle:

H2SO4 + Pb —> H2O + H2 + O2 + Pb + PbS*
(Fully charged). (Deeply discharged)

The charging cycle is pulsed to avoid damaging the plates or overheating the battery. It doesn’t always work in batteries subjected to several deep discharge cycles or those with low quality plates.

*not a balanced equation but representative of the elements and compounds that form during the deep discharge of a lead-acid battery. Lead sulphide (PbS) is a harmful deposit that affects battery performance, particularly in the area of current delivery.

 

That sounds like a possibly expensive charger. Can you bring it to an auto store or somewhere like a shop to get it charged?

 

Such chargers aren’t expensive… maybe $100-200 US. Not expensive compared to a new battery or a repeat service charge. Peace of mind over turning a battery over to a kid at AutoZone so he/she can do god knows what to it, with gods knows what charger, in the back room.