What is draining my car battery and how do I make it stop?

April 6, 2015 5:57 PM

I have a 2007 Mazda 3 with battery problems. I'm looking for: 1. Any thoughts as to what the underlying issue might be (and how expensive it should be to diagnose/fix); and, 2. Recommendations for anglo mechanics in Montreal (specifically Verdun, ideally).

The battery on my Mazda 3 started turning up dead almost every morning beginning in December. I bought a new battery and things seemed to be better for a month or two, but now the problem is back, and worse than before.

Any time the car sits for more than about 12 hours without being driven, the battery is likely to be completely dead, to the degree that the power locks won't even work. If I get a boost or put it on a trickle charger, the car drives fine once it starts.

I've discovered that if I unhook the battery leads under the hood and then reconnect them when I want to drive somewhere, the battery remains fully charged even when left for days, so it definitely seems that something is drawing an unreasonable amount of power even when the car is off. There are no apparent lights or anything that stay on.

What could be causing it? How expensive a problem is this likely to be? And, assuming it's something I am unlikely to be able to easily fix myself, does anyone know of any good, honest, and affordable English-speaking mechanics close to my location in Verdun, Montreal?

posted by 256 to Travel & Transportation (16 answers total)

A friend had a very similar sounding problem with his CRV. After a couple months of jump starts and agonizing, it turned out that the battery leads and connectors needed to be cleaned and tightened (even though they appeared to be fine). After that, all his battery issues went away.

Please don't ask me how to do this, I don't know from cars.
posted by phunniemee at 6:05 PM on April 6, 2015

You have a short circuit somewhere. I had something similar, it turned out to be one of the motors for the power windows, but it could be anywhere. I guess if you are handy with a multimeter you might be able to find it yourself but I was told (by the mechanic that fixed my car) that sometimes it is very difficult to find the cause. Normally it is fairly obvious though.
posted by AndrewStephens at 6:07 PM on April 6, 2015 [1 favorite]

Within the alternator are diodes or rectifiers, which are one-way valves for electricity. One or more of them may be shot. The diagnostic is pretty easy.
posted by jet_silver at 6:08 PM on April 6, 2015

Had a similar but not identical issue with my 2011 Mazda 3. It turned out to be bad connectors on the battery cable. It was a quick and inexpensive fix at my local mechanic. He found it by doing a quick electrical system check. You may want to start there. It cost under $30.
posted by galimatias at 6:22 PM on April 6, 2015

When I had this problem, it was one of the ventilation fan motors drawing power. It was under warranty, so I don't know what it would have cost. I doubt the part was terribly expensive (by car part standards), and the labor for replacing it was not much, but it took a long time to find the actual source of the problem (labor $$$).
posted by primethyme at 6:32 PM on April 6, 2015

How handy are you? do you have access to an ammeter? Test the current on the battery lead, while pulling fuses. See what fuse makes the amp draw go to zero.
posted by notsnot at 6:32 PM on April 6, 2015 [4 favorites]

I just checked a few Mazda forums, and there were a few threads covering a battery drain issue related to problems with the trunk latch and trunk light on the 2007 - 2010 Mazda 3. The latch feels closed, but the light stays on. The way they diagnosed it was to use the security feature on the key fob (hit lock twice to activate and get the horn beep), and if the horn didn't beep, the trunk was being read as ajar by the computer, so they would reclose the trunk until they got the beep.
posted by gimli at 6:45 PM on April 6, 2015 [1 favorite]

I had EXACTLY this problem on a 2007 Mazda 3. Stubbornly got several batteries (under warranty after the first one) over the course of six months, dealership couldn't diagnose even after leaving the car there for days on end. Obviously knew it was a short or a drain somewhere, but impossible to find. All kinds of speculation, testing. It wasn't a consistent set of circumstances that would trigger it, so it was just a mystery. Thankfully they didn't charge me for the lack of diagnosing.

