ketchup

iTrader: (3)

OK, so I FINALLY got around to hooking up my volt meter to the car/battery.

Set on 10 amps DC, it showed a drain of 0.04.

I removed the fuse for my power windows and the reading dropped to 0.03.
I removed the fuse for the radio/interior lights/clock and it dropped to 0.02.

So now what am I supposed to do? I'd like to keep the power windows (I hated using the crank, especially for the passenger side!!) and I would like to keep the clock functional. I've removed the radio/speakers years ago and pulled out the A/C system.

So what is causing the power windows and radio/clock/int. lights to drain the battery?

91DBSexyTime

maybe some random wire is touching something? but then again the fuse would pop. is your alternator properly charging the battery? it seems like something is staying on or a wire behind the fuse box in touching something.

iskate2

put those fuses back and try other fuses to see if something else is more of a drain.

cdmef

about a fuse going if a wire shorted out, not necessarily. I had a wire short out near the ecu and it caused one of my injectors to stay open, just dumping gas.... and other weird ish.

Janos Lin

I'd say look for a bigger drain. Are you sure you have your meter hooked up properly? Amperage is measured in series, voltage in parallel. The average battery for our cars can sustain about 30-50amps per hour for 20 hours before dropping dead. A drain of .04 amps would take 20,000 hours to kill your battery. The optima red top for our cars comes in at 50 amps per hour for 20 hours.


Janos

grog

one thing i found to cause a drain was the seatbelt retractors when the door is left open for a long time, if your door (ajar) sensor isnt funtioning it may cause a drain with the door shut.

one easy way to confirm this is to listen for a slight clunking noise while you pull the horn fuse (same cuircut) first with the door open (to familiarize yourself with the noise) then with it shut, it should clunk with the door open but not when its shut.

honda dealers have to fix seatbelt problems for free, if this is the problem replacing the retractors should fix the drain.

and for testing the drain use a "test light", disconnect the negative battery cable then connect the test light to the negative on the battery and the other end to a good body ground. it will light up when the drain is happening. then pull your fuses one by one until you have isolated the cuircut that is the problem

ketchup

iTrader: (3)

I've tested for the seat-belt/door issue and I don't have that problem. I've even left the seat-belt sensors unplugged.

I followed these instructions:
http://www.ehow.com/how_2249402_find...ery-drain.html

"A normal reading is usually under .035 amps."

I had the Kragen battery for the crx but recently switched to the Odyssey pc680 and if I leave the car in the garage for 2-3 nights with the alarm on, the battery won't have enough juice to crank the engine over. Even with the alarm off, it only takes a matter of days for the battery to drain.

DCRB

well, i hate to say it, but no matter what you will always have a drain with the clock, it always has constant power going to it (as long as the battery isn't dead) to keep the time accurate.

sanimalp

is your pc680 used by any chance? Brand new, they are 17 amp hours, which equates to only .85 amp/hour drain over 20 hours to sufficiently discharge the battery. Used, it is anyone's guess what the remaining charge capacity is.

you have a lot of options actually.

A) harbor freight makes a trickle charger, its like $15 and you plug it in if your car is going to sit a long period of time. in your case, long period means anything over 20 hours, given the choice of battery and your current load.

B) buy a new battery if yours is old or used. This may only fix the problem for a year or so..

C) install a master kill switch, similar to what NHRA requires, but maybe more discrete. your security system will no longer work with this scenario, but it would effectively kill power to your entire car, so theives would have to figure it out anyway. A side effect would be your clock will never be right again..

D) try to trace the miniscule drain you already know is there to reduce it.

In my opinion, those issues are ordered in sequence least difficult to most difficult. easiest would be A, hardest D. Honda already spent probably millions of dollars of electrical design to get the drain where it is today. Granted you probably could care less about some of the electrical nice-to-haves that they were worried about, but from what you have said, you only have 2 variables to play with, except your security system is probably the biggest drain, and that was not included in the "optional" systems, and for good reason.

If it were me, i would do C).. Its less costly than b in money terms, and cheaper than d in terms of time.

ketchup

iTrader: (3)

Yea but it shouldn't kill the battery within a few days. It's new BTW. Should I just leave that fuse out? I don't really care for the clock (I have to reset the time every time the battery dies or when I have to leave it disconnected).

I like that idea except for the disabling my alarm part. Wouldn't I be able to install a kill switch that cuts everything else off except for the alarm?


