Help finding short in car

Thread Starter

Volttrekkie

Joined Jul 27, 2017
63
The battery in my Accord runs down slowly. Takes about 3 days. I have to disconnect it when not using car. Sometimes I forget and 3 days later the battery is dead and I have to use my battery charger. Finally, I got out my multimeter and searched.
I connected my meter in series with the negative terminal of the battery. It read .3A DC. That is the parasitic load? I opened the fuse box under the dash and pulled out all the fuses one by one. Nothing. I did the same with the fuse box under the hood. Nothing. .3A DC all the time.
What do I check next?
 

ericgibbs

Joined Jan 29, 2010
18,766
I suspect his alternator rectifiers, its not charging the battery fully when running also its discharging a partly charged battery in 3 days.
 

Thread Starter

Volttrekkie

Joined Jul 27, 2017
63
Either I measured it wrong or my meter is not good enough. So, I started again. Here is what I found:
Connected in series at the negative terminal, I am reading .06 to .1 amp. It keeps changing. This morning it was .06 amps. a little later it was .08 amps. Now at 2pm it is .1 amp. Battery was disconnected all this time. Florida. Very hot.
Unplugging every fuse one at a time did not change the reading more than .01 amp. Probably just a glitch. I found the connectors to the alternator and disconnected them both. Amp draw did not change. Mechanic on you tube says it should not be more than 50miliamps.
I am thinking bad battery. But I think it was changed.
Car starts and runs fine.
 

Hymie

Joined Mar 30, 2018
1,277
Discharging at 0.1A would reduce the battery charge capacity by 7.2Ah over three days. Given that the battery will be at least a 40Ah capacity, such a loss should not be noticeable in terms of the battery performance.

Before spending out on a new battery, check that the battery voltage is around 14V with the engine revving at around 2,000rpm. A reputable garage will be able to check the battery condition – otherwise you might be wasting your money buying a new battery.

If the battery is more than 5 years old, then it is the fair bet that the battery has reached the end of its life.
 

dl324

Joined Mar 30, 2015
16,839
If the battery is more than 5 years old, then it is the fair bet that the battery has reached the end of its life.
The battery in my truck is over 20 years old. It always started the truck until I stopped driving it over a year ago. Even after sitting for a year, it wasn't completely dead and only needed a 15 minute charge to start the engine.

They don't make batteries like that anymore... Sears Diehard Gold.
 

Thread Starter

Volttrekkie

Joined Jul 27, 2017
63
I must be missing something. If the battery was disconnected, what was providing the current?
The ampmeter was connected in series with the battery at the negative terminal. The cable was removed and one probe was put on the negative terminal and the other was put on the cable. If there is no drain, the amp reading should be zero. Apparently, if there is a short the current is draining from the battery to the chassis to earth ground? As I understand it, the negative terminal is connected to chassis but the positive terminal is not. If this is not correct, please somebody tell me.
Oh, and in Florida, battery life is only 2 years old. If you don't wear sunglasses, you will get cataracts too. I found that out.
 

Glenn Holland

Joined Dec 26, 2014
703
Have you checked under the seats?

If you are having a problem finding a short in your car, that's where they usually get lost. I've got the same problem with finding shorts except it's in the laundry instead of the car. However, track and field runners usually carry their shorts in a bag so they can be easily found:


OOOOOOOOOOOOPPPPPPPPPPS - This should have been posted in the Jokes section of AAC. :p
 
Last edited:

gerty

Joined Aug 30, 2007
1,305
Saw this the other night, kinda like your symptoms. I realize it's a different make/model but he explains some of his troubleshooting techniques ... He has several other videos you can search..
 

jpanhalt

Joined Jan 18, 2008
11,087
I rarely read the comments to YouTube posts, but in this case I did scan them briefly. Seems some users found that the problem was related to tin whiskers from the lead-free solder that was used. Others referred to something with the "motors" (meter movements?).

Which gets to, what year Accord is it? If you isolate it to the instrument cluster, it might be worth trying a DIY repair. I did that once with a Buick long ago, found a faulty solder joint, and saved lots.
 
If there is no drain, the amp reading should be zero. Apparently, if there is a short the current is draining from the battery to the chassis to earth ground? As I understand it, the negative terminal is connected to chassis but the positive terminal is not. If this is not correct, please somebody tell me.
There is always some drain, probably less than 20 mA. The memory pre-sets in the radio is one example. Modern cars may not even have them. The BCM may control the interior lights, so it has to be on all of the time. Some modules might stay on a fter you turn off the cat for some time. You won;t be able to disconnect the alternator with a fuse. There are fusible links in a car, some are part of the harness/ I did have a car where wires were the fuse and they were located in a fuse box as a wire jumper. There is always the wet spider web. Sometimes it's wet and sometimes it's dry.

That's a very real issue with analog telephone and a real pain to fix. It's disconnect in a binary search fashion if you can until the problem goes away. Disonnect half. Which half is the problem present. bisect that half etc. Sometimes guessing works, but wet spider webs in phone sockets willl do it all the time.

For cars, it could be whiskers or wet module because of a water leak. A mouse or squirrel my have chewed wires in the engine compartment. Yep, happened to me. The spark plug wires and the MAF sensor. Surprising, the car started. Nice spectacle in the engine compartment at dusk though.

Your getting numbers that almost seem reasonable. It would be nice if you could test your meter using a battery, resistor and meter or just use a resistor. 100 ma at 100 mV would be a R=.1/.1 or 1 ohm resistor. Make it able to draw a few amps. 4 ohm 100W resistors might be easy to come by. Opening a door might draw a few amps.

