Woke up and the 5 month old battery was dead. Recharged it, and the car started right up. Next morning, tried to start it and it was dead again. Recharged it, disconnected the positive battery terminal, and I have reattached it, started it, detached it and repeated the process for 5 days now. Clearly, there is a power drain. I suspect it has something to do with the radiator fan, which is electric, and wired to allow the fan to go on without a key in the ignition because several months ago I noticed that it would not turn off by itself, but rather, after it had run the usual 10 or 15 minutes it used to, I would turn the car back on, shut it off, and then the fan would turn off. Anyway, how can I go about diagnosing this problem? The car is a 1990 Acura Integra, 225K, so I really don’t want to sink alot of money into it.
You can start by narrowing down the affected circuit. Install an ammeter between the positive battery terminal and the corresponding cable and note the reading. Have someone start pulling fuses one by one until the current level takes a sudden drop. The last fuse pulled will, naturally, be the one for the circuit with the problem. Once you’ve found that, you can go about trying to further localize the problem.
Have you confirmed that the battery is actually getting charged by the alternator?
Before you get to deep, pull the car into a dark garage (or wait until night) and look through the windows and just make sure there isn’t a light on somewhere. While you’re at it, listen. Listen for things like a fuel pump that won’t shut off or a automatic radio ant. motor that won’t stop cranking. This probably isn’t the case, but it’s worth trying.
Whenever I had a problem with the battery draining constantly, the alternator wasn’t working properly.
I have taken the alternator out of the equation by charging the battery with a trickle charger. I have not used the alternator at all for the battery. It is an extra car so I have simply started it and then turned it off, then once again disconnected the cable.
It is much wiser to disconnect the negative battery terminal, and if replacing the battery, disconnect negative first and reconnect negative last. This is because inadvertently touching the cable end, or a tool connected to it or the battery terminal, will not create a short circuit with the attendant nasty spark and arc welding that can happen when only the positive is disconnected.
The ideal approach is using an ammeter as suggested (but between the negative terminal and negative cable end for safety’s sake), but DC ammeters tend to be pricey and not in the average Joe’s tool chest. If there’s a significant drain it will light up a 12V bulb that is wired in series between the cable end and terminal, and this can be used as a sort of “go/no go” ammeter. Assuming the bulb lights up, start removing fuses one at a time as mentioned above until the bulb goes out or dims noticeably. If this happens, you have narrowed the problem down to something protected by that particular fuse. Then you replace the fuse and start disconnecting the items on that fuse circuit until the light goes out/dim again. That’s your culprit.
If none of the fuses affect it, disconnect the wiring to the alternator. If the bulb responds, there’s an internal short in the alternator and it needs to be replaced.
If there’s no response to those tactics, advanced techniques and/or equipment is called for. I’d take it to a shop that is known to be competent in electrical trouble shooting.
Arrgh! My second sentence above was incomplete. It should have read: “This is because inadvertently touching the cable end, or a tool connected to it or the battery terminal, to any metal on the body or engine will not create a short circuit…”
Even if the bulb doesn’t light up, you can still use it as a diagnostic tool, provided you have meter which can measure DC volts. Connect the bulb as suggested, and then connect the meter terminals across the light bulb and select the volts range which gives you a significant reading without “pinning” the meter. You’re not going to measure the actual current this way, since the resistance of the bulb is unknown and varies with the current, anyway. You’re only concerned about relative readings here.
Another way of locating the circuit with the trouble is to disconnect 1/2 the fuses, and leave it overnight. If it’s drained, it’s one of those fuses, so do it again, disconnecting half of those (1/4 of the fuses). Doing this a few times and you will soon isolate the fuse circuit with the problem.
This takes longer, and leaves you with a drained battery each morning, but it doesn’t require you to purchase any tools for this.
I would not suggest doing this, as draining a lead acid battery is very bad on it.
The best method is an ammeter, of course.