I checked it again with the alarm on and it showed 0.17 (every time the led blinked it jumped up to 0.19).

sanimalp

It all depends on how complex the alarm system is(i.e how many circuits it has to power), as well as how it was wired. My guess is that the alarm system is the major cause of the drain in the first place, so even after installing a kill switch, if you leave your alarm intact, it may still have the battery drain issue.

My first order of business would be to figure out how much amperage that alarm system pulls when it is armed, if you want to keep the system despite the kill switch. There might be some documentation to go along with it that says specifically what its drain is, but that may also be dependent on how it was installed and if it was done correctly. If that system draws more than a quarter amp /hour, then you wont really be solving the problem with a kill switch while leaving the security system in.

In my opinion, I would get rid of the alarm, and instead replace it with a very carefully hidden electrical kill switch. You live in Cali though.. so i don't blame you for not wanting to remove the alarm system.

ketchup

iTrader: (3)

Yeap, it's a keeper. Well for now I guess I could install a master kill and use that when the car is parked in the garage and arm the alarm for when it's parked shortly on the street (not overnight).

Keep in mind though, I've had this drain problem since I bought the car - before the alarm was ever installed, although I'm sure the alarm doesn't help the situation but then again it alone should not cause this. And I've gone thru a few batteries, so I'm sure that there is something IN the car that's causing this drain. I'm just having a hell of a time finding it and I'd hate to pay someone to try and find it when all they are going to blame is the aftermarket alarm.

Crx Jimmy

something is wrong. I can keep my pc680 in the car hooked up for over a month and it will start right up. is the battery new or used?

sanimalp

He said it was new. That was my first question as well.

It makes me think that there is something up with his reading, because the math doesn't work out. For the tiny drain he is seeing, the battery should be handling it just as CRX Jimmy said.. no issues.

I think there must be some other drain happening that we are not observing with the battery disconnected for testing. Its my favorite game!! simply by disturbing the system to test it, you change its behavior. we need to find another way to test the system, but without turning his multimeter into plasma in his hands..

Ill think of something.. i just need some time.. unless anyone else has commentary on testing it in a different way that is still accurate.

91akira

Quote: Set on 10 amps DC, it showed a drain of 0.04.

A 0.04 drain is the acceptable current drain for hondas. You have memory in the ecu that needs to power. Clock maybe. You will never never have a 0 reading.

sanimalp

Memory in the ECU is eeprom, so no electricity needed to maintain its state. a relay in the ignition switch circuit is what initially provides power to the ecu.

I agree that a .04 drain is acceptable, but i don't believe that reading is correct, because he says the car drains the battery in 24hrs or less when he leaves it sit. I think the number is not correct simply because the math does not work out. The pc680 should be able to handle a .85 amp drain over a 20 hour period before being dead. He is observing a draw 21 times smaller than that, but the battery is still dead within that time period. So one of these variables is incorrect. He said the battery is new, and also that he has had this problem with other pc680's, so the remaining variable is the .04 drain.

Thats why we need a better way to test it to be certain.

As a side note, How is the battery mounted? Is there any metal near the contacts that could be grounding the battery out when it is properly secured in the car, which would also be causing the false .04 amp draw reading because it is temporarily removed for testing?

civickid03

hahah a car battery cannot take 20-30 amps draw for more than a few hours, so the guy above is mistaken.

on mercedes benz a 1 amp draw will be enough to drain the battery overnite (these are 780 EN rated batteries).

Dont use an amp-clamp when testing - they arent as accurate.
Use a multimeter with it set on 10 amp - one lead on the negative of the battery one lead on the negative battery cable.. sorry if you already knew this just making sure

.04amps is fine and wont discharge the battery even over several days

anything over .10 amps by MB standards is too high of a draw. check ur alarm siren if u have one

sometimes a battery will discharge when it is old, ive had no-starts come into my shop and had no draw, the battery just didnt hold a charge, something to keep in mind.

user 93082343

This may all be true but perhaps the battery isn't dead? Perhaps the battery is draining to the point where it cannot start. Since voltage decays as usages gets longer, this means (I can only assume) that CCA will decay as well with this. Granted a start may not draw [too] much power, but CCA is a lot different then running current spikes in a system.