FINALLY they found the problem. Was a short in the power junction box (in the front? I don't know these things). Was not cheap ($1200 in the Boston area), but was a huge huge relief to finally get diagnosed. And I haven't had one second of trouble since then (12/2013). Good luck!
posted by clone boulevard at 6:50 PM on April 6, 2015

Yep, if nothing specific to your car/devices shows up then

You have a short circuit somewhere.

As said above. Pulling the fuses with a multimeter-type device handy will go part of the way to solving the problem if it's happening *every* time you park the car and don't pull the battery leads. Notice, if it's happening *every* time then it's actually much better than if it were random or intermittent because diagnosing the offending wire or gadget is going to be that much easier.

Alternatively you could install a push button breaker on the battery or a switch inline with the offending device/circuit (which, again, you find via the fuse pulling method) as a cheap way of validating your diagnosis and/or solving the problem in a hack-ish, if workable, way.
posted by RolandOfEld at 6:51 PM on April 6, 2015

You almost assuredly have a short. Unless you don't.

My husband and I both needed new batteries (same make and year model) not so long ago, and I'm fine but his was being horrible, which I didn't know until he confessed he had to drive every day or he'd have to call AAA for another jump.

I finally made him take the car to Toyota, assuming they could find the short. They tested the battery (originally replaced by AAA) and found out it was just shitty and they called in and found it was under warranty, replaced it with one out of their own stock for free, and sent him on his way. It's been fine since.
posted by Lyn Never at 7:04 PM on April 6, 2015

I was going to suggest alternator, as well. It's an issue I had once with a good battery. It just couldn't keep the charge.
posted by SpacemanStix at 7:30 PM on April 6, 2015

Best answer: Obviously haven't seen the car, but this "battery dies when connected consistently" problem to me reads as:

1. the battery IS charging, or it would be run down regardless
2. the car IS starting if the batteries been disconnected, so the starter isn't faulty in some way/drawing too much current/etc
3. the battery itself is good, or disconnecting it wouldn't make a difference.

So, as suggested, something is causing load even when the car is "off". The problem is that nothing is really off anymore, in the 21st century.

My pot shot guess, having seen similar problems on several vehicles of family/friends from various manufacturers, is a fucked up security controller/security computer, or if it's integrated in to the main ECU then... that. Some bit of it that's supposed to stop drawing current when you turn the car off and press lock is sticking. It might be a relay, some switch might be faulty or intermittent causing it to loop through turning some light or feature off over and over, it might never get in to the fully off state because of a bad connection or faulty wire or something.

Any time the car sits for more than about 12 hours without being driven, the battery is likely to be completely dead

Means it's not some little light in the dash board. Killing the battery flat in 12 hours if it's really and truly charged up(you're driving for more than 20 minutes or something silly after you jump it right? god i wish new cars had real ammeters to show charge load) and it dies that fast then something is pulling some serious juice. You can run a boombox on a car battery for 8-10 hours and have it still have enough grunt to start the car, much less one little light. Something substantial is stuck on here.

8 years seems like the optimum length of time for modern cars to start getting stupid minor issues like this. At least it's nothing major and mechanical, but it sucks, because if it is one of the computers/controllers i bet it costs like $3-500 or even more for the part unless you can source one on ebay or something. And the labor sucks, if any wiring itself needs to be replaced in the harness or elsewhere around it. The total bill on my dads subaru was around 1k, for exactly this issue.

A friend of mine, knowing full well he needs to swap the whole ECU(once again, same exact issue), has been connecting/disconnecting the battery on his nissan Z for several years now. He could buy the parts and fix it, he has the money, but it just pisses him off too much. Last i talked to him i think he was about to rig a high amperage switch in to the battery line for like $20 with some crimp connections and heatshrink... I have a switch like that on my bus/RV because something in the ECU never quite sleeps right either anymore and... does the same thing.


