Hammmy

I purchased my civic used with 50k miles on it about a month ago from a dealer. Twice since then when I have tried to start it, I heard a slight clicking noise for under a second when I turn the key and then nothing. Both times I was able to jumpstart it no problem. After the first time I was able to drive for three weeks before it happened again, yesterday morning.
Also, after the failed start, the door locks would go frantic for about 5 seconds, locking and unlocking repeatedly. I don't know if that is relevant.

Any ideas? I'm bringing it to the dealer for some other work this weekend and would like to be a little more informed before I do.
edit: I forgot to add that the battery was brand new upon purchase.

ezone

Most common thing is the AC clutch relay stuck on, this can kill the battery overnight. Easy to prove IF one can catch it in the act. Tough to prove if it isn't stuck on when someone looks at it.

Computer modules on the network not going to sleep is another issue, not nearly as common though.

Hammmy

So I brought it to the dealer last weekend, told them about the problem, and they tested for drains and of course found nothing. Battery died again on me today. Bringing it back tomorrow and I have no idea how they are going to go about diagnosing it.

If I knew the slightest about cars I'd attempt a diagnosis myself.

If the AC clutch relay is the pproblem, why would it only get stuck on some times and not others?

One more question: the car is dead in my driveway, does it make a difference if I jump the car and drive it to the dealership or have it towed? In terms of diagnostics.

ezone

Not tough.
Eyeballs and a flashlight, and a little knowledge.
Can you tell if the compressor clutch is engaged and the compressor is running?
Now, can you tell the same thing----is it still running when it shouldn't be?
If the compressor is running while it shouldn't be, then tap on the clutch relay. If it turns off, the relay is the cause.

It's a well known problem, and relays in general have had the same issues for decades. It's just a common issue with a certain relay in these particular cars.
The bulletin that addresses this even says it can be intermittent.
The CAUSE is a physical issue, not an electrical issue (even though the RESULT is an electrical issue).
When it sticks on, all it may take is a small tap to get it to release. Closing the door or flicking the relay with a finger can make it release.

One hurdle is catching it in the act of being broken (stuck on).
The other hurdle is getting someone that is any good with electricity to either figure out this problem, or figure out why you keep (apparently) killing batteries.

Out of all the people in our shop that can use the battery testers, there are only 2 people in our shop that would have any hope of actually solving a problem, and only one of us can "adapt on the fly". The other is "strictly by the book" as far as diagnosing most problems.
Even then, the problem MUST be present so it can be proven, otherwise all we can do is use educated guesses (or maybe alternative testing).

OTOH: It COULD be as simple as a bad battery.
Or not.
All original batteries are suspect.

I get them way too often with a semi-weak battery and a complaint, but the almighty tester (that MUST be used because it makes a fancy printout that MUST be believed----and Honda won't accept any warranty claim without a "failed" test recorded and documented) might not think the battery is bad at all. The problem is, these testers that MUST be used -- are not infallible. The people that don't know what they are doing can only believe what the almighty tester tells them, they have no inkling that this tester can ever be wrong. They have no idea that there are a half dozen other tests that can be done on batteries.



Who pays for the tow?


In terms of the diag, the only thing that matters is whether or not the problem is still active --- and if someone smart enough to diagnose it is on the job.
I don't care that the battery is dead, that's just the result of the problem.

If the problem is gone once the car starts, then it doesn't matter how the car gets there....The battery will charge up on the way there, or they will charge it up once it gets there.
And if the problem is gone once it gets there, the result will probably be a wasted trip.

maaaak

You can trace slow battery drain problems at home while they are occurring simply, for just pennies and in less time than a trip to the dealer's. Pull the battery terminal (either side, make sure you have the radio code) with the car off and place a brake light bulb (old style) between the terminal and the connector with the terminal touching the plus side of the bulb (pointy end) and the connector touching the side of the bulb base. If there is a small drain (the type that takes a long time to drain the battery) the bulb will light up, if it's a large drain you have bigger problems and it will blow the bulb (you might get by with a larger wattage bulb like an old headlight and some jumpers). If it lights up start pulling fuses one at a time until it goes out. That will be the problem circuit.
This will work with the air conditioner relay problem because it's how I traced the drain in my daughter's civic. A quick visit to this site tells me it's a relay problem. Thanks, I hope this helps chase down those nagging intermittent battery drain problems.

ezone

Old thread is old.
This ain't the 60's any more. A light bulb has no accuracy whatsoever. Almost every car on the road now will light a small bulb using your test method, and it is normal.