So, it's like you can't be playing inside the vehicle fusbox without disabling the dome circuit for sometime. Unfortunately, many problems are there.

What you MAY NOT know is that there are recessed tabs on blade fuses where you can probe. Sometimes locating a bad fuse means looking for 12 V across the fuse with the circuit on.

Similarly, the fuse has a very small resistance and you can measure the voltage drop across the fuse. I've seen tables of amperage and resistance. You don;t seem to be in that realm.

Make sure the battery is clean too. You do have a self-discharge possibility too. Check battery witth an electrolyte tester. It can catch failures that some battery testers won;t I know the hard way.
 

Thread Starter

Volttrekkie

Joined Jul 27, 2017
63
It's a 2006 Honda Accord. But well I finally took the battery to Autozone. It tested bad. I will see if the same thing happens with the new battery. My guess is it won't. I checked every fuse line. How can that be.
 

cork_ie

Joined Oct 8, 2011
428
.3A = 300mA is high for a parasitic drain. Less than 100mA is normal for any modern car.
Checking for parasitic drain is difficult and tricky in any modern vehicle as networks are running for some time after car is shutdown and locked.
It is much easier and more accurate to measure voltage drop than trying to measure current directly.

The foolproof way is:
1) Get a rough idea of the current drain with a DC clampmeter i.e. 400mA x12V is typical for a 5W boot(trunk) or glovebox light staying on
2) If current is less than 500mA, then most clampmeters are not accurate enough for this kind of fault finding.
3) Connect a new battery clamp to battery + and join to original battery clamp with a jump lead,
4) If at all possible try to simulate locked car with a door open to allow access to interior fusebox. i.e.ensure door switches appear closed and all locks shut. If bonnet (hood) switch is fitted make sure switch is in closed position. If there is ultrasonic intruder monitoring in interior make sure it is in disabled mode. In almost every car ultrasound can be temporarily turned off.
5) Lock car as normal with remotes or key in door. Wait ten minutes to allow all networks go into sleep mode
Ensure current is below 500mA with a clampmeter.
6 ) Insert a 1 Ohm 20 Watt power resistor between new battery clamp and original one. Then carefully remove the jump lead. All current now flows through the resistor.
7) Now measure voltage drop across the resistor. 300mA draw will cause a drop of 300 mV which can be accurately measured
but is not enough to disturb or wake the networks.
8) Check all live fuses, one by one.
9) Do not remove any fuses as you may wake the networks. Instead set voltmeter to 200 uV scale and check for voltage drop across each live fuse. Any fuse with even the tiniest voltage drop is drawing current.
10) Any suspect circuit replace fuse with 2 Ohm resistor soldered to two wires. Again wait for networks to shut down and measure voltage drop.

This will guarantee you being able to find your current draw on each suspect fused circuit.
 

Thread Starter

Volttrekkie

Joined Jul 27, 2017
63
.3A = 300mA is high for a parasitic drain. Less than 100mA is normal for any modern car.
Checking for parasitic drain is difficult and tricky in any modern vehicle as networks are running for some time after car is shutdown and locked.
It is much easier and more accurate to measure voltage drop than trying to measure current directly.

The foolproof way is:
1) Get a rough idea of the current drain with a DC clampmeter i.e. 400mA x12V is typical for a 5W boot(trunk) or glovebox light staying on
2) If current is less than 500mA, then most clampmeters are not accurate enough for this kind of fault finding.
3) Connect a new battery clamp to battery + and join to original battery clamp with a jump lead,
4) If at all possible try to simulate locked car with a door open to allow access to interior fusebox. i.e.ensure door switches appear closed and all locks shut. If bonnet (hood) switch is fitted make sure switch is in closed position. If there is ultrasonic intruder monitoring in interior make sure it is in disabled mode. In almost every car ultrasound can be temporarily turned off.
5) Lock car as normal with remotes or key in door. Wait ten minutes to allow all networks go into sleep mode
Ensure current is below 500mA with a clampmeter.
6 ) Insert a 1 Ohm 20 Watt power resistor between new battery clamp and original one. Then carefully remove the jump lead. All current now flows through the resistor.
7) Now measure voltage drop across the resistor. 300mA draw will cause a drop of 300 mV which can be accurately measured
but is not enough to disturb or wake the networks.
8) Check all live fuses, one by one.
9) Do not remove any fuses as you may wake the networks. Instead set voltmeter to 200 uV scale and check for voltage drop across each live fuse. Any fuse with even the tiniest voltage drop is drawing current.
10) Any suspect circuit replace fuse with 2 Ohm resistor soldered to two wires. Again wait for networks to shut down and measure voltage drop.

This will guarantee you being able to find your current draw on each suspect fused circuit.
What you are saying is very interesting. But I think I measured wrong. As I said, it was .5 or .6 to 1amp. It got higher as the day got hotter in Florida. I finally took the battery to Autozone and they told me it was definitely bad. I installed a new one. Hopefully that was the problem. Only time will tell. But I am suspicious. I took off both battery terminals and measured OL between them. I guess that is because there is a switch between them. But when I measure from the engine casing to the negative battery terminal I get continuity. Which makes sense to me because I heard the negative terminal is wired to chassis in a car? But I also get continuity from the positive terminal to the engine casing. So, both sides of the load are wired to chassis but they are separated by a switch?
 
Top