On preview, the pulling fuses suggestion is good. The problem is that when i've encountered this problem in the past, to make it stop drawing current you had to pull the fuses that you needed to drive the car. Like the ones that powered the ECU/security computer and let the starter interlock work, or let the ECU run at all, or that were shared with the gauge cluster, or something. It might be helpful in diagnosing it, but it wont be helpful in "oh, i can just drive for a while without a stereo!" being the outcome more than likely. Who knows though, you might get lucky.
posted by emptythought at 3:02 AM on April 7, 2015

To clarify, the pulling fuses suggestion is to identify, by a process of elimination, what circuit is causing the problem. With a bit of luck, a lot of patience, and a multimeter (or circuit tester) you then go over everything in that circuit to find the culprit.
posted by GeeEmm at 4:11 AM on April 7, 2015

Best answer: 1. the battery IS charging, or it would be run down regardless
2. the car IS starting if the batteries been disconnected, so the starter isn't faulty in some way/drawing too much current/etc
3. the battery itself is good, or disconnecting it wouldn't make a difference.

Agreed.

OP: You have done good basic problem solving and I think we can be confident that this is a 'weird current draw' as you concluded. This is a complex issue, so all the people guessing about battery connections and alternators and the like with no automotive knowledge can probably stop, now.... ;)

The problem is that when i've encountered this problem in the past, to make it stop drawing current you had to pull the fuses that you needed to drive the car.

The fuses are only pulled when the car is parked (when the draw is causing a problem). The draw doesn't matter when you're driving as the battery and alternator are clearly more than good enough to overcome the draw when driving or it'd grind to a halt after 5 miles anyway.

Pulling a fuse (start taking notes to do this a little scientifically/smartly) is a really good method to try and find out what it is. If you just want this dealt with, the dealer or (more likely) a good local specialist for vehicle electrics is your best bet. BUT he'll have to crawl all over it with a multimeter and work it out anyway, which is expensive, so if you can be bothered to save some money, here's what I'd suggest to shortcut the cost of that process:

1: Buy a booster pack and leave it charged in the house at all times or in the car if you are parked for long periods (or overnight somewhere. This is so you can always start the car after your experiments.
2: If you have somewhere important to be, disconnect the battery when you park rather than tale the risk!
3: When you have the time (ie the car will be parked long enough that it would definitely have gone flat before). (What I'd do) is to pull 10- 25% of the fuses (assuming two fuse boxes and a LOT of fuses) until you find the 25% that causes the problem. Then pull half of them one night, half the next night until you establish what area the draw is from. So you are narrowing down the problem to within 25% of the fuses, then to within (50% of that 25%) and so on until you get to where you need to be. Doing it one by one will take FOREVER.
4: Record battery voltage at the battery before parking and after returning. Record ALL of this and which fuses were disconnected each time.

So basically, park. Pull out 25% of fuses (noting position - camera phone/photo and notes etc). Leave car for enough time for battery to have gone flat. If it has gone flat, recharge and next time you park pull next 25% of fuses, replacing originals. Do ALL the fuses like this to make sure that it is only one circuit that is the problem. Once you have isolated it down to one quarter of the fuses, start the same process of subdividing within that subset of fuses until you know the fuse/circuit the draw is on. Then either repair it or take your notes to a dealer/specialist and tell them you have a draw on on that circuit and can they fix it (if it is past your time/effort/skill level to fix).

Does that make sense? I've only had one cup of coffee so I am happy to clarify.
posted by Brockles at 5:53 AM on April 7, 2015

Test the current on the battery lead, while pulling fuses. See what fuse makes the amp draw go to zero.

This is a good idea, too, if you are good enough with an ammeter. Also done when the car is parked and not running. There will be some small draw on some circuits, but if you can isolate which one it is that way because there will be a bigger drop in current with the fuse pulled on the bad circuit. If you get two with a big difference/current drop pulling the fuse, then you can test it with the fuse removal technique next time you park to see if the battery does go flat.
posted by Brockles at 5:58 AM on April 7, 2015

It might be simpler to pull all the fuses and then put the ammeter leads in place of each fuse until you find the circuit drawing significant current.
posted by exogenous at 6:27 AM on April 8, 2015