You could also try using an LED and 1000 ohm potentiometer in series. (The LED in series with the potentiometer would be inserted between one of the battery posts and the battery cable.) Start with the potentiometer at 1000 ohms, and then slowly crank its resistance down while watching the LED. If you see the LED light up while decreasing the potentiometer’s resistance, you have a current leak. If you never see the LED light up - even when the potentiometer is at 0 ohms - you either have a very very small current leak (less than a couple mA) or you do not have a current leak.
If you do LED and potentiometer, only use two of the potentiometer’s pins… the slider and one of the ends. And make sure the LED is the proper polarity.
Ah guys, the clock, the stay alive current for the computers, and the radio memory all use battery voltage. I have a pretty good feeling that in both of your suggestions a perfectly normal car would indicate a “draw”
I think the clue is in the OP.
The first step would be to charge the battery, and have it tested. I don’t care if it is only 5 months old, it could be bad.
Start the car and measure the voltage at the battery with the engine running. It should be between 13.5-15V If below or above these numbers, suspect the alternator. Change the voltmeter to AC voltage, measure voltage at the battery. If >0.4V you have a shorted diode in the alternator.
Assuming all is OK at this point
Then drive the car until normal engine temp is achieved. Park the car. Listen to the fan. Time how long it takes to shut off. If it runs much past 10 minutes, I suspect that a relay is sticking causing the problem. (I have seen this exact problem on other cars)
If the fan shuts off on time, use an ampmeter as described above.
ONE VERY IMPORTANT NOTE:
You will have to duplicate the conditions when the draw exists. Meaning if you lock the car and the battery goes, dead, when testing you have to lock the car. Ask me just how I came across this little gem of wisdom. :smack:
Hokay-I’ll bite. How did you learn?
A three year old car came in with a didn’t start complaint. I test the battery, it is flat, and of an age where it is probably bad. I grab a new battery (warranty) and put it in the car. Start the car, check alternator output and ship that sucker.
Next AM same car is in the drive with a battery was dead this AM complaint.
Godamn crappy parts department gave me a bad battery. So I get another one. Fill it, charge it, and test it. All OK. I put it in the car and test the alternator output. OK, I check starter draw, normal. I look for a battery drain, there is none. I ship the car and put yesterday’s battery on a slow charge to bring it back.
Next AM guess what was in my service drive? Yup you got it.
I have the shop put the customer in a rental car, as I am convinced that the customer is somehow leaving a light on or something. I replace battery #2 with Battery #1 (which tested good after a recharge). I go over all my tests from the day before, everything is stone normal. I park the car and lock it and lock the keys in my tool box. I want to be sure that nobody fucks with this car just to mess with my mind.
Next AM the battery is D E A D.
I swap the battery again and start going over all my test for a 3rd time. Every single test is perfect. Nothing is out of the ordinary. I even hook up a ammeter (back then a test light was the standard draw finder) the draw is only about 20mA. Which is not nearly enough to cause a battery to go dead overnight.
Anyway it boils down to I am standing in the shop next to the car, looking at the car, and the parking space where the battery went dead the night before. What is different I ask myself?
That’s outside, this is inside. No that can’t be it.
The windows are all up. I roll up the windows , no change.
I locked the car outside. I don’t lock a car I am working on. So I lock the car and all of a sudden there is a 10A draw, due to a sticking lock relay.
I replace the bad relay and all is better.
Learn from the mistakes of others, you won’t live long enough to make them all yourself.
I pondered that as a possibility, but said no-the door lock solenoids wouldn’t handle the constant voltage and would fail. Interesting. Thanks.
You may be right… I have no idea what the “standby current” is when the car is off. I was under the impression that it was less than 1 mA, in which case the LED trick would work. But if it’s in the 10 mA to 20 mA range, you’ll need an ammeter.
On modern cars, 15-30 mA is common, some are close to 50 mA.
On my cars it can take up to 10 minutes before standby is reached. Draw then is 25mA with periodic spikes to 35mA and every once and a while a 200mA spike is seen.
There are lots and lots of computers that do things behind the scenes.