To the OP 0.04 should be fine, I am not sure what could be causing it, perhaps it is the coil in a relay somewhere, or perhaps the micro controller in the ECU has a watchdog timer running. Now-a-days watchdogs in complete power reserve modes can draw micro or even pico amps of power, but 20 years ago it is understandable to draw milliamps in this state. Or it may be something of the combination of things like this. Also, I am not sure if the ECU has any pull-up or pull-down resistors in it which could also cause minute drain on the battery. When your battery is "drained" what is the voltage? Also, you could likely be low on battery acid or have a dead cell in your battery, which would still be conductive but lower the capacity.

But hey...I am only a junior in computer engineering working at Cisco while I am 19.....so what do I know?

-John

civickid03

i dont know why everyone is trying to blame .04 amps for the cause..

you all can overengineer the hell out of this problem but is really simple considering there's no abnormal draw..

id bet money its the battery

if not the alternator isnt putting out the right amount of amps, but that would be more obvious to notice (dont try using a multimeter you'll blow the fuse) - see if you can get your hands on a midtronics tester, that will tell you for sure if its the battery. how tight is ur belt?

JOE BD-0

But it may be a faulty electric door latch mechanism that's draining your battery....
does your seatbelt beeper/door ajar alarm beep? is it working?
have you tried to isolate the drain to particular side of the car?

Faulty alternator diodes have also been known to cause a drain, but that condition is rare.

I would imagine that a real battery drain should be measured in amps, not 1/10th's of an amp????? I suspect a bad parasitic drain testing procedure at this point....

BTW, For the record-I'm definitely not the electrical genius you were asking for in the title of this thread....

sanimalp

civickid03, the pc680 is rated at 17 amp hours. This value gives an indication of how much current draw per unit time the battery can handle. Both ampere draw and hours are variable in this equation, but it all indicates when the battery will no longer provide sufficient current given its rating.

a couple examples..

How long will the battery last with .04 amp current draw?

17 amp hours / .04 amps = 425 hours.

How much amperage is being drawn from the battery if it is for all intents dead at 24 hours?

17 amp hours / 24 hours = .708 amps

Using the last equation, it shows that his car, turned off, would have to be drawing equal to or greater than .708 amps to not be starting in 24 hours.

.708 != .04

This mathematically shows that the observed .04 amp drain is not sufficient to kill his battery in 24 hours. It wont even be enough to kill it in 48 hours. it will take approximatly 425 hours to kill it. It points to the fact that the .04 amp draw is not a correct measurement, but by no fault of the OP's. My thought is that there has to be something going on that is different from the test to what is really happening in the car. but what is it? it looks like joe BD-0 and i are on the same train of thought..

Now to be fair, we also know the current provided by a battery is not linear if there is no source of recharge, so the previous numbers are approximations, but sufficient to show that it is very unlikely that with a .04 amp draw, the battery would be dead. Even if we, say, cut the time in half(!) for the .04 amp draw, it wont be sufficient. civickid03, if you've got some math that shows otherwise, i would love to see it. He said its a new battery, it did it with other new pc680 batteries, and it was doing it when he got the car.

sparksandtabs, I think you have some interesting thoughts on what is going on in terms of the cpu power draw, and now that i think of it, it does save state in some way because of the CEL lights. The only problem is it should have shown up in his test, and it didnt.


I don't know about the op, but i would rather "over-engineer" a problem than "over-throw money" at it..

ketchup

iTrader: (3)

as for my procedure that yielded these results, I followed this http://www.ehow.com/how_2249402_find...ery-drain.html

So with the alarm on it's draining ~0.18

This is the first pc680 battery that I've had and it's new.

The previous batteries were "Autolite" batteries and they are the factory rated battery.

The alternator is fine.

Normally after not being able to find the problem, my solution was to leave the car parked in the garage with the battery disconnected. Basically anytime the car sits more than a day, that would be my practice. When the car was used as a DD I was able to park with the alarm on and without draining the battery too badly but would not turn over right away.

JOE BD-0

what about the "parasitic draw" test in the factory service manual? sorry, I did not read your link.....btw, is your amps-meter in line with the system?.....good luck....

sanimalp

So the alarm is in fact the problem. .17 amps makes much more sense. with that draw, the pc680 is rated to last only 4 days, and with the drawdown not being linear, it is more likely 3 days or a little less.

I think this one is solved.. the op needs either a higher amp hour battery, a trickle charger, or a kill switch, or needs to leave his alarm off

ketchup

iTrader: (3)

Maybe I'll throw in that master kill like suggested and use that along with the alarm when parked outside. When the car is garaged, it'll be kill switch'd.