You can get a DVOM with an ammeter at Sears pretty cheap. Much better than using a light bulb.


Absolutely 100% wrong.

maaaak

Who needs accuracy you're just looking for a battery drain. The normal things in a modern car that SLIGHTLY drains a battery when the car is turned off, such as the clock, WILL NOT, I repeat WILL NOT, light a brake light bulb (not a small bulb), not enough juice but any abnormal drain in the system will usually light the bulb. If the bulb glows when the car is off you have a leak that is large enough to drain the battery in just a few hours. The proof is when I pulled the air conditioner fuse the bulb went out completely even though the usual things, like the clock, were still operating.

I suggested using a bulb because many of us have them setting in the toolbox and it saves the time and expense of running to buy an ammeter. And, yes, I own an ammeter but I still use the bulb method most of the time. It's quick, accurate enough and, if the leak exceeds the limit of my ammeter, I don't have to worry about blowing its fuse. I usually have several bulbs in my toolbox that are expendable. Perhaps I shouldn't have used the term "old style" when referring to the brake light bulb, they have been used until fairly recently and might still be used on some newer cars.

And your comment that a large drain will not burn out the bulb as being "Absolutely wrong" is just plain wrong. The element in the bulb acts as a fuse and will only take so much power. If it was hooked up in parallel it would not blow out under any circumstance but hooked up in series with a power drain it will blow out if the drain goes beyond its capacity just like a fuse would. I've had it happen to me when testing for a leak so you're just wrong.

By the way electricity still works the same way it did in the 60's, that hasn't changed one bit. Anyone who reads this can be confident this method will work for them despite apparent claims to the contrary. I've used it dozens of times with much success.

Sincerely,
Retired auto mechanic who has worked on modern cars and their computer systems. I'm not THAT old, I was still in school in the 60's.

ezone

Ok, I might agree with using a 1157 type bulb as opposed to something small like a #194 for the draw check..... but most average non-mechanical people won't know any difference.
A bulb is a bulb is a bulb, to most people.
Just like "all cars have 4 wheels, therefore they are all the same".

My earlier point was that one can't tell the difference between a 28mA draw (which would be normal), and a draw of something like 60mA (excessive) using only a light bulb.



However:You are still 100% wrong on this.

12v bulb, 12v battery. The bulb will handle every last volt that battery can put out.
The filament will only glow bright or dim.
It will not burn out.
It is not a fuse.
It is a light bulb.

The bulb filament actually becomes a current limiting device when used in a series circuit like you described.
Still not a fuse.

ezone

I apologize if that reply 9 hours ago sounded harsh. Crappy day at work.

Jess5734

I have the same problem. Although I smell something hot and it doesn't smell like oil. I'm fixing to shut my engine off and take my ammeter and pull the fuses one by one until I find which circuit it's draining into and then looking at what I need to fix it. If anyone has any ideas what it might be please let me know. I have already shut the engine off once and pulled a battery terminal and when I connect it or disconnect it it sparked pretty good so I know I have an electrical drain somewhere!

ezone

What Y/M/M car do you have?

Flip down the back seat so you can make sure the trunk light isn't on. Same for map lights and sunvisor lights.

Use the ammeter first. Turn everything OFF that you can find, close the doors, etc.
Connect ammeter. If you have to take off a battery cable to connect it, then you need to wait up to 10-60 minutes for all computers to "go to sleep" (time varies with model and year, newer cars take longer to time out than older cars).


Post what your actual measured draw is after the various computers have timed out.
~2A-check compressor clutch relay stuck on